WO2022165832A1 - Method, system and brain keyboard for generating feedback in brain - Google Patents

Method, system and brain keyboard for generating feedback in brain Download PDF

Info

Publication number
WO2022165832A1
WO2022165832A1 PCT/CN2021/075967 CN2021075967W WO2022165832A1 WO 2022165832 A1 WO2022165832 A1 WO 2022165832A1 CN 2021075967 W CN2021075967 W CN 2021075967W WO 2022165832 A1 WO2022165832 A1 WO 2022165832A1
Authority
WO
WIPO (PCT)
Prior art keywords
brain
time
feedback
concepts
cerebral cortex
Prior art date
Application number
PCT/CN2021/075967
Other languages
French (fr)
Chinese (zh)
Inventor
张鸿勋
Original Assignee
张鸿勋
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 张鸿勋 filed Critical 张鸿勋
Priority to PCT/CN2021/075967 priority Critical patent/WO2022165832A1/en
Priority to CN202180016100.1A priority patent/CN115335102A/en
Publication of WO2022165832A1 publication Critical patent/WO2022165832A1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M21/00—Other devices or methods to cause a change in the state of consciousness; Devices for producing or ending sleep by mechanical, optical, or acoustical means, e.g. for hypnosis

Definitions

  • the present invention relates to a brain science, in particular to a method, a system and a brain keyboard for generating feedback in the brain.
  • the brain is the main component of the human central nervous system, including about 86 billion nerve cells (neurons) and hundreds of billions of glial cells. Each neuron usually has hundreds to thousands of synapses, and the number of synapses in the brain is estimated to be about 10 15 .
  • the weight of an adult brain is about 1.2-1.6 kg, and the main component is blood. Although the weight of the brain is only 2%-4% of the body weight, its oxygen consumption can account for 1/4 of the total oxygen consumption.
  • the blood flow of the brain accounts for 15% of the total blood output of the heart, and its power consumption is about 25W.
  • the brain is the most complex organ of the human body. It is the "command center” that accepts external stimuli, generates sensations, forms consciousness and thinking, and issues instructions and drives actions.
  • the cerebral cortex is the material basis of higher neural activity and the organization that produces thinking.
  • the left and right hemispheres of the cerebral cortex can be divided into 5 lobes: frontal lobe, temporal lobe, parietal lobe, occipital lobe and insula lobe; among them, the frontal lobe and temporal lobe are traditionally considered to be related to language, emotion, and memory.
  • the research on the representation of the language center and the physiological basis of memory formation and retrieval, the formation and influence of emotions, etc. are still in the early stage.
  • the research of brain application is a key field in the field of brain science. It aims to analyze the nervous system structure of the brain and the material basis of psychological activities, and to develop algorithms or models of advanced brain functions by using methods and means such as information science and computer science. , to promote the development of artificial intelligence, robotics and other fields.
  • the present invention proposes a method, a system and a brain keyboard for generating feedback in the brain, so as to generate desired feedback in the brain.
  • the present invention provides a method for generating feedback in the brain, comprising: determining a sequence having a time length, wherein all time periods within the time length are The sequence includes one or more concepts; and the sequence is perceived by the brain in a natural manner for the length of time to generate desired feedback in the brain.
  • a system for generating feedback in the brain comprising generating means and transmitting means, the generating means being configured to generate a sequence of concepts having a temporal length, wherein the The one or more concepts are included in a plurality of periods of the time length; the delivery device is configured to cause the sequence to be perceived by the subject's brain in a natural manner during the time length, in the subject's hope-generating feedback in the brain of the user.
  • the present invention provides a brain keyboard, which includes a keyboard and a processor, the keyboard includes a plurality of keys, at least one or more keys correspond to one or more concepts; the processor is processed by configured to receive keystrokes from the keyboard to form a sequence having a length of time, wherein the sequence includes the one or more concepts for a plurality of periods of the length of time; wherein the sequence is at the time Perceived by the brain in a natural manner over length produces desired feedback in the brain.
  • the present invention can make the subject's brain perceive a sequence of concepts in a natural way, so as to generate desired feedback in the subject's brain, and the desired feedback corresponds to a desired experience, such as relaxation, calm, self-confidence, Joy, contentment, bravery, health, excitement, success, beauty, etc.
  • the method and system provided by the present invention can not only make ordinary people feel the various experiences they want, but also can be used to treat or prevent psychological diseases or mental diseases.
  • the method and system provided by the present invention have low cost, no harm to subjects, easy operation and remarkable effect.
  • FIG. 1 is a schematic structural diagram of an MRI apparatus according to an embodiment of the present invention
  • Figure 2 is a schematic diagram of the model in terms of morphemes disclosed by Huth et al.
  • Figure 3 is a schematic diagram of the principal component analysis of the semantic model in terms of morphemes disclosed by Huth et al.
  • FIG. 4 is a flow chart of a method for generating feedback in the brain according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a method for dividing target feedback into multiple activation modes by time according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of a time slice obtained after the target feedback is sliced by time according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of obtaining time slices after identifying and merging time slices according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of acquiring one or more concepts corresponding to an activation mode of a time slice according to an embodiment of the present invention
  • FIG. 10 is a schematic block diagram of a system for generating feedback in the brain according to an embodiment of the present invention.
  • FIG. 11 is a schematic block diagram of a system for generating feedback in the brain according to another embodiment of the present invention.
  • FIG. 12 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention.
  • FIG. 13 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention.
  • FIG. 14 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention.
  • Fig. 15 is a schematic diagram of the physical fatigue recovery ability evaluation score of the first round of test subjects according to an embodiment of the present invention.
  • Figure 16 is a schematic diagram of the physical fatigue recovery ability evaluation score of the target test group during the second round of testing according to an embodiment of the present invention
  • FIG. 17 is a schematic diagram of the physical fatigue recovery ability evaluation score of the reference group in the second round of testing according to an embodiment of the present invention.
  • FIG. 18 is a schematic diagram of the ratio of the average deep relaxation degree of brain waves of the target test group during the second round of testing according to an embodiment of the present invention.
  • Fig. 19 is a schematic diagram of the ratio of the average deep relaxation degree of brain waves of the reference group during the second round of testing according to an embodiment of the present invention.
  • 20 is a schematic diagram of the sleep quality score value of the first round of testing subjects according to an embodiment of the present invention.
  • 21 is a schematic diagram of the sleep quality score value of the target test group during the second round of testing according to an embodiment of the present invention.
  • 22 is a schematic diagram of the sleep quality score value of the reference group during the second round of testing according to an embodiment of the present invention.
  • Figure 23 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale of the subject during the first round of testing according to an embodiment of the present invention
  • Figure 24 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale in the second round of the test target test group according to an embodiment of the present invention.
  • Figure 25 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale in the second round of the test reference group according to an embodiment of the present invention.
  • Figure 26 is a schematic diagram of the reverse scoring average of the physical problem assessment form of the first round of test subjects according to one embodiment of the present invention.
  • Figure 27 is a schematic diagram of the average positive score of the physical problem improvement assessment form for the second round of test target test groups according to one embodiment of the present invention.
  • FIG. 28 is a schematic diagram of the average positive score of the physical problem improvement evaluation form of the second-round test reference group according to an embodiment of the present invention.
  • the evolution of life on earth for billions of years can be summed up in two basic evolutionary ways: one is the biological evolutionary way to realize the transmission of genetic information, that is, the genome composed of nucleotide sequences generates proteins and cell tissues according to certain rules. , resulting in an infinite variety of living organisms; the other is an evolutionary way of realizing non-genetic information transmission through language.
  • the latter way of evolution distinguishes human beings from other living beings, has a mind, and promotes cultural progress and prosperity.
  • the evolution of life in terms of mind includes five levels from low to high: neural level, psychological level, language level, thinking level and cultural level; the neural level and psychological level are shared by humans and animals, which can be called low-level Cognitive level; language level, thinking level and cultural level are unique to human beings, also known as higher-order cognitive level.
  • Language is the basis of human cognition. Human thinking is formed on the basis of abstract conceptual language. Language and thinking together build the human knowledge system, and even define the entire human society. The accumulation of knowledge forms culture and cultural decisions. In turn, it promotes the development of human society.
  • neurolinguistics is a very important research field. Neurolinguistics studies the relationship between language and brain function. Its purpose is to explain the neural and psychological mechanisms of human language understanding, production, acquisition, and learning. The research object is the interaction between the human nervous system and human language. Understand how the human brain receives, stores, processes and extracts linguistic information.
  • the cerebral cortex in the human brain is the material basis of advanced neural activities and the organ that produces thinking, which dominates all activities within the body and coordinates the balance between the body and the external environment.
  • the cerebral hemisphere is divided into five parts by means of the sulci and gullies on the surface of the cerebral cortex, namely the frontal lobe, the temporal lobe, the occipital lobe, the parietal lobe and the insular lobe.
  • the temporal lobe is related to perception, hearing and memory;
  • the occipital lobe is related to vision;
  • the parietal lobe is related to touch, temperature, pressure, and pain;
  • the insula is related to the autonomic function of the brainstem, and also processes Taste information.
  • Broca's area German anatomist and neuropathologist Carl Wernicke named the left posterior superior temporal gyrus region of the brain the Wernicke District and proposed that different areas of the brain are specialized for different language tasks.
  • Korbinian Brodmann divided the surface of the brain into differently numbered regions, called Brodmann, based on the cellular structure and function of each region. )Area.
  • Brodmann's area is widely used in the neurolinguistic field to study the location of specific language "modules" in the brain, for example, Broca's area processes speech motor production, while Wernicke's area processes auditory speech comprehension.
  • the left brain is responsible for memory, time, language, judgment, arrangement, classification, logic, analysis , writing, reasoning, inhibition of the five senses, etc., the way of thinking has the characteristics of continuity, continuity and analysis. Therefore, the left brain is called the “conscious brain”, “academic brain”, “language brain” or “logical brain”.
  • the right brain is mainly responsible for spatial image memory, intuition, emotion, physical coordination, visual perception, art, music rhythm, imagination, inspiration and insight. Therefore, the right brain is also known as the "instinct brain”, “subconscious brain”, “creative brain”, “music brain” or “artistic brain”.
  • Neurolinguistics examines physiological brain responses using sentence processing experiments, ELAN, N400, and P600 brain responses obtained from ERP techniques, and then compares the results of physiological brain responses with predictions from sentence processing models proposed by psycholinguists It can reflect the rationality of different sentence processing models.
  • neurolinguistics can also guide psycholinguistics to propose new theories on the structure and organization of language based on knowledge of brain physiology by "generalizing knowledge of neural structure into language structure".
  • pathological experimental methods were used to study the actual language process and speculate on the language mechanism of the brain.
  • pathological experimental method analyzes the language status of patients with brain injury from the perspective of neuropsychology, and uses the analysis of brain injury regions to understand the process of language generation and its neuropsychological mechanism.
  • Electroencephalography uses non-invasive EEG electrodes to record potential changes at different cranial brain locations. Electrocorticography technology is more in-depth, using ECoC electrodes built into the cerebral cortex to collect potential changes in deep cortical activity.
  • Electrocorticography although more accurate, is an invasive technique for acquiring brain signals.
  • Magnetoencephalography technology uses a particularly sensitive ultra-cold electromagnetic detector to measure the extremely weak brain magnetic waves in the brain, so as to obtain the changes in the electric field distribution in the brain.
  • Event-related potential technology belongs to a kind of evoked potential method.
  • Described brain imaging technology includes computed tomography (Computed Tomography, CT), such as X-ray CT, ultrasound CT, ⁇ -ray CT; Positron Emission Computed Tomography (Position Emission Computed Tomography, PET); Single Photon Emission Computed Tomography (Single-Photo Emission Computed Tomography, SPECT); Magnetic Resonance Imaging (MRI); Functional Magnetic Resonance Imaging (fMRI); Near Infrared Spectroscopy (NIRS); Functional Near Infrared Spectroscopy (functional Near Infrared Spectroscopy, fNIRS); cerebral angiography; photoacoustic imaging (Photoacoustic Imaging, PAI); fast functional photoacoustic microscopy (fast-functional RAM) and so on.
  • computed tomography such as X-ray CT, ultrasound CT, ⁇ -ray CT
  • Positron Emission Computed Tomography Positron Emission Computed Tomography (Position Emission Computed Tomography, PET
  • fMRI non-invasive brain information acquisition techniques
  • FIG. 1 it is a schematic structural diagram of an MRI apparatus.
  • the MRI equipment mainly includes a magnet system, a radio frequency system and a computer image reconstruction system.
  • the magnet system is mainly used to generate two magnetic fields, one is the static magnetic field, also known as the main magnetic field; the other is the gradient coils.
  • a radio frequency system includes a radio frequency (RF) generator and a radio frequency (RF) receiver.
  • a radio frequency generator is used to generate a short and strong radio frequency field, which is applied to the sample in a pulsed manner to cause the hydrogen nuclei in the sample to produce NMR phenomena.
  • the radio frequency receiver is used to receive the NMR signal, amplify it and send it to the computer image reconstruction system.
  • the computer image reconstruction system converts the analog signal into a digital signal through the A/D converter for the signal sent by the radio frequency receiver.
  • the D/A converter is added to the image display to display the image of the layer to be observed with different gray levels according to the size of the NMR.
  • the product sum of the time basis functions and the space basis functions that are independent of each other is used to represent the magnetic resonance signal to obtain a functional magnetic resonance imaging image with high spatial resolution and high temporal resolution;
  • EPI Echo Planar Imaging
  • fNIRS technology uses the difference in the absorption rate of oxyhemoglobin and deoxyhemoglobin in the brain tissue to the near-infrared light with wavelengths in the range of 600-900nm, and directly detects the cerebral cortex in real time. information on hemodynamic activity.
  • the brain activity can be deduced from the changes in the hemodynamic activity information.
  • the equipment involved may include fNIRS detectors, light sources, probes and computer systems.
  • fNIRS detectors include forehead detectors and whole-brain detectors. The light source and detector are placed on the probe, which is in contact with the organism under test, such as the brain, and the probe is wired to a computer system.
  • the computer system controls the on and off of the light source by sending out control signals, the detector inputs its measurement data to the computer system, and obtains brain function images through AD conversion and processing of the signals.
  • fNIRS technology provides a powerful monitoring method for the monitoring of brain activity.
  • FIG 2 is a schematic diagram of the model in terms of morphemes disclosed in the article by Huth et al.
  • Huth et al. used fMRI imaging to measure brain BOLD (Brain Blood-Oxygen Level Dependent) feedback during seven subjects listening to a 2-hour story told in natural language.
  • Each word (Word) is projected into a 985-dimensional space based on the statistics of word co-occurrences.
  • the morphological aspect of the model reflects how word occurrence affects BOLD feedback.
  • Figure 3 is a schematic diagram of principal component analysis of the semantic model in terms of morphemes disclosed in Huth et al. Principal component analysis reflects 4 important semantic dimensions in the brain, and the color map is obtained after RGB colorization.
  • Huth et al. led to some important conclusions: First, there is a clear correlation between words and the distribution of semantically selective regions in the cerebral cortex. On the other hand, among different individuals, the distribution of semantically selected regions corresponding to the same word is also highly consistent.
  • a "concept” refers to a unit of thought activity that has a defined meaning.
  • a concept may be a Chinese character (character) or a word (word); it may also be an English word (word) or phrase (phrase).
  • word word
  • phrase phrase
  • the strokes of Chinese characters, foreign letters, Japanese katakana, etc. that do not have specific meanings cannot become units of thinking activities, and they are not concepts called in this article.
  • the concepts of this paper are clearly linked to language. Concepts expressed in different languages may have the same meaning; however, in most cases the two are not the same.
  • the Chinese "table” and the English “table” are not consistent in the extension of the concept, so they can be regarded as different concepts.
  • a subset of concepts can be derived based on existing linguistic classifications and the frequency of co-occurrence of statistically identical concepts.
  • Several representative concepts are included in this subset of concepts. These representative concepts can form a morphological space.
  • Other concepts can be expressed as projections of one or more representative concepts in the space of the morpheme aspect of the subset.
  • the subset can express other concepts that are not present in the subset's representative concepts.
  • the subset under that language can express the meaning of the language's concepts, thereby reducing the language to a sequence of subsets of representative concepts that can be processed quickly.
  • the subsets of representative concepts can also be classified into different subsets of experience based on the experience expressed.
  • experience refers to the feedback formed in the brain due to experience, which can be a fixed feedback formed by multiple practices; it can also be a single feedback formed by a single experience.
  • Experiences include, but are not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, beauty, and more.
  • the subset of experience formed by the categorized representative concepts can more easily form feedback in the brain corresponding to the experience it represents.
  • feedback in the brain refers to responses in the cerebral cortex to external stimuli.
  • different regions in the cerebral cortex will be active, and further lead to changes in the metabolism of various active substances, such as oxygen, glucose, etc., which are sensed by EEG signal detection means.
  • the activation patterns of the cerebral cortex represent the spatial distribution of regions of the cerebral cortex's active state. The different activation patterns indicate that different regions in the cerebral cortex are active. Even when a person is asleep, several areas of the cerebral cortex are active, albeit in fewer areas than in the awake state. If you look at changes in multiple regions of the cerebral cortex that are active over time, such changes actually indicate changes in the activation pattern of the cerebral cortex.
  • Cortical activation patterns are rapidly time-varying. The durations of the different activation modes are short, such as hundreds of milliseconds to a few seconds. Different activation patterns may arise in response to different external stimuli, or they may arise spontaneously in response to higher brain functions. Therefore, the different activation patterns of the cerebral cortex are independent of each other. Conversely, independent activation patterns also correspond to different external stimuli or the results of higher brain functions.
  • the cerebral cortex will experience multiple different activation patterns within a period of time, such as a dozen or so seconds. Therefore, the changes in the active regions of the cerebral cortex during this period can be sliced in time (eg, 100 ms) and the active regions of the cerebral cortex in each time slice can be regarded as an activation pattern.
  • Activation patterns in adjacent time slices can be continuations of the same activation pattern and, therefore, are not independent of each other.
  • the activation patterns in adjacent time slices are different activation patterns and are independent of each other. Therefore, through the independent relationship between the activation patterns in each time slice, the changes in the activation patterns of the cerebral cortex during this period can be obtained.
  • the simulated feedback refers to the formation of feedback in the brain that simulates the desired feedback.
  • the analog feedback will match the target feedback as closely as possible, although the two need not be exactly the same.
  • the goal feedback represents a hopeful experience including, but not limited to, relaxation, calm, confidence, joy, satisfaction, bravery, health, excitement, success, beauty, and the like.
  • the simulated feedback transfers the desired experience from another cerebral cortex to the subject's cerebral cortex by simulating the target feedback.
  • “perceived by the brain in a natural manner” refers to the input of concepts into the brain from the outside world through the human body's own perception methods, including: hearing, sight, touch, smell, and taste. For example, listening to a text, reading a text, tactilely sensing a text in Braille, smelling a certain smell or tasting a certain taste.
  • “perceived by the brain in a natural way” does not include communicating with the brain through invasive or non-invasive means such as acupuncture, electric shocks, surgery, etc.
  • invasive or non-invasive means such as acupuncture, electric shocks, surgery, etc.
  • the brain-computer interface if the brain-computer interface communicates with the brain in the same way as the human body itself perceives, it is also within the scope of this term.
  • the brain-computer interface uses non-human perception methods, such as applying voltage through electrodes in a certain area of the brain, then it is not within the scope of the present invention. Understandably, communicating with the brain in a natural way is more efficient, has fewer side effects, and is more conducive to acceptance.
  • brain keyboard refers to an input device that includes a plurality of keys. At least one or more keys correspond to one or more concepts. In some embodiments, multiple keys correspond to multiple concepts of the representative subset of concepts. In other embodiments, multiple keys correspond to multiple concepts of the experience subset. These keys can be multiple physical keys or multiple virtual keys on the input interface.
  • a key of the brain keyboard is "pressed"
  • the concept corresponding to the key is entered.
  • Multiple keys being “pressed” in succession result in a sequence with a length of time. The sequence includes the one or more concepts entered by the keys of the brain keyboard over multiple periods of the time length.
  • a brain keyboard may be considered a dedicated device capable of applying the methods of the present invention.
  • the brain keyboard may also include other components such as a processor, memory, display, speakers, etc., thereby becoming a device capable of interacting with the brains of users and subjects.
  • a "psychological disorder” refers to a disorder of thinking, emotion, and behavior that deviates from the usual norms of social life.
  • Psychological disorders include, non-exhaustive, depressions, major depressions, treatment-resistant depression and treatment-resistant bipolar depression, bipolar disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression , postpartum depression, premenstrual dysphoric disorder (PMDD), situational depression, atypical depression, mania, anxiety, attention deficit disorder (ADD), attention deficit disorder with hyperactivity (ADDH), and attention Deficit/Hyperactivity Disorder (AD/HD), Bipolar and Manic Disorders, Obsessive-Compulsive Disorder, Hyperphagia, Premenstrual Syndrome, Substance Addiction or Abuse, Nicotine Addiction, Psycho-sexual Dysfunction, and Pseudosexuality one or more of bulbar disease.
  • mental disorder refers to a disorder in which cognitive, emotional, volitional, or behavioral disorders result from dysfunctional brain functions.
  • Psychiatric disorders include, non-exhaustive, schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette one or more of the diseases.
  • non-disease psychotic state refers to a cognitive, emotional, volitional, or behavioral disordered state of mind that does not have a pathological basis.
  • Non-disease mental disorders include, but are not exhaustive, one or more of: lack of confidence, timidity, sensitivity, inattention, weakness of will, obsessive-compulsive behavior, fear of exams, fear of speaking.
  • the present invention aims to transplant experiences from another brain into another brain by simulating feedback in the other brain by means of the means by which concepts are perceived by the brain in a natural way.
  • the high degree of consistency of feedback formed between different human individuals after perceiving the same concept becomes the basis for experience transplantation in a natural way; deep learning neural network models make such simulations a reality. While the simulated and targeted feedback will not be exactly the same, similar experiences can also be formed in the brain. This low-cost, harmless, and easily accepted experience transplant into the brain will undoubtedly be a major advance in the field of brain science.
  • FIG. 1 is a flow chart of a method for generating feedback in the brain according to an embodiment of the present invention.
  • a desired experience is determined.
  • the desired feedback formed in the brain is an analog of the target feedback.
  • Goal feedback is a temporal change in activation patterns in the cerebral cortex of another brain that represents an experience.
  • Such experiences include, but are not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, and beauty. Therefore, determining the desired experience can also be understood as determining the simulated target feedback.
  • the desired feedback does not have to be exactly the same as the target feedback, but it can be as close to the target feedback as possible. In this way, the experience of hope can be obtained in the brain.
  • test phobia As another example, some people have severe test phobia. Even if you study well, you can still have a lack of confidence before the test and can seriously affect your test scores. Such test phobia is also difficult to change through human will. For those with this need, the desired feedback can be identified as the experience of confidence or exam confidence.
  • target feedback is feedback formed in the brain of a person with a certain experience.
  • targeted feedback is obtained by collecting feedback on the desired experience in the brains of individuals with the desired experience. For example, if you want to gain the experience of deep relaxation, you can choose individuals with deep relaxation experience to be religious ceremonies, and choose the feedback for the desired experience as the feedback formed in the brains of these individuals after they complete religious activities such as meditation and prayer. .
  • By recording the feedback formed in the cerebral cortex for a period of time through brain signal detection technology it can be used as the target feedback.
  • a different individual may be selected, or feedback in the cerebral cortex following different activities of an individual may be selected as the target feedback.
  • the present invention does not make any limitation here.
  • target feedback is recorded using fMRI data obtained by scanning the brain with a functional magnetic resonance apparatus.
  • fMRI magnetic resonance apparatus
  • the optional range of activities to reproduce the desired experience will be very small, usually only through wearable devices, such as VR glasses, etc., so that individuals with the desired experience have the feeling of participating in the activities that reproduce the desired experience .
  • MRI image data, CT image data, SEPECT image data, and PAI image data similar to fMRI image data, impose many limitations on activities that reproduce desired experiences.
  • the fNIRS image data obtained by the frontal region fNIRS detector, the whole-brain fNIRS detector, or the wearable detection device used in the NIRS image data are more suitable for more complex activities.
  • EEG or magnetoencephalography data, etc. obtained by invasive or non-invasive electrodes do not require the brain to be placed in a large device, so there are few restrictions on complex activities.
  • fMRI data is taken as an example to illustrate the technical solution of the present invention.
  • other brain signals eg, magnetoencephalography
  • data obtained from brain information detection techniques may also be employed in the protocols of the present invention.
  • the present invention does not make any limitation here.
  • a 3D convolutional neural network model can be chosen to learn feedback from the cerebral cortex that wishes to experience.
  • a CNN that uses two-dimensional images can also be selected.
  • the fMRI data recorded after the individuals with the desired experience reproduced the desired experience activity were used as training samples. Since language, age, gender, religious belief, education level, occupation or previous occupation among individuals have a great influence on the desired experience, these human-related parameters are used as training parameters.
  • the construction of the neural network model is performed using convolutional sparse coding (CSC).
  • Convolutional sparse coding is an unsupervised learning method of linear convolution. The model is simpler, more intuitive, and easy to analyze and understand.
  • the individual's language, age, gender, religion, education, occupation or previous occupation are input as parameters to the trained neural network model.
  • the neural network model outputs the target feedback that best matches the individual's experience. In this way, it can not only solve the problem that there are fewer training samples and cannot obtain an accurate neural network model, but also reduce the impact of individual differences between training samples on target feedback.
  • the neural network model applied to the present invention is not limited to CNN, and other feature classification models can also be used; nor is it limited to three-dimensional data obtained by fMRI as training samples.
  • Other brain signal or brain information data can also be applied here.
  • magnetoencephalography (MEG) data can also be processed very well with the model established by CSC.
  • a database of target feedback for each experience is established.
  • the multiple target feedback in the database is categorized by the different experiences targeted.
  • parameters related to people such as language, age, gender, religious belief, education level, occupation or previous occupation, can also be the basis for further classification of target feedback. If parameters related to people such as language, age, gender, religious belief, education level, occupation or previous occupation are further considered, the matching degree of target feedback that can be obtained is higher.
  • the neural network model constructed as above can be used to update the target feedback stored in the database.
  • the accuracy of the target feedback database is not as good as that of the trained neural network model, it is more convenient and faster to use, and it can also save the previous steps.
  • the target feedback is divided into a plurality of activation modes by time.
  • Targeted feedback in the cerebral cortex consists of temporal changes in activation patterns in the cerebral cortex.
  • the activation pattern includes the spatial distribution of the activated parts of the cerebral cortex.
  • target feedback can be thought of as a temporally changing sequence of multiple activation patterns in the cerebral cortex.
  • the step of dividing the target feedback into multiple activation modes by time includes:
  • Step S1301 to slice the target feedback by time.
  • the length of each time slice is 10-50ms. As shown in the schematic diagram in FIG. 6 , the length of each time slice is 20ms, and it is numbered as A00001, A00002, etc. to facilitate subsequent processing. Although a smaller time slice can improve the accuracy of the simulation, if the time length is too small, it will not only result in a large amount of data to be processed and a huge amount of computation, but also a too small time slice may not match the perception in the natural way. , resulting in a large amount of interference data. After this step, a set of time slices is obtained, where each time slice has a unique number.
  • Step S1302 Determine the activation pattern of the cerebral cortex in each time slice, that is, the spatial distribution and signal intensity of the activated part.
  • a representative activation mode in the time slice is acquired as the activation mode of the time slice.
  • There are many ways to obtain a representative activation pattern for example, selecting the spatial distribution of the activation part at the middle of the time slice as the representative activation mode; or, calculating the average spatial distribution of the superposition of the activation parts at each time in the time slice as a representative sexual activation mode. Of course, other methods can also be applied to this.
  • Step S1303 Calculate whether the activation patterns of the cerebral cortex in adjacent time slices are independent of each other. If the activation patterns of the cerebral cortex in two adjacent time slices are non-independent, it means that the activation pattern lasts for at least two time slices, then the two time slices are merged in step S1304, if the two time slices are The activation modes of the cerebral cortex in the time slice are independent of each other, and the two activation modes are recorded as different activation modes in step S1305.
  • Step S1306 after judging whether all adjacent time slices have been processed, if not, repeat the foregoing steps until the adjacent time slices are independent. If all adjacent time slices have been processed, then the target feedback has been divided into multiple activation modes that last an integer number of time slices at this point. As shown in Figure 7, each time slice, such as B0001 and B0002, has a certain time length, and its activation mode is different from that of other time slices, that is, another time slice set is obtained at this time, in which the Each time slice includes an activation pattern and has a corresponding time period, and all time slices in the set form a temporally continuous sequence.
  • the difference in the activation pattern of the cerebral cortex in adjacent time slices is calculated. If the difference exceeds a predetermined range, the activation patterns in the two time slices are considered to be independent of each other.
  • correlation coefficients are calculated for the activation patterns of the cerebral cortex in adjacent time slices. If the correlation coefficient exceeds a predetermined threshold, the activation patterns in the two time slices are considered to be independent of each other.
  • other methods can also be applied to this.
  • the target feedback is divided into multiple time periods, and each time period corresponds to an activation mode.
  • step 140 one or more concepts corresponding to each activation mode among the plurality of activation modes of the target feedback are identified.
  • the cerebral cortex of the cerebral cortex has a high degree of consistency in its distribution and signal intensity in response to the same concept. Signal strength also varies.
  • the distribution of activation regions in the cerebral cortex corresponding to each concept in a group comprising a plurality of concepts and their signal intensities are obtained.
  • the distribution of activation areas in the cerebral cortex corresponding to a concept and its signal strength are defined as the "concept pattern" corresponding to the concept.
  • a conceptual schema database is established to store conceptual schemas for a plurality of concepts. Take fMRI data as an example, by recording the fMRI data of the feedback formed in the cerebral cortex when a subject listens to a piece of text expressed in natural language. After data processing, the fMRI data is corresponding to the concept in the text, and the concept pattern reflected by the fMRI data corresponding to the concept can be obtained.
  • a conceptual schema database can be established. It will be understood by those skilled in the art that other types of brain signals or data from brain detection techniques can also be used to build the conceptual pattern database. In other manners, concepts can also be associated with their corresponding conceptual schemas to obtain a conceptual schema database. The present invention does not make any limitation here.
  • the corresponding conceptual pattern can be determined according to the activation area and its distribution and signal strength.
  • Mode 1 compare the conceptual schema 1 with multiple conceptual schemas stored in the conceptual schema database, and find out one concept or a combination of multiple concepts.
  • the activation mode corresponding to these concept combinations is the superposition of one or more concept modes in the concept combination, and the activation mode of the concept combination is close to the activation mode of the target feedback.
  • an activation mode in the target feedback corresponds to a combination of one or more concepts, or, an activation mode in the target feedback is decomposed into one or more concepts in the combination of one or more concepts.
  • a deep learning neural network model is used to obtain a conceptual combination of activation patterns in target feedback.
  • a convolutional neural network model CNN
  • Other neural network models for pattern recognition such as deep belief network DBN model, recurrent neural network RNN model, etc., can also be applied here.
  • fMRI data taking fMRI data as an example, two or more different words (corresponding to different concepts) are read out to the subject, and the fMRI data fed back in the cerebral cortex of the subject is recorded.
  • the dataset formed by different words and corresponding fMRI data is used as the training set to train the neural network model.
  • the trained neural network model takes fMRI data as input and outputs a plurality of different concept combinations, and the plurality of different concept combinations are ordered from high to low matching with the fMRI data as input.
  • Multiple concept combinations corresponding to one activation pattern in the target feedback can be obtained by using the trained neural network model.
  • typically only the best matching concept combination is used. When the matching degrees of several concept combinations are not much different, all these concept combinations are reserved to facilitate subsequent selective use.
  • a sequence having a length of time is determined, wherein the sequence includes one or more concepts over multiple time periods of the length of time.
  • the target feedback is divided into a plurality of activation modes, each activation mode lasting an integer number of time slices of time length. Further, each activation mode may correspond to one or more concept combinations.
  • the duration of the concept combination is the same as the duration of the active mode, which is also the duration of an integer number of time slices.
  • target feedback can be decomposed into conceptual composition sequences of multiple temporal lengths. Further, one of the concept combination sequences of multiple time lengths is selected as the concept combination sequence corresponding to the target feedback. As shown in FIG.
  • activation mode 1 includes three concepts: concept 1-concept 3; activation mode 2 includes one concept: concept 4; activation mode 3 includes one concept: concept 5.
  • the starting times corresponding to these five concepts are 0, T1, T2, T3, T4, and T5, respectively.
  • each concept includes both a start time and a duration period.
  • the criteria for selecting a concept combination sequence may vary. For example, the selection can be made according to the corresponding relationship between the desired experience and the concept combination, and the concept that does not appear frequently in a certain experience can be eliminated. For example, if the desired experience is "calm”, then concept combinations including concepts such as “fierce” and “intense” are not selected as much as possible. For another example, the selection can be made according to the association relationship between the concept combinations, and the concept combinations that are relatively poorly correlated or difficult to be correlated with each other can be eliminated. For example, the concept combination before and after in the sequence is all related to "flying", then the concept combination in the middle also selects the concept combination related to "flying".
  • the selection can be made according to the theme of the whole concept combination, and the concept combination with poor correlation can be eliminated.
  • the concept combinations are all related to the "sea", then when there are concept combinations related to the sea to choose from, other concept combinations can be eliminated.
  • they may all be reserved for subsequent steps.
  • a piece of text is formed according to the concept combination of the time instant and the duration of time within the time period.
  • the sequence includes a plurality of concepts with temporal attributes, and one of the temporal attributes is the duration of the concept.
  • Different concepts have different durations in the sequence.
  • the duration of a concept in a sequence is not entirely determined by the number of syllables, but by the association of the concept with the desired feedback in the brain. Therefore, in a piece of text formed by a combination of concepts, each concept has a definite start time and duration. This is also known as the temporal property of a concept in a sequence.
  • the concepts of "applause” and “take off” appeared in the concept combination.
  • the duration of the two can be different according to the time slice determined by the activation mode. For example, the duration of "applause” is greater than the duration of "take off”.
  • the starting moment of "elegance” is the 2nd second after the sequence playback starts, and the duration is 0.5 seconds; The start time is 2.5 seconds from the beginning of the sequence playback, and the duration is 0.2 seconds.
  • words are added to the paragraph to form content with coherent semantics. Since it is perceived in a natural way, a way that is more conducive to the brain's acceptance will also achieve better results. Therefore, it is necessary to form coherent semantics. In some embodiments, forming a certain theme is a more favorable technical solution.
  • auxiliary words and the like are added to the concept combination to coherent the semantics.
  • a concept corresponds to one or more phrases (or groups of words). For example, combine “graceful” and “jumping” into phrases like “jumping gracefully”; “flying” and “elf” into phrases such as “flying elves.” If there are two words of the same nature in the concept combination, they can also be directly juxtaposed. For example, “brave” and “strong” are directly combined as “strong and brave.” If the conceptual combination of two words is far apart, the combination can be based only on the properties of the words. Although difficult to understand, the semantics are still coherent. For example, “red” and “sit down” are combined directly with adjectives followed by the former entity word as “red sit down”; “temple” and “fly” are combined directly with nouns as “flying temple".
  • adverbs, prepositions, etc. are added to the contextual combination to coherent semantics. There may be a big difference between the before and after concept combinations, so adverbs, prepositions, etc. need to be added to make the pronunciation coherent. For example, the previous concept combination is “applause begins”, and the latter concept combination is “elegantly landed”, then the modified semantics of contextual coherence can be "beginning with applause, gracefully landed”.
  • a small number of entity words are added to the contextual concept combination to coherent the semantics.
  • entity words there may be a large difference between the before and after concept combinations, so it is necessary to add entity words to make the speech coherent. For example, when combining “red sit down”, “flying spirit” and “flying temple", you can make the following modification “sit down in red” , see “flying spirit” and “flying temple”; where “in” and “in” are added prepositions, and “see” is an added entity word.
  • the added particles, prepositions, adverbs, and entity words should be as short as possible in the sequence to minimize the impact of feedback on the original sequence definition.
  • the added entity words should be as common and habitual actions or things as possible.
  • the text may be modified by manual proofreading.
  • the text may be modified to be as close as possible to the subject matter related to the desired experience.
  • the timing and/or duration of the occurrence of one or more concepts within the time period is adjusted according to the coherent semantic content.
  • a piece of text may sound oddly rhythmic, affecting the feedback effect in the brain, if the timing and/or duration of the multiple concepts defined by the target feedback exactly occurs within the time period. Therefore, in some embodiments, the timing and/or duration of one or more concepts in the time period may need to be adjusted so that the entire text sounds more natural.
  • the adjustment should not exceed the preset range, otherwise the feedback in the cerebral cortex may be changed, and the effect of replicating the experience will not be achieved.
  • the adjustment of the start time does not exceed 20ms, and the adjustment of the duration does not exceed 50ms. This not only helps to adjust the reading rhythm of the text, but also does not deviate too much from the target feedback.
  • the sequence includes multiple concepts, each with its own defined start time and duration. Such text is suitable for human or machine reading or other natural way of being perceived by the brain.
  • the sequence is perceived by the brain in a natural manner for the length of time to generate the desired feedback in the brain.
  • the sequence can be perceived by the brain in a natural way, such as auditory, visual, tactile, so that feedback can be generated in the brain.
  • the sequence is played in a speech output device, or presented in a video, or in Braille.
  • the brain is able to perceive multiple concepts in the sequence at a defined moment onset and duration and generate desired feedback in the brain.
  • Feedback of hope is an analog of the sequential way in which the concept of goal feedback is combined, enabling the transplantation of goal experience from one human individual to another.
  • the methods of the present invention can be used for the treatment or prevention of psychological disorders or methods of psychiatric disorders including depression, major depression, treatment-resistant depression and treatment-resistant bipolar depression, bipolar depression, bipolar depression Phase disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression, postpartum depression, premenstrual dysphoric disorder (PMDD), situational depression, atypical depression, mania, anxiety, attention deficit Disorders (ADD), Attention Deficit Disorder with Hyperactivity (ADDH) and Attention Deficit/Hyperactivity Disorder (AD/HD), Bipolar and Manic Disorders, Obsessive Compulsive Disorder, Hyperphagia, Premenstrual Syndrome symptoms, substance addiction or abuse, nicotine addiction, psycho-sexual dysfunction, and pseudobulbar disorder.
  • the mental disorders include: schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette's disease one or more of.
  • the present invention also provides a system for generating feedback in the brain.
  • 10 is a functional block diagram of a system for generating feedback in the brain according to one embodiment of the present invention.
  • the system includes a brain keyboard 1 and a delivery device 2, the brain keyboard 1 being configured to generate a sequence of concepts having a length of time including the One or more concepts; the delivery device 2 is configured to transmit the sequence to the subject's brain for the time length, where desired feedback is generated.
  • the brain keyboard 1 includes the keyboard 11 and the processor 12 .
  • the keyboard 11 includes a plurality of keys, and at least one or more keys correspond to one or more concepts.
  • the keyboard 11 can be a physical keyboard, such as a common computer keyboard, which includes various letters, numbers, characters, etc., or a virtual keyboard, such as a virtual keyboard application running in the computer, which displays
  • the interface has multiple keys for the user to enter one or more concepts.
  • the keyboard further includes keys with a time length. When inputting a concept, you can also input the corresponding time, such as start time and duration.
  • the processor 12 is connected to the keyboard 11, receives key operations from the keyboard, and forms a concept sequence with a time length according to the key operations.
  • the processor 12 is connected to a delivery device 2, which receives the sequence of concepts generated by the processor 12 and sends it to a user, so that the sequence of concepts can be used by the user in a natural way over its length of time. Brain perception, resulting in desired feedback in the user's brain.
  • the transmitting device 2 may be a device transmitting visual, auditory or tactile information, such as a display, which converts the concept sequence into video information, so that the user's brain can perceive the concept sequence in the video information.
  • the transmitting device 2 is an audio playing device, which converts the conceptual sequence into audio information, especially voice information, so that the user's brain can perceive the conceptual sequence in the audio information or voice information.
  • the delivery device 2 can be made into a Braille reader that can be read by a blind person, through which the blind user can perceive the sequence of concepts, thereby generating the desired feedback in his brain.
  • a switch is provided in the conveying device 2 to control the time when the concept sequence is sent out.
  • FIG. 11 is a schematic block diagram of a system for generating feedback in the brain according to another embodiment of the present invention.
  • this embodiment also includes a data center 3 and a sample processing system 4.
  • the data center 3 stores various data, such as brain feedback sample data and corresponding concept sequences, and various sample data, Such as cerebral cortex data from different people and different experiences.
  • the data center 3 includes one or more databases, such as a database storing cerebral cortex data and corresponding human-related parameters from people with various experiences, a conceptual schema database, and the like.
  • the cerebral cortex data of each type of experience can be multiple people and multiple types of data. The data is obtained through various means, when information is collected from people's brains when they have a certain experience.
  • Said experience or experience includes, but is not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, beauty.
  • the human-related parameters include, but are not limited to, language, age, gender, religious belief, education level, occupation or previous occupation, and the like.
  • the cerebral cortex data can be fMRI data, and of course can also be MRI image data, CT image data, SEPECT image data, NIRS image data, fNIRS image data, PAI image data, EEG or magnetoencephalography data, and so on.
  • the sample processing system 4 includes at least a deep learning neural network model module 41 and a pattern recognition model module 42, and the deep learning neural network model module 41 is used to process cerebral cortex data obtained from different people in the same experience to obtain the brain.
  • Temporal changes in activation patterns in the cortex the data format for the temporal changes in activation patterns varies with the original data.
  • the pattern recognition model module 42 is used for comparing the change of the activation pattern of the cerebral cortex in the one time period with the activation pattern of a plurality of concepts in the cerebral cortex, so as to obtain the temporal variation of the activation pattern in the cerebral cortex corresponding to the cerebral cortex. Change one or more concepts, and get a concept sequence with time length sorted by the time of appearance and duration of the concepts.
  • the models used by the two model modules in this embodiment may be a combination of one or more of the product neural network CNN model, the deep belief network DBN model, and the recurrent neural network RNN model, or may be other various types, using Models of various other algorithms, those of ordinary skill in the art can implement the deep learning neural network model and pattern recognition model described in the present invention with reference to modeling methods in related fields, because the establishment of the model is not the focus of the present invention, It is not repeated here.
  • the present invention provides a method for the treatment or prevention of a mental illness or psychiatric disorder by utilizing the aforementioned method of generating feedback in the brain using the system provided in FIG. 10 or FIG. 11 , Mental illness or mental illness can be treated or prevented.
  • the mental disorders include: depression, major depression, treatment-resistant depression and treatment-resistant bipolar depression, bipolar disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression, postpartum depression Disorders, Premenstrual Dysphoria (PMDD), Situational Depression, Atypical Depression, Mania, Anxiety Disorders, Attention Deficit Disorder (ADD), Attention Deficit Disorder with Hyperactivity (ADDH), and ADHD Dysactivity disorder (AD/HD), bipolar and manic disorders, obsessive-compulsive disorder, bulimia, premenstrual syndrome, substance addiction or abuse, nicotine addiction, psycho-sexual dysfunction, and pseudobulbar one or more of the symptoms.
  • the mental disorders include: schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette's disease one or more of.
  • a system for generating feedback in the brain comprising generating means 1a and transmitting means 2a, the generating means 1a being configured to generate a feedback having a length of time a sequence of concepts in which the one or more concepts are included in a plurality of periods of the time length; the delivery device 2a is configured to transmit the sequence to the subject in a natural manner during the time length
  • the desired feedback in the subject's brain as perceived by the brain.
  • the production device 1a includes a target feedback determination device 11a and a target feedback analysis device 12a.
  • the target feedback determining means 11a determines the target feedback according to the experience to be obtained by the subject.
  • the production device 1a includes a target feedback database 16a, which includes target feedback data corresponding to any of the foregoing experiences.
  • the target feedback database 16a stores the feedback data formed in the brain of religious ceremonies after performing religious activities such as meditation and prayer.
  • the feedback data is the feedback data formed in the cerebral cortex for a period of time recorded by the brain signal detection technology.
  • the target feedback database 16a records the feedback data formed in the brain of an outstanding athlete for a period of time before participating in the competition event he is good at.
  • the feedback data includes, but is not limited to, MRI image data, fMRI image data, CT image data, SEPECT image data and PAI image data, fNIRS image data, and the like.
  • the target feedback determination device 11a includes a target feedback deep learning neural network model.
  • the neural network model inputs human-related parameters and experience of an individual (subject) according to the training result, and outputs target feedback that best matches the experience of the individual.
  • the people-related parameters include the individual's language, age, gender, religious belief, education level, occupation or previous occupation, and the experience includes the aforementioned relaxation, calmness, self-confidence, pleasure, satisfaction, courage, Any of health, excitement, success, beauty, etc.
  • the target feedback analysis device 12a includes a target feedback time slicing module 120a and an activation mode correlation analysis module 121a. After obtaining the target feedback, the target feedback determination device 11a sends it to the target feedback analysis device 12a.
  • the target feedback time slice module 120a of the target feedback analysis device 12a acquires the target feedback data, and slices it according to a certain time period to obtain time slices with the same time period, as shown in FIG. 6 . Wherein, the time period of the time slice cannot be too large or too small, and in one embodiment, 10-50 ms is better.
  • the activation pattern correlation analysis module 121a performs correlation analysis of the activation pattern for each time slice, for example, determines the activation pattern of the cerebral cortex in each time slice, the activation pattern includes the spatial distribution of the activated part and the signal intensity, and then recalculates Whether the activation patterns of the cerebral cortex in adjacent time slices are independent of each other. For example, compare whether the spatial distribution of the activation part of the cerebral cortex in two time slices is the same or whether the difference is within the allowable range, and then compare the signal intensity of the corresponding activation part. The difference in the spatial distribution of the activation part and the difference in signal intensity are both Within the allowable range, it is determined that the two are not independent, otherwise the two are independent. After the correlation analysis of the activation patterns, a plurality of corresponding time slices are obtained, and each time slice corresponds to an activation pattern, as shown in FIG. 7 .
  • the target feedback analysis device 12a includes a concept combination identification module 122a.
  • the production device 1a includes a conceptual pattern database 17a, which stores feedback data formed in the brain by the combination of one or more concepts, ie, corresponding to the distribution of activation areas in the cerebral cortex and their signal strengths .
  • the concept combination recognition module 122a includes a concept recognition neural network model. According to the training result, the neural network model inputs an activation pattern of the target feedback, such as data corresponding to a time slice in Figure 7, and outputs one or more concept combinations matching the activation pattern.
  • the production device 1a further includes a semantic module 13a configured to modify the sequence of concepts to form coherent semantics.
  • the concept combination identification module 122a obtains a concept combination sorted by time, and each concept has time attributes, such as time and duration.
  • time attributes such as time and duration.
  • auxiliary words, prepositions, adverbs, content words, etc. to combine the semantics of the concepts described in coherence. For example, adding “please” and “de” to the concepts "listen”, “gurgling” and “flowing water” to obtain the text "please listen to the gurgling water” with coherent semantics.
  • the timing and/or duration of the multiple concepts' occurrence within the time period is adjusted after additional vocabulary has been added.
  • the delivery device 2a includes one or more devices for hearing, vision, or touch, so that the brain perceives the sequence in a natural way, or a combination thereof, correspondingly, the system further includes: Conversion device 3a.
  • the conversion device 3a includes an audio conversion module 30a, a video conversion module 31a, a Braille conversion module 32a, and the like.
  • the audio conversion module 30a is used to convert the text with coherent semantics representing the target feedback of an experience into corresponding audio information. For example, voice messages read aloud by humans or machines. Further, corresponding background music can also be configured for the voice information according to the desired experience, for example, Chinese and foreign classical music, various natural onomatopoeia, and the like.
  • the video conversion module 31a selects a corresponding piece of video or multiple pieces of video in the material library according to the text with coherent semantics fed back by the target representing an experience, and connects the multiple pieces of video into a whole piece of video image, and according to the transmission device format, and store it in the corresponding format.
  • the Braille conversion module 32a can convert text with coherent semantics representing target feedback of an experience into outputable Braille.
  • the above-mentioned conversion device 3a can also be integrated in the production device 1a as a conversion module of the production device 1a.
  • the transmission device 2a can be various, for example, the audio playback device 20a for hearing includes but is not limited to speakers, earphones, audio players, etc.; wherein, the video playback device 21a for vision includes but is not limited to various displays, such as Desktop computer monitors, laptop computer monitors; various mobile terminal displays, such as mobile phone displays, flat panel displays, etc.; various large-screen displays; Braille printers, electronic Braille readers, etc.
  • the transmission device 2a may also be a device that has both audio and video playback, such as a VR/AR device, including but not limited to VR/AR glasses, VR/AR head-mounted displays, and the like.
  • a VR/AR device including but not limited to VR/AR glasses, VR/AR head-mounted displays, and the like.
  • the transmission device 2a may also include some professional and non-professional audio-visual rooms and studios.
  • the audio-visual room may be a professional or home audio-visual room including audio playback equipment, video playback equipment, signal source equipment, etc.
  • the production device 1a sends the converted audio and video information to the signal source equipment of the video room, then the audio-visual room can be used in the audio-visual equipment.
  • the room causes the subject to perceive the concepts in the sequence auditorily and visually, thereby generating the desired feedback in their brains and obtaining the desired experience.
  • Studio can be professional or non-professional studios including cameras (such as full-frame cameras or digital backs), lenses, lights, curtains, background props, etc. In the studio, by means of actors' performances, subjects are allowed to perceive the concepts in the sequence in auditory and visual ways, so as to generate desired feedback in their brains and obtain desired experiences.
  • the system for generating feedback in the brain further includes a storage device 4a, as shown in the system block diagram of FIG. 14 .
  • the storage device 4a is configured to store brain feedback sample data and corresponding conceptual sequences, which may be one or more of local memory, local database, and cloud database.
  • the stored brain feedback sample data and the corresponding concept sequence, the aforementioned target feedback database, concept pattern database, etc. are all located in the cloud database, and the production device 1a, when needed, uses any existing communication mechanism to send data from Obtaining the required data from the cloud database can save local storage space.
  • some modules in the production device 1a can be placed in the cloud, and the powerful cloud computing can be used to obtain the required modules, which can be executed by the specific transmission device. audio and video information.
  • the system for generating feedback in the brain also includes a sample processing device 5a configured to obtain corresponding brain feedback sample data and corresponding concept sequences using cerebral cortex data obtained from different people during the same experience as samples.
  • the obtained brain feedback sample data and the corresponding concept sequence are stored in the storage device 4a.
  • the sample processing system 5a includes a deep learning neural network model module 51a and a pattern recognition model module 52a, the deep learning neural network model module 51a being configured to process cerebral cortex data obtained from different people while having the same experience , to obtain the temporal change of the activation pattern in the cerebral cortex; the pattern recognition model module 52a compares the change in the activation pattern of the cerebral cortex with the activation pattern of multiple concepts in the cerebral cortex in the one time period, To obtain one or more concepts corresponding to the temporal changes of the activation patterns in the cerebral cortex, and to obtain a concept sequence with a time length by sorting the concept's appearance time and duration.
  • a brain keyboard is proposed, as shown in FIG. 10, which includes a keyboard 11 and a processor 12, the keyboard includes a plurality of keys, at least one or more keys correspond to one or more keys a concept; the processor forms a sequence having a time length by receiving key operations from the keyboard, wherein the sequence includes the one or more concepts for a plurality of periods of the time length; wherein the sequence is Being perceived by the brain in a natural manner for that length of time produces a desired feedback in the brain.
  • the speaker By taking the brain feedback of Buddhist believers after meditation as the target feedback of the "deep relaxation” experience, the speaker repeatedly played the sequence of "deep relaxation” obtained by the method of the present invention to the subjects by voice for 13 minutes.
  • the subject's deep relaxation evaluation index was used to determine whether the desired feedback was formed; the evaluation index included physical fatigue recovery ability, sleep quality and the proportion of deep relaxation presented by brain waves.
  • the deep relaxation concept sequence was broadcast live at a fixed time for 13 minutes every day for 7 consecutive days, and questionnaires were distributed to the subjects to evaluate their physical fatigue recovery ability. A total of 120 valid questionnaires were harvested after 7 days.
  • the gender ratio of the subjects is 5:1; 77% of the subjects are adults between the ages of 30 and 50; 40% have a bachelor's degree or above, of which 6 have master's degrees and 2 have doctoral degrees; Urban users accounted for 21.67%.
  • Another 10 experienced meditators served as a reference group. The 10 people in the reference group used their own methods to meditate and relax for 40 minutes a day.
  • Brainlink Lite is a smart headband EEG acquisition smart device produced by Shenzhen Hongzhi Technology Co., Ltd. The device is a single-channel headband with a sampling rate of 512, with 3 dry electrodes on the forehead.
  • FIG. 15 it is a schematic diagram of the average score of physical fatigue recovery ability evaluation according to an embodiment of the present invention. From the average score of 5.53 before the test to the average score of 7.41 on the seventh day of the test, an increase of 34.00%.
  • Figure 16 is a schematic diagram of the average score of the test group's assessment of physical fatigue recovery ability during the four statistics. It can be seen from the figure that after the first day of testing, the physical fatigue recovery ability of the 32 test group members increased from the previous average of 4.50 points to 5.19 points, an increase of 15.33%; the physical fatigue recovery ability after 7 days increased from the previous 4.50 points To 6.56 points, an increase of 45.78%; the physical fatigue recovery ability after 14 days, from the previous 4.50 points to 7.63 points, an increase of 69.56%.
  • Figure 17 is the average score of the evaluation of the physical fatigue recovery ability of the reference group in the second round of testing at 4 counts. After the first day of meditation, it rose to 5.5 from 5.2 on the previous day, an increase of 5.77%; after 7 consecutive days, the physical fatigue recovery ability increased from the previous 5.2 to 6.1, an increase of 17.31%; after 14 consecutive days, the physical fatigue recovery ability increased from before The 5.2 rose to 6.8, an increase of 30.77%.
  • FIG. 18 is a schematic diagram of the ratio of the average deep relaxation degree of the brain waves calculated according to the brain wave data of the target test group.
  • the average deep relaxation ratio of the brainwaves on the day before the test was 7.53%; the average deep relaxation ratio of the brainwaves after the test on the first day was 10.75%; on the seventh day, it was 11.66%.
  • the proportion of deep relaxation in brainwaves showed an increasing trend over time as a whole.
  • FIG. 19 is the ratio of the average deep relaxation degree of the brain wave calculated according to the brain wave data of the reference group.
  • the average proportion of deep relaxation before the first day of meditation and brain waves was 1.50%, after the first day of meditation, the average proportion of deep relaxation was 4.20%; after 7 days, it was 1.9%, and there was no increasing trend.
  • Brain wave data is the most direct data reflecting deep relaxation.
  • the brain wave data obtained after applying the method of the present invention reflects the successful transplantation of the deep relaxation experience of meditation.
  • the increasing trend indicates that as multiple applications of the method of the present invention generate feedback in the brain, this experience is increasingly becoming an experience of the subject itself.
  • Figure 20 is a schematic diagram of the sleep quality scores of 120 people in the first round of tests.
  • the 7-day sleep quality score increased from the previous average of 5.8 to an average of 7.58, an increase of 30.69%.
  • FIG. 21 is a schematic diagram of the sleep quality score values of 32 people in the target test group in the second round of testing.
  • the average score of sleep quality on the day before the test was 5.25 points. After the first day of testing, it increased to 5.88 points, an increase of 12.00%; after 7 days, it increased to an average of 6.75 points, an increase of 28.57%; after 14 days, it increased to an average of 7.5 points, an increase of 28.57%. up 42.86%.
  • the sleep quality of the target test group gradually increased over time.
  • Figure 22 is a schematic diagram of the sleep quality score values of 10 people in the reference group in the second round of testing.
  • the average score of sleep quality before meditating was 6.2 points. After the first day of meditation, it increased to 6.5 points, an increase of 4.84%; after 7 days, it increased to an average of 6.5 points, an increase of 16.13%; after 14 days, it increased to an average of 7.4 points, an increase of 19.35 points. %.
  • sleep quality reflects the degree of deep relaxation, and on the other hand, it also reflects the overall changes in the subjects after experience transplantation. Due to the widespread nature of sleep problems, it is very common for subjects to have sleep problems, which is one of the reasons for their willingness to participate in this test. The application effect of this method reflects that such experience transplantation is not short-term and fleeting, but can produce changes to the subject as a whole. This is almost the same as the subjects themselves getting a similar experience. Further, the effect of such a global change on the treatment and prevention of a psychological disorder or psychiatric disorder in a subject is also clear.
  • the "calm” experience sequence was sent to the subjects in the form of voice through the speaker, and the duration of the sequence was repeated for 15 minutes.
  • the hopeful feedback is that subjects are able to reduce anxiety, increase inner peace, and reduce physical problems.
  • the calm concept sequence was played at a fixed time for 15 minutes every day for 7 consecutive days, and a questionnaire was distributed to 120 subjects to assess their anxiety level.
  • Another 10 experienced meditators served as a reference group. The 10 people in the reference group used their own methods to meditate and relax for 40 minutes a day.
  • Figure 23 is a schematic diagram of the average reverse score of the STAI Spielberg Anxiety Scale for 120 subjects in the first round of testing. After 7 consecutive days of testing, the level of calm and reassurance changed from the previous average of 2.63 to 1.84, an improvement of 30.04%.
  • FIG. 24 is the average reverse score of the STAI Spielberg Anxiety Scale for the subjects of the 32 target test groups in the second round of testing.
  • the level of peace of mind changed from 2.53 to 2.47, an improvement of 2.37%; after 7 days, the level of peace of mind changed from the previous average of 2.53 to 2.06, an improvement of 18.58%; after 14 days, the level of peace of mind From the previous average of 2.53 points to 1.70 points, an improvement of 32.81%.
  • the level of calm and reassurance showed an increasing trend over time.
  • Figure 25 is the average of the reverse scores of the STAI Spielberg Anxiety Scale for 10 subjects in the reference group in the second round of testing. Subjects in the reference group changed from 2.6 to 2.5 the day before meditation, an improvement of 3.85%; after 7 days, the level of peace of mind changed from 2.6 to 2.4, an improvement of 7.69%; after 14 days, the level of peace of mind changed from 2.6 to 2.3, Improved by 11.54%.
  • Figure 26 is a schematic diagram of the reverse scoring average of the physical problem assessment form for 120 subjects in the first round of testing.
  • 120 subjects completed the test, and the severity of the physical problem (reverse scoring) score changed from the previous 7.32 points to 4.31 points, an improvement of 41.12%.
  • FIG. 27 is a schematic diagram of the average positive score of the physical problem improvement assessment form of the 32 subjects in the target test group in the second round of testing.
  • the physical problems of the 32 subjects in the target test group improved from 3.90 points on the day before the start to 4.56 points on the first day of testing, an increase of 16.92%; after 7 days, it rose to 4.56 points. 5.78, an increase of 48.21%; 14 days later, it rose to 7.59, an increase of 94.12%.
  • Figure 28 is a schematic diagram of the average positive score of the physical problem improvement assessment form for 10 subjects in the reference group in the second round of testing.
  • the improvement of the physical problems of the subjects in the reference group increased from 4.8 points on the day before the start to 5.4 on the first day, an increase of 12.50%; after 7 days, it increased to 5.5, an increase of 14.58%; after 14 days, it increased to 6.4, An increase of 33.33%.
  • the method provided by the present invention can more effectively improve physical problems.
  • the marked improvement in anxiety, as well as physical problems due to anxiety, illustrates the high efficacy of the methods of the present invention in the treatment of psychological and psychiatric disorders.
  • the method of the present invention can of course be better applied to prevent the occurrence of these problems.
  • the methods of the present invention are expected to produce better therapeutic and prophylactic effects if used in combination with other drugs.

Abstract

A method, a system, and a brain keyboard for generating a feedback in a brain. Said method comprises: determining a sequence having a length of time, wherein the sequence comprises one or more concepts within multiple periods of the length of time; and the brain sensing the sequence in a natural manner within the length of time so as to produce a desired feedback in the brain. The system comprises a generation apparatus and a transmission apparatus. The brain keyboard comprises a keyboard and a processor, wherein the keyboard comprises multiple keys, and at least one or more keys correspond to one or more concepts; the processor is configured to receive key operations from the keyboard to form a sequence having a length of time, wherein the sequence comprises the one or more concepts within multiple periods of the length of time. The method and the system have low cost, easy operations and no harm to a subject.

Description

在大脑中产生反馈的方法、系统及大脑键盘Method, system and brain keyboard for generating feedback in the brain 技术领域technical field
本发明涉及一种脑科学，特别地涉及一种在大脑中产生反馈的方法、系统及大脑键盘。The present invention relates to a brain science, in particular to a method, a system and a brain keyboard for generating feedback in the brain.
背景技术Background technique
大脑是人体中枢神经系统的主要组成部分，包括大约860亿个神经细胞(神经元)已经千亿计的胶质细胞。每一个神经元通常拥有几百个到几千个神经突触，大脑的突触数量估计约有10 15个之多。一个成年人的大脑重量约为1.2-1.6千克，主要成分是血液。虽然大脑的重量只是人体重量的2％-4％，但是其耗氧量却能占到总耗氧量的1/4。大脑的血流量占到心脏输出总血量的15％，其消耗的功率大约为25W。 The brain is the main component of the human central nervous system, including about 86 billion nerve cells (neurons) and hundreds of billions of glial cells. Each neuron usually has hundreds to thousands of synapses, and the number of synapses in the brain is estimated to be about 10 15 . The weight of an adult brain is about 1.2-1.6 kg, and the main component is blood. Although the weight of the brain is only 2%-4% of the body weight, its oxygen consumption can account for 1/4 of the total oxygen consumption. The blood flow of the brain accounts for 15% of the total blood output of the heart, and its power consumption is about 25W.
大脑是人体最为复杂的器官，是接受外界刺激、产生感觉、形成意识和思维，发出指令和驱动行动的“指挥部”。大脑皮层是高级神经活动的物质基础，是产生思维的组织。大脑皮层的左右半球可以分为5叶：额叶、颞叶、顶叶、枕叶和岛叶；其中，传统上认为额叶和颞叶被认为与语言、情感、记忆有关。然而，对于语言中枢的表象以及记忆的形成和检索的生理基础，情绪的形成和影响等的研究还都处于初期阶段。The brain is the most complex organ of the human body. It is the "command center" that accepts external stimuli, generates sensations, forms consciousness and thinking, and issues instructions and drives actions. The cerebral cortex is the material basis of higher neural activity and the organization that produces thinking. The left and right hemispheres of the cerebral cortex can be divided into 5 lobes: frontal lobe, temporal lobe, parietal lobe, occipital lobe and insula lobe; among them, the frontal lobe and temporal lobe are traditionally considered to be related to language, emotion, and memory. However, the research on the representation of the language center and the physiological basis of memory formation and retrieval, the formation and influence of emotions, etc. are still in the early stage.
目前，脑科学研究的三个主要方向是大脑的结构和功能研究(特别是大脑的高级功能)、脑疾病的研究和脑应用的研究。脑应用的研究是脑科学领域的一个重点领域，其旨在解析大脑的神经系统结构和心理活动的物质基础的同时，利用信息科学、计算机科学等方法和手段，开发大脑高级功能的算法或模型，推动人工智能、机器人等领域的发展。At present, the three main directions of brain science research are the study of the structure and function of the brain (especially the higher functions of the brain), the study of brain diseases and the study of brain applications. The research of brain application is a key field in the field of brain science. It aims to analyze the nervous system structure of the brain and the material basis of psychological activities, and to develop algorithms or models of advanced brain functions by using methods and means such as information science and computer science. , to promote the development of artificial intelligence, robotics and other fields.
然而，利用大脑的结构和功能研究的生理基础，以脑应用研究的方法和手段解决脑疾病领域的有关问题的跨领域研究和应用成果在本领域中却未有报道。However, the cross-domain research and application achievements that utilize the physiological basis of brain structure and function research to solve related problems in the field of brain diseases with the methods and means of brain application research have not been reported in this field.
发明内容SUMMARY OF THE INVENTION
针对现有技术中存在的技术问题，本发明提出了一种在大脑中产生反馈的方法、系统和大脑键盘，用以在所述大脑中产生希望的反馈。In view of the technical problems existing in the prior art, the present invention proposes a method, a system and a brain keyboard for generating feedback in the brain, so as to generate desired feedback in the brain.
为了解决上述技术问题，根据本发明的一个方面，本发明提供了一种在大脑中产生反馈的方法，其包括：确定具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括一个或多个概念(concept)；以及将所述序列在所述时间长度内以自然方式被大脑感知，以在所述大脑中产生希望的反馈(desired feedback)。In order to solve the above-mentioned technical problem, according to one aspect of the present invention, the present invention provides a method for generating feedback in the brain, comprising: determining a sequence having a time length, wherein all time periods within the time length are The sequence includes one or more concepts; and the sequence is perceived by the brain in a natural manner for the length of time to generate desired feedback in the brain.
根据本发明的另一个方面，本发明提供了一种在大脑中产生反馈的系统，其包括生成装置和传送装置，所述生成装置经配置以生成一个具有一个时间长度的概念序列，其中在所述时间长度的多个时段内包括所述一个或多个概念；所述传送装置经配置以将所述序列在所述时间长度内以自然的方式被受试者的大脑所感知，在受试者的大脑中的产生希望的反馈。According to another aspect of the present invention, there is provided a system for generating feedback in the brain, comprising generating means and transmitting means, the generating means being configured to generate a sequence of concepts having a temporal length, wherein the The one or more concepts are included in a plurality of periods of the time length; the delivery device is configured to cause the sequence to be perceived by the subject's brain in a natural manner during the time length, in the subject's hope-generating feedback in the brain of the user.
根据本发明的另一个方面，本发明提供了一种大脑键盘，其包括键盘和处理器，所述键盘包括多个按键，至少一个或多个按键对应一个或多个概念；所述处理器经配置以接收来自所述键盘的按键操作形成具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括所述一个或多个概念；其中，所述序列在所述时间长度内以自然方式被大脑感知在所述大脑中产生希望的反馈。According to another aspect of the present invention, the present invention provides a brain keyboard, which includes a keyboard and a processor, the keyboard includes a plurality of keys, at least one or more keys correspond to one or more concepts; the processor is processed by configured to receive keystrokes from the keyboard to form a sequence having a length of time, wherein the sequence includes the one or more concepts for a plurality of periods of the length of time; wherein the sequence is at the time Perceived by the brain in a natural manner over length produces desired feedback in the brain.
本发明通过前述方法，可以自然的方式使受试者大脑感知一段概念序列，以在受试者大脑中产生希望的反馈，所述希望的反馈对应着希望的经验，例如 放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽等。本发明提供的方法及系统不但可以使普通人感受到其所希望的种种经验，还能够用于治疗或预防心理性疾病或精神疾病。本发明提供的方法和系统成本低，对受试者无伤害，易操作，效果显著。Through the aforementioned method, the present invention can make the subject's brain perceive a sequence of concepts in a natural way, so as to generate desired feedback in the subject's brain, and the desired feedback corresponds to a desired experience, such as relaxation, calm, self-confidence, Joy, contentment, bravery, health, excitement, success, beauty, etc. The method and system provided by the present invention can not only make ordinary people feel the various experiences they want, but also can be used to treat or prevent psychological diseases or mental diseases. The method and system provided by the present invention have low cost, no harm to subjects, easy operation and remarkable effect.
附图说明Description of drawings
下面，将结合附图对本发明的优选实施方式进行进一步详细的说明，其中：Below, the preferred embodiments of the present invention will be described in further detail in conjunction with the accompanying drawings, wherein:
图1是根据本发明的一个实施例中的MRI设备的一个简要结构示意图；FIG. 1 is a schematic structural diagram of an MRI apparatus according to an embodiment of the present invention;
图2是Huth等人在文章《自然语音揭示了平铺人类大脑皮层的语义图》中公开的语素方面的模型示意图；Figure 2 is a schematic diagram of the model in terms of morphemes disclosed by Huth et al.
图3是Huth等人在文章《自然语音揭示了平铺人类大脑皮层的语义图》中公开的语素方面的语义模型的主元分析示意图；Figure 3 is a schematic diagram of the principal component analysis of the semantic model in terms of morphemes disclosed by Huth et al.
图4是根据本发明的一个实施例的一种在大脑中产生反馈的方法流程图；4 is a flow chart of a method for generating feedback in the brain according to an embodiment of the present invention;
图5是根据本发明的一个实施例的对目标反馈按时间划分为多个激活模式的方法流程图；5 is a flowchart of a method for dividing target feedback into multiple activation modes by time according to an embodiment of the present invention;
图6是根据本发明的一个实施例的目标反馈按时间进行切片后得到时间切片示意图；6 is a schematic diagram of a time slice obtained after the target feedback is sliced by time according to an embodiment of the present invention;
图7是根据本发明的一个实施例的对时间切片识别、合并后得到时间切片示意图；FIG. 7 is a schematic diagram of obtaining time slices after identifying and merging time slices according to an embodiment of the present invention;
图8是根据本发明的一个实施例的获取与一个时间切片的激活模式对应的一个或多个概念的示意图；8 is a schematic diagram of acquiring one or more concepts corresponding to an activation mode of a time slice according to an embodiment of the present invention;
图9是根据本发明的一个实施例的多个概念的时序示意图；9 is a timing diagram of various concepts according to one embodiment of the present invention;
图10是根据本发明一个实施例的在大脑中产生反馈的系统的原理框图；10 is a schematic block diagram of a system for generating feedback in the brain according to an embodiment of the present invention;
图11是根据本发明另一个实施例的在大脑中产生反馈的系统的原理框图；11 is a schematic block diagram of a system for generating feedback in the brain according to another embodiment of the present invention;
图12是根据本发明又一个实施例的在大脑中产生反馈的系统的原理框图；12 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention;
图13是根据本发明再一个实施例的在大脑中产生反馈的系统的原理框图；13 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention;
图14是根据本发明又一个实施例的在大脑中产生反馈的系统的原理框图；14 is a schematic block diagram of a system for generating feedback in the brain according to yet another embodiment of the present invention;
图15是根据本发明的一个实施例的第一轮测试受试者身体疲劳恢复能力评估分数示意图；Fig. 15 is a schematic diagram of the physical fatigue recovery ability evaluation score of the first round of test subjects according to an embodiment of the present invention;
图16是根据本发明的一个实施例的第二轮测试时目标测试组身体疲劳恢复能力评估分数示意图；Figure 16 is a schematic diagram of the physical fatigue recovery ability evaluation score of the target test group during the second round of testing according to an embodiment of the present invention;
图17是根据本发明的一个实施例的第二轮测试时参照组身体疲劳恢复能力评估分数示意图；FIG. 17 is a schematic diagram of the physical fatigue recovery ability evaluation score of the reference group in the second round of testing according to an embodiment of the present invention;
图18是根据本发明的一个实施例的第二轮测试时目标测试组的脑波平均深度放松程度占比值示意图；FIG. 18 is a schematic diagram of the ratio of the average deep relaxation degree of brain waves of the target test group during the second round of testing according to an embodiment of the present invention;
图19是根据本发明的一个实施例第二轮测试时参照组的脑波平均深度放松程度占比值示意图；Fig. 19 is a schematic diagram of the ratio of the average deep relaxation degree of brain waves of the reference group during the second round of testing according to an embodiment of the present invention;
图20是根据本发明的一个实施例第一轮测试受试者睡眠质量评分值示意图；20 is a schematic diagram of the sleep quality score value of the first round of testing subjects according to an embodiment of the present invention;
图21是根据本发明的一个实施例第二轮测试时目标测试组的睡眠质量评分值示意图；21 is a schematic diagram of the sleep quality score value of the target test group during the second round of testing according to an embodiment of the present invention;
图22是根据本发明的一个实施例第二轮测试时参照组的睡眠质量评分值示意图；22 is a schematic diagram of the sleep quality score value of the reference group during the second round of testing according to an embodiment of the present invention;
图23是根据本发明的一个实施例第一轮测试时受试者的STAI斯皮尔伯格焦虑量表反向计分平均值示意图；Figure 23 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale of the subject during the first round of testing according to an embodiment of the present invention;
图24是根据本发明的一个实施例第二轮测试目标测试组的STAI斯皮尔伯格焦虑量表反向计分平均值示意图；Figure 24 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale in the second round of the test target test group according to an embodiment of the present invention;
图25是根据本发明的一个实施例第二轮测试参照组的STAI斯皮尔伯格焦虑量表反向计分平均值示意图；Figure 25 is a schematic diagram of the reverse score average value of the STAI Spielberg Anxiety Scale in the second round of the test reference group according to an embodiment of the present invention;
图26是根据本发明的一个实施例第一轮测试受试者的身体问题评估表的 反向计分平均值示意图；Figure 26 is a schematic diagram of the reverse scoring average of the physical problem assessment form of the first round of test subjects according to one embodiment of the present invention;
图27是根据本发明的一个实施例第二轮测试目标测试组的身体问题改善评估表的正向计分平均值示意图；以及Figure 27 is a schematic diagram of the average positive score of the physical problem improvement assessment form for the second round of test target test groups according to one embodiment of the present invention; and
图28是根据本发明的一个实施例第二轮测试参照组的身体问题改善评估表的正向计分平均值示意图。FIG. 28 is a schematic diagram of the average positive score of the physical problem improvement evaluation form of the second-round test reference group according to an embodiment of the present invention.
具体实施方式Detailed ways
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
在以下的详细描述中，可以参看作为本申请一部分用来说明本申请的特定实施例的各个说明书附图。在附图中，相似的附图标记在不同图式中描述大体上类似的组件。本申请的各个特定实施例在以下进行了足够详细的描述，使得具备本领域相关知识和技术的普通技术人员能够实施本申请的技术方案。应当理解，还可以利用其它实施例或者对本申请的实施例进行结构、逻辑或者电性的改变。In the following detailed description, reference may be made to the accompanying drawings, which are considered a part of this application to illustrate specific embodiments of the application. In the figures, like reference numerals describe substantially similar components in the different figures. The specific embodiments of the present application are described in sufficient detail below to enable those of ordinary skill with relevant knowledge and technology in the art to implement the technical solutions of the present application. It should be understood that other embodiments may also be utilized or structural, logical or electrical changes may be made to the embodiments of the present application.
地球上几十亿年的生命进化历程可以总结为两种基本进化方式：一种为实现基因信息传递的生物学进化方式，即由核苷酸序列组成的基因组按照一定的规则生成蛋白质和细胞组织，从而产生无限多样性的生命有机体；另一种为通过语言实现非基因信息传递的进化方式。后一种进化方式使人类区别于其他生命体，具有了思维，并促进了文化上的进步和繁荣。The evolution of life on earth for billions of years can be summed up in two basic evolutionary ways: one is the biological evolutionary way to realize the transmission of genetic information, that is, the genome composed of nucleotide sequences generates proteins and cell tissues according to certain rules. , resulting in an infinite variety of living organisms; the other is an evolutionary way of realizing non-genetic information transmission through language. The latter way of evolution distinguishes human beings from other living beings, has a mind, and promotes cultural progress and prosperity.
根据蔡曙山主编、2015年人民出版社出版的《人类的心智与认知》一书所总结的，从当代基因科学来看，心智从低级到高级的进化也决定了生命从低级 到高级的进化。生命在心智方面的进化包括从低级到高级的五个层级：神经层级、心理层级、语言层级、思维层级和文化层级；其中的神经层级和心理层级是人和动物共有的，可称为低阶认知层级；语言层级、思维层级和文化层级为人类特有的，也称为高阶认知层级。语言是人类认知的基础，在抽象概念语言的基础上形成了人类思维，语言和思维共同构建了人类的知识体系，甚至定义了整个人类社会，而知识的积淀又形成了文化，文化的繁荣也反过来促进了人类社会的发展。According to the book "Human Mind and Cognition" edited by Cai Shushan and published by People's Publishing House in 2015, from the perspective of contemporary genetic science, the evolution of mind from low to high also determines the evolution of life from low to high . The evolution of life in terms of mind includes five levels from low to high: neural level, psychological level, language level, thinking level and cultural level; the neural level and psychological level are shared by humans and animals, which can be called low-level Cognitive level; language level, thinking level and cultural level are unique to human beings, also known as higher-order cognitive level. Language is the basis of human cognition. Human thinking is formed on the basis of abstract conceptual language. Language and thinking together build the human knowledge system, and even define the entire human society. The accumulation of knowledge forms culture and cultural prosperity. In turn, it promotes the development of human society.
从20世纪开始，随着脑科学的发展，人们从神经、心理、语言、思维和文化各个层级对人类认知展开了全面的研究，从而出现了研究侧重点各不同的脑科学体系分支。其中，神经语言学是一个非常重要的研究领域。神经语言学研究语言和大脑功能之间的关系，其目的在于解释人类语言的理解、产生、习得以及学习的神经和心理机制，其研究对象是人类神经系统与人类语言之间的互动关系，理解人脑如何接收、存储、加工和提取语言信息。Since the 20th century, with the development of brain science, people have carried out comprehensive research on human cognition from all levels of neurology, psychology, language, thinking and culture, resulting in the emergence of branches of brain science systems with different research focuses. Among them, neurolinguistics is a very important research field. Neurolinguistics studies the relationship between language and brain function. Its purpose is to explain the neural and psychological mechanisms of human language understanding, production, acquisition, and learning. The research object is the interaction between the human nervous system and human language. Understand how the human brain receives, stores, processes and extracts linguistic information.
1987年，卡普兰(Caplan，D.)的《神经语言学和语言学语言障碍：导论》(由剑桥大学出版社于1987年出版)是神经语言学发展初期的代表性著作，探讨了大脑不同的区域与语言之间的关系。施特莫尔(Stemmer，B.)和惠特克(Whitaker，HA)在1998年出版了《神经语言学手册》(由美国马萨诸塞州学术1998出版社)一书，总结了神经语言学的进展，对语言处理的大脑机制进行了探讨。神经语言学领域的代表著作是荷兰歌德堡大学教授伊丽莎白·艾尔森(Elisabeth Ahlsén)和约翰·本杰明斯(John Benjamins)在2006年发行的《神经语言学导论》(Ahlsen.E.(2006).Introduction to Neurolinguistics.)。在该书中两位作者总结了利用各种方法和技术对语言处理和语言学习的大脑机制研究的最新发展。例如，在处理语言信息时大脑会遵循的过程，不同大脑区域在语言处理中的相互作用，以及当受试者产生或感知其母语以外的其他语言时大脑激活的位置等等。虽然神经语言学取得了很多进步，然而对于记忆的形 成和搜索过程以及语言的形成机制的理解却仍停留在比较宏观的阶段，并未形成神经网络层面的有效描述。In 1987, Caplan, D.'s "Neurolinguistics and Linguistics Language Disorders: An Introduction" (published by Cambridge University Press in 1987) is a representative work in the early development of neurolinguistics, exploring different brains. the relationship between region and language. Stemmer (B.) and Whitaker (Whitaker, HA) published the book "Handbook of Neurolinguistics" (Massachusetts Academic Press 1998) in 1998, summarizing the progress of neurolinguistics , which explores the brain mechanisms of language processing. The representative work in the field of neurolinguistics is the "Introduction to Neurolinguistics" published in 2006 by professors Elisabeth Ahlsén and John Benjamins of Gothenburg University in the Netherlands (Ahlsen.E.(2006) .Introduction to Neurolinguistics.). In this book, the authors summarize recent developments in the study of the brain mechanisms of language processing and language learning using various methods and techniques. For example, the processes that the brain follows when processing linguistic information, the interaction of different brain regions in language processing, and where the brain activates when subjects produce or perceive languages other than their native language. Although a lot of progress has been made in neurolinguistics, the understanding of the formation and search process of memory and the formation mechanism of language is still at a relatively macro stage, and an effective description at the neural network level has not been formed.
人类大脑中的大脑皮质是高级神经活动的物质基础，是产生思维的器官，其主导着肌体内部的一切活动，并协调肌体与外部环境的平衡。大脑半球借助于大脑皮质表面呈现出的沟与壑分成五个部分，即额叶、颞叶、枕叶、顶叶和岛叶，他们各有一定的机能分工，例如额叶与推理、计划、情感、部分语言及运动有关；颞叶与感知、听觉和记忆有关；枕叶与视觉有关；顶叶与触觉、温度、压力和疼痛等有关；岛叶与脑干的自主功能有关，同时还处理味觉信息。The cerebral cortex in the human brain is the material basis of advanced neural activities and the organ that produces thinking, which dominates all activities within the body and coordinates the balance between the body and the external environment. The cerebral hemisphere is divided into five parts by means of the sulci and gullies on the surface of the cerebral cortex, namely the frontal lobe, the temporal lobe, the occipital lobe, the parietal lobe and the insular lobe. The temporal lobe is related to perception, hearing and memory; the occipital lobe is related to vision; the parietal lobe is related to touch, temperature, pressure, and pain; the insula is related to the autonomic function of the brainstem, and also processes Taste information.
在对语言处理和大脑的研究过程中，最早在特定大脑区域和语言处理之间建立联系的人之一是法国外科医生—保罗·布罗卡。布罗卡通过对许多说话有缺陷的人进行了尸检而发现大多数人脑部受损(或病变)在左额叶上，现在称为Broca区。德国解剖学家、神经病理学家卡尔·韦尼克(Carl Wernicke)将大脑中的左后颞上回区域命名为韦尼克区(Wernicke District)，并提出大脑的不同区域专门用于不同的语言任务。二十世纪初的科尔比尼亚·布罗德曼(Korbinian Brodmann)根据每个区域的细胞结构和功能将大脑的表面划分为不同编号的区域，这些区域被称为布罗德曼(Brodmann)区。布罗德曼(Brodmann)区被广泛应用在神经语言领域中对大脑特定语言“模块”位置的研究，例如，Broca区处理语音的运动产生，而Wernicke区处理听觉语音理解。During his research on language processing and the brain, one of the first to establish a link between specific brain regions and language processing was French surgeon Paul Broca. By performing autopsies on many people with speech defects, Broca found that most of the brain damage (or lesions) was in the left frontal lobe, now called Broca's area. German anatomist and neuropathologist Carl Wernicke named the left posterior superior temporal gyrus region of the brain the Wernicke District and proposed that different areas of the brain are specialized for different language tasks. In the early twentieth century, Korbinian Brodmann divided the surface of the brain into differently numbered regions, called Brodmann, based on the cellular structure and function of each region. )Area. Brodmann's area is widely used in the neurolinguistic field to study the location of specific language "modules" in the brain, for example, Broca's area processes speech motor production, while Wernicke's area processes auditory speech comprehension.
在20世纪60年代以后连续多年进行裂脑人实验后，美国心理生物学家斯佩里提出左右脑的特化和分工：左脑负责记忆、时间、语言、判断、排列、分类、逻辑、分析、书写、推理、抑制的五感等，思维方式具有连续性、延续性和分析性等特点。因此，左脑被称作“意识脑”、“学术脑”、“语言脑”或“逻辑脑”。右脑主要负责空间形象记忆、直觉、情感、身体协调、视知觉、美术、音乐节奏、想象、灵感和顿悟等，思维方式具有无序性、跳跃性、直觉性等特点。所以，右脑又被称作“本能脑”、“潜意识脑”、“创造脑”、“音乐脑”或“艺 术脑”。After conducting split-brain experiments for many years after the 1960s, American psychobiologist Sperry proposed the specialization and division of labor between the left and right brains: the left brain is responsible for memory, time, language, judgment, arrangement, classification, logic, analysis , writing, reasoning, inhibition of the five senses, etc., the way of thinking has the characteristics of continuity, continuity and analysis. Therefore, the left brain is called the "conscious brain", "academic brain", "language brain" or "logical brain". The right brain is mainly responsible for spatial image memory, intuition, emotion, physical coordination, visual perception, art, music rhythm, imagination, inspiration and insight. Therefore, the right brain is also known as the "instinct brain", "subconscious brain", "creative brain", "music brain" or "artistic brain".
神经语言学的另外一个重要的工作是对心理语言学家和理论语言学家提出的理论进行测试和评估。一般而言，理论语言学家提出模型来解释语言的结构以及语言信息的组织方式；心理语言学家提出模型和算法来解释语言信息在头脑中的处理方式；神经语言学家分析大脑活动以推断生物结构(种群和个体)的方式，例如，神经元网络执行这些心理语言处理算法。举例而言，杰内特·福德(Janet Fodor)和莱恩·弗雷泽(Lyn Frazier)的“序列”模型和西奥·沃斯(Theo Vosse)和杰拉德·肯本(Gerard Kempen)的“统一模型”都是针对句子处理的不同模型。神经语言学利用句子处理实验、根据ERP技术获得的ELAN、N400和P600大脑反应来检验生理性大脑反应，再将生理性大脑反应的结果与心理语言学家提出的句子处理模型的预测结果进行比较就能够反映不同句子处理模型的合理性。另一方面，神经语言学还可以基于对大脑生理学的知识，通过“将神经结构的知识概括为语言结构”，指导心理语言学对语言的结构和组织提出新的理论。Another important work of neurolinguistics is the testing and evaluation of theories proposed by psycholinguists and theoretical linguists. In general, theoretical linguists propose models to explain the structure of language and how linguistic information is organized; psycholinguists propose models and algorithms to explain how linguistic information is processed in the mind; neurolinguists analyze brain activity to infer The way biological structures (populations and individuals), for example, networks of neurons, execute these psycholinguistic processing algorithms. For example, Janet Fodor and Lyn Frazier's "sequence" model and Theo Vosse and Gerard Kempen The "unified models" are all different models for sentence processing. Neurolinguistics examines physiological brain responses using sentence processing experiments, ELAN, N400, and P600 brain responses obtained from ERP techniques, and then compares the results of physiological brain responses with predictions from sentence processing models proposed by psycholinguists It can reflect the rationality of different sentence processing models. On the other hand, neurolinguistics can also guide psycholinguistics to propose new theories on the structure and organization of language based on knowledge of brain physiology by "generalizing knowledge of neural structure into language structure".
神经语言学领域中运用的研究方法及相关技术随着现代科技的进步也在不断发展。以下针对脑科学研究的技术手段进行详细的说明。The research methods and related technologies used in the field of neurolinguistics are constantly developing with the advancement of modern technology. The following is a detailed description of the technical means of brain research.
最初，人们采用发生学实验方法和病理学实验方法来研究实际语言过程、推测大脑的语言机制。以病理学实验方法为例，其通过从神经心理学角度分析脑损伤患者的语言状况，利用脑损伤区域的分析来了解语言生成过程及其神经心理机制。Initially, genetic and pathological experimental methods were used to study the actual language process and speculate on the language mechanism of the brain. Taking the pathological experimental method as an example, it analyzes the language status of patients with brain injury from the perspective of neuropsychology, and uses the analysis of brain injury regions to understand the process of language generation and its neuropsychological mechanism.
随着科学技术的进步，脑科学出现了新的研究方法，在这些方法主要基于获取的脑信号和脑信息。脑信号的获取包括用于获取头皮脑电信号的脑电图(Electroencephalogram，EEG)技术、获取皮层脑电信号的脑皮层电图(Electrocorticography，ECoC)技术、获取脑磁波信号的脑磁图(Magnetoencephalography，MEG)技术、事件相关电位(Event-related potential， ERP)技术等。脑电图技术采用无创伤EEG电极记录不同颅骨脑部位置的电位变化。脑皮层电图技术则更加深入，采用内置于大脑皮层中的ECoC电极采集深层皮层活动的电位变化。脑皮层电图技术虽然更加准确，却是一种侵入式的脑信号获取技术。脑磁图技术利用特别敏感的超冷电磁测定器测出颅脑内极微弱的脑磁波，以此获取大脑中的电场分布的变化。事件相关电位技术则属于一种诱发电位方法，其原理是人的感觉传入系统受到特殊刺激时，在中枢神经系统会诱发特殊电位，这种诱发的电位可以说明某种功能的神经机制。With the advancement of science and technology, new research methods have emerged in brain science, which are mainly based on acquired brain signals and brain information. The acquisition of brain signals includes electroencephalogram (EEG) technology for obtaining scalp EEG signals, electrocorticography (ECoC) technology for obtaining cortical EEG signals, and magnetoencephalography (Magnetoencephalography) for obtaining electroencephalogram signals. , MEG) technology, event-related potential (Event-related potential, ERP) technology and so on. Electroencephalography technology uses non-invasive EEG electrodes to record potential changes at different cranial brain locations. Electrocorticography technology is more in-depth, using ECoC electrodes built into the cerebral cortex to collect potential changes in deep cortical activity. Electrocorticography, although more accurate, is an invasive technique for acquiring brain signals. Magnetoencephalography technology uses a particularly sensitive ultra-cold electromagnetic detector to measure the extremely weak brain magnetic waves in the brain, so as to obtain the changes in the electric field distribution in the brain. Event-related potential technology belongs to a kind of evoked potential method.
脑信息的另一种获取方法是对人脑直接成像，以观察活人大脑处理语言信息的情况。所述的大脑成像技术包括计算机断层扫描(Computed Tomography,CT)，如X射线CT、超声CT、γ射线CT；正电子发射计算机断层扫描(Position Emission Computed Tomography，PET)；单光子发射计算机断层扫描(Single-Photo Emission Computed Tomography，SPECT)；磁共振成像(Magnetic Resonance Imaging,MRI)；功能磁共振成像(functional Magnetic Resonance Imaging,fMRI)；近红外光谱(Near Infrared Spectroscopy,NIRS)；功能近红外光谱(functional Near Infrared Spectroscopy,fNIRS)；脑血管造影；光声成像(Photoacoustic Imanging,PAI)；快速功能光声显微镜技术(fast-functional RAM)等等。Another way to obtain brain information is to directly image the human brain to observe how the brain processes language information in a living person. Described brain imaging technology includes computed tomography (Computed Tomography, CT), such as X-ray CT, ultrasound CT, γ-ray CT; Positron Emission Computed Tomography (Position Emission Computed Tomography, PET); Single Photon Emission Computed Tomography (Single-Photo Emission Computed Tomography, SPECT); Magnetic Resonance Imaging (MRI); Functional Magnetic Resonance Imaging (fMRI); Near Infrared Spectroscopy (NIRS); Functional Near Infrared Spectroscopy (functional Near Infrared Spectroscopy, fNIRS); cerebral angiography; photoacoustic imaging (Photoacoustic Imaging, PAI); fast functional photoacoustic microscopy (fast-functional RAM) and so on.
除了上述方法外，还有测试大脑半球言语功能的方法，例如把一侧脑半球麻醉，研究另一侧脑半球的言语功能；或者用速示仪器向人的半侧视野出示词语，研究言语视觉的脑功能；或者给人的两耳提供有声语言信息，研究大脑两半球言语听觉功能；或者切断大脑两半球之间的耕砥依，研究裂脑人两半球的言语功能等等。In addition to the above methods, there are also methods to test the speech function of the cerebral hemisphere, such as anesthetizing one hemisphere to study the speech function of the other cerebral hemisphere; or using a quick display instrument to show words to the human hemisphere to study speech and vision Or provide voice and language information to human ears to study the speech and auditory function of the two hemispheres of the brain;
由于脑信号及脑信息获取技术及分析技术的进步，神经语言学领域研究面及深度也大大扩大并且成绩斐然。以下对功能磁共振成像fMRI和功能近红外光谱fNIRS这两个非侵入性的重要的脑信息获取技术进行更为详细的介绍。Due to the advancement of brain signal and brain information acquisition technology and analysis technology, the scope and depth of research in the field of neurolinguistics has also been greatly expanded and achieved remarkable results. Two important non-invasive brain information acquisition techniques, fMRI and fNIRS, are described in more detail below.
fMRI属于MRI中的一种成像方法，其主要原理是利用磁振造影来测量神经元活动所引发的血液动力的改变情况，即通过检测血液进入脑细胞的磁场变化而实现脑工功能成像。如图1所示，为MRI设备的一个简要结构示意图。所述MRI设备主要包括磁铁系统、射频系统和计算机图像重建系统。磁铁系统主要用于产生两个磁场，一个为静磁场，又称主磁场；另一个为梯度场(gradient coils)。目前大部分设备采用超导磁铁来产生主磁场，磁场强度为0.2T-7.0T，常见的为1.5T和3.0T，并且另有匀磁线圈(shim coil)以协助主磁场达到高均匀度。采用梯度场线圈用来产生并控制磁场中的梯度以生成梯度场，实现NMR信号的空间编码。其包括三组线圈组，产生x、y、z三个方向的梯度场，线圈组的磁场叠加起来，可得到任意方向的梯度场。射频系统包括射频(RF)发生器和射频(RF)接收器。射频发生器用于产生短而强的射频场，以脉冲方式加到样品上，使样品中的氢核产生NMR现象。射频接收器用于接收NMR信号，放大后送入计算机图像重建系统。计算机图像重建系统对射频接收器送来的信号经A/D转换器，将模拟信号转换成数字信号，根据与观察层面各体素的对应关系，经计算机处理，得出层面图像数据，再经D/A转换器，加到图像显示器上，按NMR的大小，用不同的灰度等级显示出欲观察层面的图像。随着技术的进步，还出现了各种用于解决提高成像速度、提高空间分辨率的改进技术。例如，采用彼此独立的时间基函数和空间基函数的乘积和表示磁共振信号以得到具有高空间分辨率和高时间分辨率的功能磁共振成像图像；又例如，采用在一次激励核磁性后，顺序地高速反转梯度磁场，有序地收集多个回波信号的EPI(回波平面成像)下的扫描方法以解决图像位置偏移而产生的图像失真的问题，从而使得fMRI能够更好地应用于脑功能研究中。fMRI is an imaging method in MRI. Its main principle is to use magnetic resonance imaging to measure the changes in hemodynamics caused by neuronal activity, that is, to achieve brain functional imaging by detecting changes in the magnetic field of blood entering brain cells. As shown in FIG. 1 , it is a schematic structural diagram of an MRI apparatus. The MRI equipment mainly includes a magnet system, a radio frequency system and a computer image reconstruction system. The magnet system is mainly used to generate two magnetic fields, one is the static magnetic field, also known as the main magnetic field; the other is the gradient coils. At present, most devices use superconducting magnets to generate the main magnetic field, the magnetic field strength is 0.2T-7.0T, the common ones are 1.5T and 3.0T, and there is another shim coil to help the main magnetic field achieve high uniformity. The use of gradient field coils to generate and control gradients in the magnetic field to generate gradient fields enables spatial encoding of NMR signals. It includes three sets of coil groups, which generate gradient fields in three directions of x, y, and z. The magnetic fields of the coil groups are superimposed to obtain gradient fields in any direction. A radio frequency system includes a radio frequency (RF) generator and a radio frequency (RF) receiver. A radio frequency generator is used to generate a short and strong radio frequency field, which is applied to the sample in a pulsed manner to cause the hydrogen nuclei in the sample to produce NMR phenomena. The radio frequency receiver is used to receive the NMR signal, amplify it and send it to the computer image reconstruction system. The computer image reconstruction system converts the analog signal into a digital signal through the A/D converter for the signal sent by the radio frequency receiver. The D/A converter is added to the image display to display the image of the layer to be observed with different gray levels according to the size of the NMR. With the advancement of technology, various improved techniques have also appeared to solve the problem of increasing the imaging speed and increasing the spatial resolution. For example, the product sum of the time basis functions and the space basis functions that are independent of each other is used to represent the magnetic resonance signal to obtain a functional magnetic resonance imaging image with high spatial resolution and high temporal resolution; The scanning method under EPI (Echo Planar Imaging) that sequentially reverses the gradient magnetic field at high speed and collects multiple echo signals in an orderly manner can solve the problem of image distortion caused by image position shift, so that fMRI can better applied to the study of brain function.
fNIRS技术基于大脑神经活动会导致局部的血液动力学变化，利用脑组织中的氧合血红蛋白和脱氧血红蛋白对波长位于600-900nm区间的近红外光吸收率的差异特性，实时、直接地检测大脑皮层的血液动力学活动信息。通过血 液动力学活动信息的变化情况可以反推出大脑的神活动情况。所涉及的设备可包括fNIRS探测器、光源、探头和计算机系统。fNIRS探测器包括前额区探测器和全脑探测器。光源和探测器置于探头上，并与被测生物体，如大脑相接触，探头通过电线连接到计算机系统。计算机系统通过发出控制信号控制光源的亮与熄，探测器将其测量数据输入到计算机系统，经过信号的AD转换、处理等得到脑功能图像。随着硬件设备在制造上的完善、数据处理方法的改进，fNIRS技术为脑活动的监测提供了有力的监测手段。Based on the local hemodynamic changes caused by brain neural activity, fNIRS technology uses the difference in the absorption rate of oxyhemoglobin and deoxyhemoglobin in the brain tissue to the near-infrared light with wavelengths in the range of 600-900nm, and directly detects the cerebral cortex in real time. information on hemodynamic activity. The brain activity can be deduced from the changes in the hemodynamic activity information. The equipment involved may include fNIRS detectors, light sources, probes and computer systems. fNIRS detectors include forehead detectors and whole-brain detectors. The light source and detector are placed on the probe, which is in contact with the organism under test, such as the brain, and the probe is wired to a computer system. The computer system controls the on and off of the light source by sending out control signals, the detector inputs its measurement data to the computer system, and obtains brain function images through AD conversion and processing of the signals. With the improvement in the manufacture of hardware equipment and the improvement of data processing methods, fNIRS technology provides a powerful monitoring method for the monitoring of brain activity.
2016年4月27日，《自然》(Nature)杂志发表了第一作者为亚历山大·胡思(Alexander G.Huth)的的文章《自然语音揭示了平铺人类大脑皮层的语义图(Natural speech reveals the semantic maps that tile human cerebral cortex)》，其中公开了一项重要的研究成果。Huth等人使用大脑成像技术绘制了一幅大脑语义地图，从中可以清楚地看到大脑不同区域如何表征985个常见英语词汇及其含义。On April 27, 2016, "Nature" published the first author of the article "Natural speech reveals the semantic map of the human cerebral cortex. the semantic maps that tile human cerebral cortex), which published an important research result. Using brain imaging techniques, Huth et al. created a semantic map of the brain that clearly shows how different regions of the brain represent 985 common English words and their meanings.
图2是Huth等人在文章中公开的语素方面的模型示意图。Huth等人使用fMRI成像方法测量了七位受试者听了2小时自然语言陈述的故事过程中大脑的BOLD(Brain Blood-Oxygen Level Dependent，大脑血氧水平依赖性)反馈。每个词(Word)被投影到一个985个维度的基于词同时出现的统计结果建立的空间中。语素方面的模型反映了词的出现如何影响BOLD反馈。图3是Huth等人在文章中公开的语素方面的语义模型的主元分析示意图。主元分析反映了大脑中4个重要的语义维度，并经过RGB上色后得到色彩图。Huth等人的工作得到了一些重要的结论：首先，词语与大脑皮层中语义性选择区域的分布存在明显的关联。另一方面，在不同的个体之间，相同词语对应的语义性选择区域的分布也高度一致。Figure 2 is a schematic diagram of the model in terms of morphemes disclosed in the article by Huth et al. Huth et al. used fMRI imaging to measure brain BOLD (Brain Blood-Oxygen Level Dependent) feedback during seven subjects listening to a 2-hour story told in natural language. Each word (Word) is projected into a 985-dimensional space based on the statistics of word co-occurrences. The morphological aspect of the model reflects how word occurrence affects BOLD feedback. Figure 3 is a schematic diagram of principal component analysis of the semantic model in terms of morphemes disclosed in Huth et al. Principal component analysis reflects 4 important semantic dimensions in the brain, and the color map is obtained after RGB colorization. The work of Huth et al. led to some important conclusions: First, there is a clear correlation between words and the distribution of semantically selective regions in the cerebral cortex. On the other hand, among different individuals, the distribution of semantically selected regions corresponding to the same word is also highly consistent.
在上述基础研究成果上，人们又取得了进一步的研究成果，并且对语言有了更深的认识。《自然通信》(Nature Communications)在2020年4月20日发 表了题目为《通过映射语义关系的皮层表示来连接大脑中的概念(Connecting concepts in the brain by mapping cortical representations of semantic relations)》文章。通过与Huth等人类似的语素方面预测模型及脑成像技术，文章得出结论：语义类别和关系都由空间重叠的皮层模式而不是解剖上分离的区域表示，人脑不仅使用分布式网络对概念进行编码，而且还对概念之间的关系进行编码。Based on the above basic research results, people have achieved further research results and have a deeper understanding of language. Nature Communications published an article titled "Connecting concepts in the brain by mapping cortical representations of semantic relations" on April 20, 2020. Using morphological aspect prediction models and brain imaging techniques similar to those of Huth et al., we conclude that semantic categories and relationships are represented by spatially overlapping cortical patterns rather than anatomically separated regions, and the human brain uses not only distributed networks for conceptual encodes, but also encodes the relationships between concepts.
脑科学的上述进展将人们对大脑高级功能的理解推进到了一个新的阶段。如诺姆·乔姆斯基(Noam Cginsky)等在2017年9月18日发表在《自然》杂志上名称为《语言、思想和大脑》文章中所建议，不要将语言等同于“语音”或“交流”，而应将语言最好地描述为一种生物学确定的计算认知机制。如本申请的发明人的研究所证明的，利用这一确定的认知机制，可以使得语言成为工具，在大脑中产生特定的生理学反馈，从而提高人类的心智，促进思维的转变，推动文化的进步。The above-mentioned advances in brain science have pushed people's understanding of higher-level functions of the brain to a new stage. Do not equate language with "speech" or "communication", but language is best described as a biologically determined computational cognitive mechanism. As demonstrated by the research of the inventors of the present application, utilizing this identified cognitive mechanism, language can be used as a tool to generate specific physiological feedback in the brain, thereby enhancing the human mind, promoting the transformation of thinking, and promoting the development of culture. progress.
在本文以下的描述中，提供以下术语的定义以帮助理解本发明。In the following description herein, definitions of the following terms are provided to assist the understanding of the present invention.
如本文所使用的，“概念”所指为具有确定含义的思维活动的单元。例如，概念可以是中文的字(character)或者词(word)；也可以是英文的词(word)或者短语(phrase)。但是，不具备特定含义的汉字笔画、外文字母、日文片假名等无法成为思维活动的单元，也就不是本文所称的概念。如所发现的，本文的概念与语言存在着明显的关联。不同的语言所表示的概念可能具有相同的含义；然而，大多数情况下两者的含义并不相同。例如，中文的“桌子”与英文的“table”其实从概念的外延上就并不一致，因此，可以被认为是不同的概念。而外来语的例子中，如“计算机”和“computer”之间差别就非常小，可以认为是相同的概念。在一些实施例中，由于不同的语言之下的概念有着很大的区别，相同的语言的人之间更容易在大脑中形成相似的反馈。As used herein, a "concept" refers to a unit of thought activity that has a defined meaning. For example, a concept may be a Chinese character (character) or a word (word); it may also be an English word (word) or phrase (phrase). However, the strokes of Chinese characters, foreign letters, Japanese katakana, etc. that do not have specific meanings cannot become units of thinking activities, and they are not concepts called in this article. As found, the concepts of this paper are clearly linked to language. Concepts expressed in different languages may have the same meaning; however, in most cases the two are not the same. For example, the Chinese "table" and the English "table" are not consistent in the extension of the concept, so they can be regarded as different concepts. In the case of loanwords, the difference between "computer" and "computer" is very small and can be considered as the same concept. In some embodiments, it is easier for people of the same language to develop similar feedback in the brain due to the large differences in the concepts underlying different languages.
类似的概念在大脑中可能形成相似的反馈。根据已有的语言学的分类以及统计上相同概念同时出现的频率，可以得到一个概念的子集。在这个概念的子 集中包括多个有代表性的概念。这些代表性的概念能够形成一个语素方面的空间。其他概念可以表达为该子集的语素方面的空间中一个或多个代表性概念的投影。换言之，该子集能够表达其他未在该子集代表性概念中出现的概念。对于某一特定的语言而言，在该语言下的该子集能够表达该语言的概念的含义，从而将该语言简化为能够快速处理的代表性概念形成的子集的序列。Similar concepts may form similar feedback in the brain. A subset of concepts can be derived based on existing linguistic classifications and the frequency of co-occurrence of statistically identical concepts. Several representative concepts are included in this subset of concepts. These representative concepts can form a morphological space. Other concepts can be expressed as projections of one or more representative concepts in the space of the morpheme aspect of the subset. In other words, the subset can express other concepts that are not present in the subset's representative concepts. For a particular language, the subset under that language can express the meaning of the language's concepts, thereby reducing the language to a sequence of subsets of representative concepts that can be processed quickly.
进一步地，代表性概念的子集还能够根据所表达的经验而分类为不同的经验子集。本文中的“经验(experience)”是指由于经历而形成于大脑中的反馈，其可以是多次实践而形成的固定反馈；也可以是一次体验而形成的单次反馈。经验包括但不限于放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽等等。经过分类的代表性概念形成的经验子集能够更容易地在大脑中形成其代表的经验对应的反馈。Further, the subsets of representative concepts can also be classified into different subsets of experience based on the experience expressed. In this article, "experience" refers to the feedback formed in the brain due to experience, which can be a fixed feedback formed by multiple practices; it can also be a single feedback formed by a single experience. Experiences include, but are not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, beauty, and more. The subset of experience formed by the categorized representative concepts can more easily form feedback in the brain corresponding to the experience it represents.
如本文所使用的，大脑中的“反馈”是指大脑皮层中针对外界刺激所产生的反应。如所了解的，在外界刺激下，大脑皮层中的不同区域会处于活动状态，并进一步导致各种活性物质，例如氧、葡萄糖等，的代谢发生变化，从而被脑电信号检测手段所感测。大脑皮层的激活模式代表大脑皮层活动状态的区域在空间上的分布。不同的激活模式表明大脑皮层中不同的多个区域处于活动状态。即使人处于睡眠状态，大脑皮层中仍有多个区域处于活动状态，尽管相比于清醒状态处于活动状态的区域会减少。如果考察一段时间大脑皮层中处于活动状态的多个区域的变化，这样的变化其实表明了大脑皮层激活模式的变化。As used herein, "feedback" in the brain refers to responses in the cerebral cortex to external stimuli. As is known, under external stimuli, different regions in the cerebral cortex will be active, and further lead to changes in the metabolism of various active substances, such as oxygen, glucose, etc., which are sensed by EEG signal detection means. The activation patterns of the cerebral cortex represent the spatial distribution of regions of the cerebral cortex's active state. The different activation patterns indicate that different regions in the cerebral cortex are active. Even when a person is asleep, several areas of the cerebral cortex are active, albeit in fewer areas than in the awake state. If you look at changes in multiple regions of the cerebral cortex that are active over time, such changes actually indicate changes in the activation pattern of the cerebral cortex.
大脑皮层激活模式是快速时变的。不同激活模式的持续时间很短，例如几百毫秒到几秒钟。不同激活模式可能是响应于不同的外界刺激而产生的，也可能是响应于大脑高级功能而自发产生的。因此，大脑皮层不同的激活模式之间相互独立。反之，相互独立的激活模式之间也会对应于不同的外界刺激或者大脑高级功能的结果。Cortical activation patterns are rapidly time-varying. The durations of the different activation modes are short, such as hundreds of milliseconds to a few seconds. Different activation patterns may arise in response to different external stimuli, or they may arise spontaneously in response to higher brain functions. Therefore, the different activation patterns of the cerebral cortex are independent of each other. Conversely, independent activation patterns also correspond to different external stimuli or the results of higher brain functions.
进一步地，由于激活模式的持续时间是有限的，在一段时间，例如十几秒 钟内，大脑皮层会经历多个不同的激活模式。因此，该段时间内大脑皮层中处于活动状态的多个区域的变化可以按时间进行切片(例如100毫秒)而将每个时间切片内大脑皮层中处于活动状态的多个区域作为一个激活模式。相邻的时间切片内的激活模式可以是同一激活模式的延续，因此，相互是非独立的。或者，相邻的时间切片内的激活模式是不同的激活模式，而相互是独立的。因此，通过各个时间切片内的激活模式之间的独立性关系，能够得出该段时间内大脑皮层的激活模式的变化。Further, since the duration of the activation pattern is limited, the cerebral cortex will experience multiple different activation patterns within a period of time, such as a dozen or so seconds. Therefore, the changes in the active regions of the cerebral cortex during this period can be sliced in time (eg, 100 ms) and the active regions of the cerebral cortex in each time slice can be regarded as an activation pattern. Activation patterns in adjacent time slices can be continuations of the same activation pattern and, therefore, are not independent of each other. Alternatively, the activation patterns in adjacent time slices are different activation patterns and are independent of each other. Therefore, through the independent relationship between the activation patterns in each time slice, the changes in the activation patterns of the cerebral cortex during this period can be obtained.
如本文所使用的，“希望的反馈(desired feedback)”是指在大脑中形成对于目标反馈进行模拟获得的反馈。对于目标反馈而言，模拟反馈会尽可能地与目标反馈相一致，虽然二者不必完全一致。目标反馈与希望的反馈之间会存在多个相似的激活模式和/或激活模式在时间上的变化。激活模式在时间上的变化包括顺序、持续、间隔等。在一些实施例中，目标反馈代表了一种希望经验，包括但不限于放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽等等。在一些实施例中，通过对于目标反馈的模拟，模拟反馈将希望的经验从另一个大脑皮层中转移到受试者的大脑皮层中。As used herein, "desired feedback" refers to the formation of feedback in the brain that simulates the desired feedback. For target feedback, the analog feedback will match the target feedback as closely as possible, although the two need not be exactly the same. There may be multiple similar activation patterns and/or temporal changes in activation patterns between the target feedback and the desired feedback. Variations of activation patterns in time include sequence, duration, interval, and the like. In some embodiments, the goal feedback represents a hopeful experience including, but not limited to, relaxation, calm, confidence, joy, satisfaction, bravery, health, excitement, success, beauty, and the like. In some embodiments, the simulated feedback transfers the desired experience from another cerebral cortex to the subject's cerebral cortex by simulating the target feedback.
如本文所使用的，“以自然方式被大脑感知”是指通过人体本身的感知方式，包括：听觉、视觉、触觉、嗅觉和味觉，将概念从外界输入到大脑中。例如，听取一段文字内容，阅读一段文字内容，以盲文方式通过触觉感知一段文字的内容，闻到某种气味或者品尝到某种味道。如所理解的，“以自然方式被大脑感知”不包括通过针刺、电击、手术等侵入式或非侵入式的方式与大脑进行交流。对于脑机接口而言，如果脑机接口采用的是与人体本身的感知方式相同的方式与大脑进行交流，那么其也在本术语的范围中。另一方面，如果脑机接口采用的是非人体本身的感知方式，例如在大脑的某个区域通过电极施加电压，那么其将不在本发明的范围之中。可以理解的是，以自然的方式与大脑交流能够更加高效，更少的副作用，也更有利于被接受。As used herein, "perceived by the brain in a natural manner" refers to the input of concepts into the brain from the outside world through the human body's own perception methods, including: hearing, sight, touch, smell, and taste. For example, listening to a text, reading a text, tactilely sensing a text in Braille, smelling a certain smell or tasting a certain taste. As is understood, "perceived by the brain in a natural way" does not include communicating with the brain through invasive or non-invasive means such as acupuncture, electric shocks, surgery, etc. For brain-computer interfaces, if the brain-computer interface communicates with the brain in the same way as the human body itself perceives, it is also within the scope of this term. On the other hand, if the brain-computer interface uses non-human perception methods, such as applying voltage through electrodes in a certain area of the brain, then it is not within the scope of the present invention. Understandably, communicating with the brain in a natural way is more efficient, has fewer side effects, and is more conducive to acceptance.
如本文所使用的，“大脑键盘”是指一种输入装置，其包括多个按键。至少一个或多个按键对应一个或多个概念。在一些实施例中，多个按键与代表性概念子集的多个概念相对应。在另一些实施例中，多个按键与经验子集的多个概念相对应。这些按键可以是多个物理按键，也可以是输入界面上的多个虚拟按键。当大脑键盘的一个按键被“按下”，该按键所对应的概念被输入。多个按键被连续“按下”就会得到一个具有时间长度的序列。在该时间长度的多个时段内，该序列包括所述一个或多个由大脑键盘的按键输入的概念。因此，在一些实施例中，大脑键盘可以被认为能够应用本发明的方法的专用设备。在一些实施例中，大脑键盘还可以包括处理器、存储器、显示器、扬声器等其他组件，从而成为能够与使用者和受试者的大脑之间互动的装置。As used herein, "brain keyboard" refers to an input device that includes a plurality of keys. At least one or more keys correspond to one or more concepts. In some embodiments, multiple keys correspond to multiple concepts of the representative subset of concepts. In other embodiments, multiple keys correspond to multiple concepts of the experience subset. These keys can be multiple physical keys or multiple virtual keys on the input interface. When a key of the brain keyboard is "pressed", the concept corresponding to the key is entered. Multiple keys being "pressed" in succession result in a sequence with a length of time. The sequence includes the one or more concepts entered by the keys of the brain keyboard over multiple periods of the time length. Thus, in some embodiments, a brain keyboard may be considered a dedicated device capable of applying the methods of the present invention. In some embodiments, the brain keyboard may also include other components such as a processor, memory, display, speakers, etc., thereby becoming a device capable of interacting with the brains of users and subjects.
如本文所使用的，“心理性疾病”是指思维、情感和行为上偏离通常社会生活规范的疾病。心理性疾病非穷举性地包括：凹陷、大凹陷、治疗抗性抑郁症和治疗抗性双相抑郁症、双相性精神障碍、季节性情感障碍、情绪障碍、慢性抑郁症、精神病性抑郁症、产后抑郁症、经前期焦虑症(PMDD)、情境抑郁、非典型抑郁症、躁狂、焦虑症、注意力缺陷障碍(ADD)、具有多动性的注意力缺陷障碍(ADDH)和注意力缺陷/多动性障碍(AD/HD)、双相性和躁狂性病症、强迫症、食欲过盛、月经前期综合征、物质成瘾或滥用、尼古丁成瘾、心理-性功能障碍和假性球泡症中的一者或多者。As used herein, a "psychological disorder" refers to a disorder of thinking, emotion, and behavior that deviates from the usual norms of social life. Psychological disorders include, non-exhaustive, depressions, major depressions, treatment-resistant depression and treatment-resistant bipolar depression, bipolar disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression , postpartum depression, premenstrual dysphoric disorder (PMDD), situational depression, atypical depression, mania, anxiety, attention deficit disorder (ADD), attention deficit disorder with hyperactivity (ADDH), and attention Deficit/Hyperactivity Disorder (AD/HD), Bipolar and Manic Disorders, Obsessive-Compulsive Disorder, Hyperphagia, Premenstrual Syndrome, Substance Addiction or Abuse, Nicotine Addiction, Psycho-sexual Dysfunction, and Pseudosexuality one or more of bulbar disease.
如本文所使用的，“精神疾病”是指大脑功能失调导致的认知、情感、意志或行为出现障碍的疾病。精神疾病非穷举性地包括：精神分裂症、精神分裂症情感障碍、双相性精神障碍、强迫性精神障碍、帕金森氏精神病、相向违反性精神障碍、Charles Bonnet综合征、自闭症和Tourette氏病中的一种或多种。As used herein, "mental disorder" refers to a disorder in which cognitive, emotional, volitional, or behavioral disorders result from dysfunctional brain functions. Psychiatric disorders include, non-exhaustive, schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette one or more of the diseases.
如本文所使用的，“非疾病心理失常状态”是指并不具有病理学基础的认知、情感、意志或行为的出现失常的心理状态。非疾病心理失常状态非穷举地包括：缺乏信心、胆小、敏感、注意力涣散、意志薄弱、强迫性行为、考试恐 惧、演讲恐惧中的一种或多种。As used herein, a "non-disease psychotic state" refers to a cognitive, emotional, volitional, or behavioral disordered state of mind that does not have a pathological basis. Non-disease mental disorders include, but are not exhaustive, one or more of: lack of confidence, timidity, sensitivity, inattention, weakness of will, obsessive-compulsive behavior, fear of exams, fear of speaking.
本发明旨在通过将多个概念以自然的方式被大脑感知的手段模拟另一个大脑中的反馈从而实现将另一个大脑中的经验移植到该大脑中。在感知相同的概念后，在不同人类个体之间所形成的反馈的高度一致性成为自然方式进行经验移植的基础；而深度学习神经网络模型则使得这样的模拟成为现实。虽然模拟形成的反馈与目标反馈之间不会完全相同，但是类似的经验也同样能够在大脑中形成。这种低成本、无伤害而且易于接受的大脑中的经验移植无疑将是脑科学领域的重大进展。The present invention aims to transplant experiences from another brain into another brain by simulating feedback in the other brain by means of the means by which concepts are perceived by the brain in a natural way. The high degree of consistency of feedback formed between different human individuals after perceiving the same concept becomes the basis for experience transplantation in a natural way; deep learning neural network models make such simulations a reality. While the simulated and targeted feedback will not be exactly the same, similar experiences can also be formed in the brain. This low-cost, harmless, and easily accepted experience transplant into the brain will undoubtedly be a major advance in the field of brain science.
以下通过具体的实施例，详细说明本发明的技术方案。The technical solutions of the present invention will be described in detail below through specific embodiments.
图1是根据本发明的一个实施例的一种在大脑中产生反馈的方法流程。如图所示，在步骤110，确定希望的经验。在大脑中形成的希望的反馈是目标反馈的模拟。目标反馈是另一个大脑中大脑皮层中激活模式在时间上的变化，其代表了一种经验。这样的经验包括但不限于放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽。因此，确定希望的经验也可以被理解为确定被模拟的目标反馈。希望的反馈虽然不必和目标反馈完全一致，但是可以尽可能地接近目标反馈。这样，就能够在大脑中获得希望的经验。FIG. 1 is a flow chart of a method for generating feedback in the brain according to an embodiment of the present invention. As shown, at step 110, a desired experience is determined. The desired feedback formed in the brain is an analog of the target feedback. Goal feedback is a temporal change in activation patterns in the cerebral cortex of another brain that represents an experience. Such experiences include, but are not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, and beauty. Therefore, determining the desired experience can also be understood as determining the simulated target feedback. The desired feedback does not have to be exactly the same as the target feedback, but it can be as close to the target feedback as possible. In this way, the experience of hope can be obtained in the brain.
举例而言，对于一个非常劳累的人，单纯的休息(比如睡眠)并不能够让人感到放松。因为在休息的时候，大脑也仍然在高速运转，仍处于忙碌之中。从休息中回复后，人并不会感到放松，在某些时候会觉得更累。人的意志无法改变这种体验，有些人不得不求助于药物。针对有这一需求的人，可以将希望的反馈确定为放松或者深度放松的经验。For example, for a very tired person, simply rest (such as sleep) does not make people feel relaxed. Because when you are resting, your brain is still running at a high speed and is still busy. After recovering from a break, the person does not feel relaxed and at times feels more tired. The human will cannot change this experience, and some people have to resort to drugs. For those with this need, the desired feedback can be identified as a relaxing or deeply relaxing experience.
再一个例子，某些人存在严重的考试恐惧症。即使复习得很好，在考试之前也仍然会毫无信心，并且会严重影响考试成绩。这样的考试恐惧症也很难通过人的意志来改变。针对有这一需求的人，可以将希望的反馈确定为自信或者考试自信的经验。As another example, some people have severe test phobia. Even if you study well, you can still have a lack of confidence before the test and can seriously affect your test scores. Such test phobia is also difficult to change through human will. For those with this need, the desired feedback can be identified as the experience of confidence or exam confidence.
接下来，在步骤120，基于希望的经验确定目标反馈。如所理解的，目标反馈是具备某个经验的人的大脑中形成的反馈。在一些实施例中，通过收集具备希望经验的个体大脑中对于希望经验的反馈来获得目标反馈。例如，如果希望获得深度放松的经验，可以将经具备深度放松经验的个体选择为宗教的虔诚信徒，将对于希望经验的反馈选择为这些个体进行完冥想、祈祷等宗教活动后大脑中形成的反馈。通过脑信号探测技术记录一段时间大脑皮层中形成的反馈就可以作为目标反馈。当然，如所理解的，即使对于深度放松的经验，也可以选择不同的个体，或者选择个体不同活动后的大脑皮层中的反馈作为目标反馈。本发明在此并不做任何的限制。Next, at step 120, target feedback is determined based on the desired experience. As understood, target feedback is feedback formed in the brain of a person with a certain experience. In some embodiments, targeted feedback is obtained by collecting feedback on the desired experience in the brains of individuals with the desired experience. For example, if you want to gain the experience of deep relaxation, you can choose individuals with deep relaxation experience to be religious believers, and choose the feedback for the desired experience as the feedback formed in the brains of these individuals after they complete religious activities such as meditation and prayer. . By recording the feedback formed in the cerebral cortex for a period of time through brain signal detection technology, it can be used as the target feedback. Of course, as will be appreciated, even for deeply relaxing experiences, a different individual may be selected, or feedback in the cerebral cortex following different activities of an individual may be selected as the target feedback. The present invention does not make any limitation here.
再比如，如果希望获得自信的经验，可以将具备自信经验的个体选择为运动成绩优秀的运动员，将对于希望经验的反馈选择为这些个体在参加自己擅长的比赛项目之前一段时间内在大脑中形成的反馈。例如，播放比赛之前的录音或录像，营造出比赛之前的紧张气氛，使得这些个体产生即将参加比赛的感觉。当然，如所理解的，即使对于自信的经验，也可以选择不同的个体，或者选择个体不同活动后的大脑皮层中反馈作为目标反馈。本发明在此并不做任何的限制。For another example, if you want to gain self-confidence experience, you can choose individuals with self-confidence experience as athletes with excellent sports performance, and choose the feedback for the hoped experience as those formed in the brain of these individuals for a period of time before participating in the competitions they are good at. feedback. For example, playing a pre-match audio or video recording creates a pre-match tension that makes these individuals feel like they're about to compete. Of course, as will be appreciated, even for confident experiences, it is possible to select a different individual, or to select the feedback in the cerebral cortex after the individual's different activities as the target feedback. The present invention does not make any limitation here.
再比如，如果希望获得平静的经验，可以将具备平静经验的个体选择为经历过多次事件的参与者，将对于希望经验的反馈选择为这些个体在获知自己即将参与某个事件之前一段时间内在大脑中形成的反馈。例如，向飞机修理人员告知正在飞行的飞机某个部件发生故障，而对于飞机修理人员而言，这样的故障是经常发生、且不会产生严重后果。记录下飞机修理人员获知该故障后大脑中形成的反馈。当然，如所理解的，即使对于平静的经验，也可以选择不同的个体，或者选择个体不同活动后的大脑皮层中反馈作为目标反馈。本发明在此并不做任何的限制。For another example, if you want to gain a calming experience, you can select individuals with calming experience as participants who have experienced multiple events, and select feedback on the desired experience as those individuals who have learned that they are about to participate in an event within a certain period of time. Feedback formed in the brain. For example, an aircraft repairman is informed that a component of an in-flight aircraft has failed, whereas for the aircraft repairman, such failures are frequent and have no serious consequences. Record the feedback that formed in the brain of the aircraft repair crew when they learned of the malfunction. Of course, as will be appreciated, even for calm experiences, it is possible to select a different individual, or to select as the target feedback the feedback in the cerebral cortex following the individual's different activities. The present invention does not make any limitation here.
这样的例子还可以有很多。这样的例子其实都说明，为了获得希望经验的 目标反馈，可以通过选择具备希望经验的个体以及这些个体再现希望经验的活动。在这些活动之前、之中和之后记录下具备希望经验的个体中大脑皮层中的反馈作为目标反馈。在一些实施例中，可以根据被移植的经验的应用场景来选择再现希望经验的活动。二者之间的相似程度越高，被移植的经验所能发挥的效果也就越好。There could be many more such examples. Such examples actually illustrate that in order to obtain the target feedback of the desired experience, individuals with the desired experience can be selected and the activities of these individuals to reproduce the desired experience. Feedback in the cerebral cortex of individuals with the desired experience was recorded as target feedback before, during, and after these activities. In some embodiments, the activities that reproduce the desired experience may be selected according to the application context of the transplanted experience. The higher the similarity between the two, the better the effect of the transplanted experience.
在一些实施例中，利用功能磁共振仪对其大脑进行扫描获得的fMRI数据来记录目标反馈。然而，由于fMRI的设备限制，再现希望经验的活动的可选范围会很小，通常只能通过可穿戴的设备，如VR眼镜等，使得具备希望经验的个体产生参与再现希望经验的活动的感觉。MRI图像数据、CT图像数据、SEPECT图像数据和PAI图像数据与fMRI图像数据类似，都会对再现希望经验的活动产生很多限制。In some embodiments, target feedback is recorded using fMRI data obtained by scanning the brain with a functional magnetic resonance apparatus. However, due to the equipment limitations of fMRI, the optional range of activities to reproduce the desired experience will be very small, usually only through wearable devices, such as VR glasses, etc., so that individuals with the desired experience have the feeling of participating in the activities that reproduce the desired experience . MRI image data, CT image data, SEPECT image data, and PAI image data, similar to fMRI image data, impose many limitations on activities that reproduce desired experiences.
相比而言，通过前额区fNIRS探测器、全脑fNIRS探测器获得的fNIRS图像数据或者NIRS图像数据所采用的穿戴式探测装置则更适于更加复杂的活动。当然，通过侵入式或非侵入式的电极获得的脑电图或脑磁图数据等由于不需要将大脑置于大型设备中，对于复杂活动的限制也很小。In contrast, the fNIRS image data obtained by the frontal region fNIRS detector, the whole-brain fNIRS detector, or the wearable detection device used in the NIRS image data are more suitable for more complex activities. Of course, EEG or magnetoencephalography data, etc. obtained by invasive or non-invasive electrodes do not require the brain to be placed in a large device, so there are few restrictions on complex activities.
在本发明的以下部分中，以fMRI数据为例说明本发明的技术方案。如所理解的，其他脑信号(例如脑磁图)或脑信息探测技术所获得的数据也可以应用于本发明的方案中。本发明在此不做任何限制。In the following part of the present invention, fMRI data is taken as an example to illustrate the technical solution of the present invention. As will be appreciated, other brain signals (eg, magnetoencephalography) or data obtained from brain information detection techniques may also be employed in the protocols of the present invention. The present invention does not make any limitation here.
对于不同的个体而言，具备的经验不同而经历相同的活动再现的经验也会不同。因此，即使对于针对相同经验的反馈在个体之间也会有较大的差异。在一些实施例中，通过深度学习神经网络模型来获得不同个体之间针对相同经验的共同特征，从而减少个体之间的差异所带来反映在目标反馈中的噪声。Different individuals have different experiences and experience the reproduction of the same activity differently. Thus, even feedback for the same experience can vary widely between individuals. In some embodiments, common features for the same experience among different individuals are obtained through a deep learning neural network model, thereby reducing noise reflected in target feedback caused by differences between individuals.
由于fMRI能够获得三维图像，可以选择3D卷积神经网络模型(CNN)学习希望经验的大脑皮层的反馈。为了减小计算量，也可以选择使用二维图像的CNN。将具备希望经验的个体再现希望经验活动后记录的fMRI数据作为训 练样本。由于个体之间的语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业对于希望经验的影响很大，将这些与人有关的参数作为训练参数。在一些实施例中，使用卷积稀疏编码(CSC)来进行神经网络模型的构建。卷积稀疏编码是一个线性卷积的无监督的学习的方法，模型更简单，更直观，也容易分析理解。Since fMRI can obtain three-dimensional images, a 3D convolutional neural network model (CNN) can be chosen to learn feedback from the cerebral cortex that wishes to experience. In order to reduce the amount of computation, a CNN that uses two-dimensional images can also be selected. The fMRI data recorded after the individuals with the desired experience reproduced the desired experience activity were used as training samples. Since language, age, gender, religious belief, education level, occupation or previous occupation among individuals have a great influence on the desired experience, these human-related parameters are used as training parameters. In some embodiments, the construction of the neural network model is performed using convolutional sparse coding (CSC). Convolutional sparse coding is an unsupervised learning method of linear convolution. The model is simpler, more intuitive, and easy to analyze and understand.
在一些实施例中，对于希望获得某个经验的个体，将该个体的语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业作为参数输入经训练的神经网络模型。神经网络模型根据训练的结果，输出与该个体的该经验最为匹配的目标反馈。这样，既能够解决训练样本较少，无法获得构建准确的神经网络模型的问题，也能够减少训练样本之间个体差异对目标反馈带来的影响。In some embodiments, for an individual wishing to gain a certain experience, the individual's language, age, gender, religion, education, occupation or previous occupation are input as parameters to the trained neural network model. According to the training result, the neural network model outputs the target feedback that best matches the individual's experience. In this way, it can not only solve the problem that there are fewer training samples and cannot obtain an accurate neural network model, but also reduce the impact of individual differences between training samples on target feedback.
如所理解的，应用于本发明的神经网络模型并不限于CNN，也可以使用其他的特征分类模型；也不限于fMRI所获得的三维数据作为训练样本。其他的脑信号或脑信息数据也可以应用于此。例如，以CSC建立的模型处理脑磁图(MEG)的数据也能够取得非常不错的效果。As understood, the neural network model applied to the present invention is not limited to CNN, and other feature classification models can also be used; nor is it limited to three-dimensional data obtained by fMRI as training samples. Other brain signal or brain information data can also be applied here. For example, magnetoencephalography (MEG) data can also be processed very well with the model established by CSC.
在一些实施例中，建立针对各个经验的目标反馈的数据库。数据库中的多个目标反馈按所针对的不同经验进行分类。进一步地，语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业等与人有关的参数，也可以成为目标反馈进一步分类的依据。如果进一步考虑语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业等与人有关的参数，能够获得的目标反馈的匹配度更高。进一步地，如上所构建的神经网络模型能够用于对数据库中存储的目标反馈进行更新。由此，基于希望的反馈使用数据库就能够获得与之匹配的目标反馈，而不需每次都使用经训练的神经网络模型生成最新的目标反馈。虽然目标反馈数据库的准确度不如经训练的神经网络模型，但是使用更为方便，更加快速，也能够省去之前的步骤。In some embodiments, a database of target feedback for each experience is established. The multiple target feedback in the database is categorized by the different experiences targeted. Further, parameters related to people, such as language, age, gender, religious belief, education level, occupation or previous occupation, can also be the basis for further classification of target feedback. If parameters related to people such as language, age, gender, religious belief, education level, occupation or previous occupation are further considered, the matching degree of target feedback that can be obtained is higher. Further, the neural network model constructed as above can be used to update the target feedback stored in the database. Thus, using the database based on the desired feedback, it is possible to obtain matching target feedback without using the trained neural network model to generate the latest target feedback each time. Although the accuracy of the target feedback database is not as good as that of the trained neural network model, it is more convenient and faster to use, and it can also save the previous steps.
在步骤130，对目标反馈按时间划分为多个激活模式。大脑皮层中的目标 反馈包括大脑皮层中激活模式在时间上的变化。对于每个时刻而言，激活模式包括大脑皮层中激活部分的空间分布。因此，目标反馈可以被认为是大脑皮层中多个激活模式在时间上变化的序列。At step 130, the target feedback is divided into a plurality of activation modes by time. Targeted feedback in the cerebral cortex consists of temporal changes in activation patterns in the cerebral cortex. For each moment, the activation pattern includes the spatial distribution of the activated parts of the cerebral cortex. Thus, target feedback can be thought of as a temporally changing sequence of multiple activation patterns in the cerebral cortex.
在一些实施例中，如图5所示，对目标反馈按时间划分为多个激活模式的步骤包括：In some embodiments, as shown in FIG. 5 , the step of dividing the target feedback into multiple activation modes by time includes:
步骤S1301，以将目标反馈按时间进行切片。每个时间切片的长度为10-50ms。如图6所示的示意图，每个时间切片的长度为20ms，并为其编号为A00001、A00002等，以便于后续的处理。虽然更小的时间切片能够提高模拟的精度，然而时间长度过小时，不但会导致大量的待处理数据，带来巨大的计算量，而且，过小的时间切片也可能与自然方式的感知不匹配，导致出现大量的干扰数据。经过本步骤得到一个时间切片集合，其中的每个时间切片具有唯一的编号。Step S1301, to slice the target feedback by time. The length of each time slice is 10-50ms. As shown in the schematic diagram in FIG. 6 , the length of each time slice is 20ms, and it is numbered as A00001, A00002, etc. to facilitate subsequent processing. Although a smaller time slice can improve the accuracy of the simulation, if the time length is too small, it will not only result in a large amount of data to be processed and a huge amount of computation, but also a too small time slice may not match the perception in the natural way. , resulting in a large amount of interference data. After this step, a set of time slices is obtained, where each time slice has a unique number.
步骤S1302，确定每个时间切片中大脑皮层的激活模式，即激活部分的空间分布及信号强度。在一些实施例中，获取该时间切片中有代表性的激活模式作为该时间切片的激活模式。有多种方式能够获得代表性的激活模式，例如选取该时间切片中间时刻激活部分的空间分布作为代表性的激活模式；或者，计算该时间切片中各个时刻激活部分叠加后的平均空间分布作为代表性的激活模式。当然，其他的方式也可以应用于此。Step S1302: Determine the activation pattern of the cerebral cortex in each time slice, that is, the spatial distribution and signal intensity of the activated part. In some embodiments, a representative activation mode in the time slice is acquired as the activation mode of the time slice. There are many ways to obtain a representative activation pattern, for example, selecting the spatial distribution of the activation part at the middle of the time slice as the representative activation mode; or, calculating the average spatial distribution of the superposition of the activation parts at each time in the time slice as a representative Sexual activation mode. Of course, other methods can also be applied to this.
步骤S1303，计算相邻时间切片中大脑皮层的激活模式是否相互独立。如果两个相邻的两个时间切片中大脑皮层的激活模式是非独立的，那么说明该激活模式持续了至少两个时间切片的时间，那么在步骤S1304将这两个时间切片合并，如果两个时间切片中大脑皮层的激活模式是相互独立的，则在步骤S1305将这两个激活模式记录为不同的激活模式。Step S1303: Calculate whether the activation patterns of the cerebral cortex in adjacent time slices are independent of each other. If the activation patterns of the cerebral cortex in two adjacent time slices are non-independent, it means that the activation pattern lasts for at least two time slices, then the two time slices are merged in step S1304, if the two time slices are The activation modes of the cerebral cortex in the time slice are independent of each other, and the two activation modes are recorded as different activation modes in step S1305.
步骤S1306，判断是否处理完所有相邻的时间切片后，如果没有，则在重复前述步骤，直到相邻的时间切片都是独立的。如果已经处理完所有相邻的时 间切片，则此时已将目标反馈被划分为延续整数个时间切片的多个激活模式。如图7所示，每一个时间切片，如B0001、B0002，其为具有一定时间长度，其激活模式与其他时间切片的激活模式各不相同，即此时得到了另一个时间切片集合，其中的每一个时间切片包括了一种激活模式，并具有一个对应的时间段，集合中的全部时间切片形成一个在时间上连续性的序列。Step S1306, after judging whether all adjacent time slices have been processed, if not, repeat the foregoing steps until the adjacent time slices are independent. If all adjacent time slices have been processed, then the target feedback has been divided into multiple activation modes that last an integer number of time slices at this point. As shown in Figure 7, each time slice, such as B0001 and B0002, has a certain time length, and its activation mode is different from that of other time slices, that is, another time slice set is obtained at this time, in which the Each time slice includes an activation pattern and has a corresponding time period, and all time slices in the set form a temporally continuous sequence.
在一些实施例中，在计算相邻时间切片中大脑皮层的激活模式是否相互独立时，计算相邻时间切片中大脑皮层的激活模式的差异。如果差异超过预定的范围，则认为两个时间切片中的激活模式是相互独立的。在另一些实施例中，计算相邻时间切片中大脑皮层的激活模式的相关系数。如果相关系数超过预定阈值，则认为两个时间切片中的激活模式是相互独立的。当然，其他的方式也可以应用于此。In some embodiments, when calculating whether the activation patterns of the cerebral cortex in adjacent time slices are independent of each other, the difference in the activation pattern of the cerebral cortex in adjacent time slices is calculated. If the difference exceeds a predetermined range, the activation patterns in the two time slices are considered to be independent of each other. In other embodiments, correlation coefficients are calculated for the activation patterns of the cerebral cortex in adjacent time slices. If the correlation coefficient exceeds a predetermined threshold, the activation patterns in the two time slices are considered to be independent of each other. Of course, other methods can also be applied to this.
以fMRI数据处理为例，将目标反馈的fMRI数据按一定间隔时间(如10-50ms)进行时间切片后，能够得到多个时刻大脑皮质的激活区域及其信号强度，其中信号强度反映为激活区域的颜色深浅。根据多个时刻激活模式的独立性分析能够确定多个激活模式随时间的变化的情况。由此，目标反馈被划分出多个时间段，每一个时间段对应为一个激活模式。Taking fMRI data processing as an example, after the fMRI data fed back by the target is time-sliced at a certain interval (such as 10-50ms), the activation areas of the cerebral cortex and their signal intensities at multiple times can be obtained, and the signal intensity is reflected as the activation area. shades of color. According to the independent analysis of the activation modes at multiple times, the change of multiple activation modes over time can be determined. Thus, the target feedback is divided into multiple time periods, and each time period corresponds to an activation mode.
在接下来，在步骤140，识别目标反馈的多个激活模式中各个激活模式对应的一个或多个概念。Next, in step 140, one or more concepts corresponding to each activation mode among the plurality of activation modes of the target feedback are identified.
根据大脑皮层对自然方式感知的概念的响应，不同个体之间大脑皮层对同一个概念的响应激活区域及其分布和信号强度具有高度一致性；而对不同的概念，响应激活区域的分布不同，信号强度也不尽相同。According to the response of the cerebral cortex to the concept perceived in a natural way, the cerebral cortex of the cerebral cortex has a high degree of consistency in its distribution and signal intensity in response to the same concept. Signal strength also varies.
在一些实施例，获得包括多个概念的群组中各个概念对应的大脑皮层中激活区域的分布及其信号强度。将一个概念对应的大脑皮层中激活区域的分布及其信号强度定义为该概念对应的“概念模式”。在一些实施例中，建立概念模式数据库，存储多个概念的概念模式。以fMRI数据为例，通过记录受试者倾 听自然语言表述的一段文字内容时大脑皮层中形成的反馈的fMRI数据。经过数据处理，将fMRI数据与该段文字中的概念相对应，就能够获得与概念对应的fMRI数据反映的概念模式。将这些概念和对应的概念模式存储，以此能够建立起概念模式数据库。本领域技术人员应当理解，其他类型的脑信号或者脑探测技术的数据也能够被用来建立概念模式数据库。通过其他的方式，也能够将概念与其对应的概念模式相对应从而获得概念模式数据库。本发明在此并不做任何限制。In some embodiments, the distribution of activation regions in the cerebral cortex corresponding to each concept in a group comprising a plurality of concepts and their signal intensities are obtained. The distribution of activation areas in the cerebral cortex corresponding to a concept and its signal strength are defined as the "concept pattern" corresponding to the concept. In some embodiments, a conceptual schema database is established to store conceptual schemas for a plurality of concepts. Take fMRI data as an example, by recording the fMRI data of the feedback formed in the cerebral cortex when a subject listens to a piece of text expressed in natural language. After data processing, the fMRI data is corresponding to the concept in the text, and the concept pattern reflected by the fMRI data corresponding to the concept can be obtained. By storing these concepts and corresponding conceptual schemas, a conceptual schema database can be established. It will be understood by those skilled in the art that other types of brain signals or data from brain detection techniques can also be used to build the conceptual pattern database. In other manners, concepts can also be associated with their corresponding conceptual schemas to obtain a conceptual schema database. The present invention does not make any limitation here.
对应的，从一个激活模式包括在该时间切片的时间长度内的激活区域及其信号强度，如图8所示，可以根据激活区域及其分布和信号强度可以确定与其对应的概念模式，如概念模式1。再将该概念模式1与概念模式数据库中存储多个概念模式进行比较，找出一个概念或多个概念的组合。这些概念组合对应的激活模式为该概念组合中一个或多个概念模式的叠加，并且概念组合的激活模式与目标反馈的激活模式接近。由此，将目标反馈中的一个激活模式与一个或多个概念的组合相对应，或者说，目标反馈中的一个激活模式被分解为一个或多个概念的组合中的一个或多个概念的概念模式。Correspondingly, from an activation pattern including the activation area and its signal strength within the time length of the time slice, as shown in Figure 8, the corresponding conceptual pattern can be determined according to the activation area and its distribution and signal strength. Mode 1. Then, compare the conceptual schema 1 with multiple conceptual schemas stored in the conceptual schema database, and find out one concept or a combination of multiple concepts. The activation mode corresponding to these concept combinations is the superposition of one or more concept modes in the concept combination, and the activation mode of the concept combination is close to the activation mode of the target feedback. Thereby, an activation mode in the target feedback corresponds to a combination of one or more concepts, or, an activation mode in the target feedback is decomposed into one or more concepts in the combination of one or more concepts. Conceptual pattern.
在一些实施例中，利用深度学习神经网络模型获得目标反馈中的一个激活模式的概念组合。例如，可以选择卷积神经网络模型(CNN)识别概念组合。其他的用于模式识别的神经网络模型，例如深度信念网络DBN模型、递归神经网络RNN模型等也可以应用于此。In some embodiments, a deep learning neural network model is used to obtain a conceptual combination of activation patterns in target feedback. For example, a convolutional neural network model (CNN) can be chosen to identify concept combinations. Other neural network models for pattern recognition, such as deep belief network DBN model, recurrent neural network RNN model, etc., can also be applied here.
在一些实施例中，以fMRI数据为例，向受试者读出两个或多个不同的词(对应不同概念)，记录受试者大脑皮层中反馈的fMRI数据。将不同的词以及对应的fMRI数据形成的数据集作为训练集训练神经网络模型。经训练的神经网络模式以fMRI数据为输入，输出多个不同的概念组合，并且该多个不同概念组合按与作为输入的fMRI数据的匹配度从高到低进行排序。利用经训练的神经网络模型能够获得目标反馈中的一个激活模式对应的多个概念组合。在一 些实施例中，通常仅使用匹配度最高的概念组合。当几个概念组合的匹配度相差不大时，将这几个概念组合都保留，以利于后面的选择性使用。In some embodiments, taking fMRI data as an example, two or more different words (corresponding to different concepts) are read out to the subject, and the fMRI data fed back in the cerebral cortex of the subject is recorded. The dataset formed by different words and corresponding fMRI data is used as the training set to train the neural network model. The trained neural network model takes fMRI data as input and outputs a plurality of different concept combinations, and the plurality of different concept combinations are ordered from high to low matching with the fMRI data as input. Multiple concept combinations corresponding to one activation pattern in the target feedback can be obtained by using the trained neural network model. In some embodiments, typically only the best matching concept combination is used. When the matching degrees of several concept combinations are not much different, all these concept combinations are reserved to facilitate subsequent selective use.
在步骤150，基于目标反馈，确定具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括一个或多个概念。如前所述，目标反馈被划分为多个激活模式，每个激活模式延续整数个时间切片的时间长度。进一步地，每个激活模式可以对应于一个或多个概念组合。概念组合的延续时间与激活模式的延续时间相同，也是整数个时间切片的时间长度。由此，目标反馈能够被分解为多个时间长度的概念组合序列。进一步地，在该多个时间长度的概念组合序列中选择一个作为与目标反馈对应的概念组合序列。如图9所示，为根据本发明的一个实施例的多个概念的时序示意图，图中示出了与连续三个激活模式对应的概念。在本实施例中，激活模式1包括三个概念：概念1-概念3；激活模式2包括一个概念：概念4；激活模式3包括一个概念：概念5。这五个概念对应的起始时刻分别为0、T1、T2、T3、T4、T5。激活模式1对应的时长为t1＝T3-0；激活模式2对应的时长为t2＝T4-T3；激活模式3对应的时长为t3＝T5-T4。因而，每一个概念既包括起始时间，也包括持续时间段。At step 150, based on the target feedback, a sequence having a length of time is determined, wherein the sequence includes one or more concepts over multiple time periods of the length of time. As mentioned earlier, the target feedback is divided into a plurality of activation modes, each activation mode lasting an integer number of time slices of time length. Further, each activation mode may correspond to one or more concept combinations. The duration of the concept combination is the same as the duration of the active mode, which is also the duration of an integer number of time slices. Thus, target feedback can be decomposed into conceptual composition sequences of multiple temporal lengths. Further, one of the concept combination sequences of multiple time lengths is selected as the concept combination sequence corresponding to the target feedback. As shown in FIG. 9 , which is a schematic time sequence diagram of a plurality of concepts according to an embodiment of the present invention, the figure shows concepts corresponding to three consecutive activation modes. In this embodiment, activation mode 1 includes three concepts: concept 1-concept 3; activation mode 2 includes one concept: concept 4; activation mode 3 includes one concept: concept 5. The starting times corresponding to these five concepts are 0, T1, T2, T3, T4, and T5, respectively. The duration corresponding to the activation mode 1 is t1=T3-0; the duration corresponding to the activation mode 2 is t2=T4-T3; the duration corresponding to the activation mode 3 is t3=T5-T4. Thus, each concept includes both a start time and a duration period.
在一些实施例中，选择概念组合序列的标准可以多种多样。例如，可以根据希望的经验与概念组合中的对应关系进行选择，将在某种经验中不常出现的概念剔除。例如，希望的经验是“平静”，那么，包括例如“凶猛”、“激烈”等概念的概念组合就尽量不去选择。再例如，可以根据概念组合之间的关联关系进行选择，将关联性比较差，或者难以前后关联的概念组合剔除。例如，在序列中前后的概念组合都是与“飞翔”有关，那么中间的概念组合就也选择与“飞翔”有关的概念组合。再例如，可以根据概念组合整体的主题进行选择，将关联性较差的概念组合剔除。例如，在序列中概念组合都是与“大海”有关，那么当有大海有关的概念组合可以选择时，就可以剔除其他的概念组合。在一些实施例中，如果存在多个较好的概念组合难以取舍，也可以都保留以用于后 续步骤。In some embodiments, the criteria for selecting a concept combination sequence may vary. For example, the selection can be made according to the corresponding relationship between the desired experience and the concept combination, and the concept that does not appear frequently in a certain experience can be eliminated. For example, if the desired experience is "calm", then concept combinations including concepts such as "fierce" and "intense" are not selected as much as possible. For another example, the selection can be made according to the association relationship between the concept combinations, and the concept combinations that are relatively poorly correlated or difficult to be correlated with each other can be eliminated. For example, the concept combination before and after in the sequence is all related to "flying", then the concept combination in the middle also selects the concept combination related to "flying". For another example, the selection can be made according to the theme of the whole concept combination, and the concept combination with poor correlation can be eliminated. For example, in the sequence, the concept combinations are all related to the "sea", then when there are concept combinations related to the sea to choose from, other concept combinations can be eliminated. In some embodiments, if there are multiple better concept combinations that are difficult to choose, they may all be reserved for subsequent steps.
在步骤160，根据概念组合在所述时间段内出现的时刻和持续的时间形成一段文字。At step 160, a piece of text is formed according to the concept combination of the time instant and the duration of time within the time period.
如上所述，在一个时间长度内在不同的时间段内的概念组合形成了一个序列。该序列包括多个具有时间属性的概念，所述的时间属性之一为该概念的持续时间段，不同的概念，其在该序列中的持续时间段不同。为了在大脑中得到某种特定的反馈，概念在序列中的持续时间段并不完全取决于音节的多少，而是取决于所述概念与在大脑中希望得到的反馈的关联关系。因此，概念组合形成的一段文字中，每个概念都有确定的起始时间和持续时间。这也被称为概念在序列中的时间属性。As mentioned above, concepts combined in different time periods within one time length form a sequence. The sequence includes a plurality of concepts with temporal attributes, and one of the temporal attributes is the duration of the concept. Different concepts have different durations in the sequence. In order to get a specific feedback in the brain, the duration of a concept in a sequence is not entirely determined by the number of syllables, but by the association of the concept with the desired feedback in the brain. Therefore, in a piece of text formed by a combination of concepts, each concept has a definite start time and duration. This is also known as the temporal property of a concept in a sequence.
例如，为了唤起比赛选手的“自信”的经验，概念组合中出现了“掌声”与“起跳”这两个概念。根据激活模式确定的时间切片，二者的持续时间段可以不同。例如，“掌声”的持续时间段大于“起跳”的持续时间段。在一个更具体的例子中，在“优雅”和“落地”这两个概念之间，“优雅”的起始时刻为序列播放开始的第2秒，延续时间0.5秒；而“落地”的起始时刻为序列播放开始的第2.5秒，延续时间0.2秒。For example, in order to evoke the "confidence" experience of the contestants, the concepts of "applause" and "take off" appeared in the concept combination. The duration of the two can be different according to the time slice determined by the activation mode. For example, the duration of "applause" is greater than the duration of "take off". In a more specific example, between the two concepts of "elegance" and "landing", the starting moment of "elegance" is the 2nd second after the sequence playback starts, and the duration is 0.5 seconds; The start time is 2.5 seconds from the beginning of the sequence playback, and the duration is 0.2 seconds.
在步骤170，在该段文字中增加文字以形成有连贯语义的内容。由于是以自然的方式感知，那么更利于大脑接受的方式取得的效果也会更好，因此，有必要形成连贯语义，在一些实施例中，形成确定的主题，是更为有利的技术方案。At step 170, words are added to the paragraph to form content with coherent semantics. Since it is perceived in a natural way, a way that is more conducive to the brain's acceptance will also achieve better results. Therefore, it is necessary to form coherent semantics. In some embodiments, forming a certain theme is a more favorable technical solution.
在一些实施例中，在概念组合中增加助词等以连贯语义。例如，将概念对应于一个或多个短语(或词组)。例如，将“优雅”和“跳起”组合为“优雅地跳起”；将“飞舞”和“精灵”组合成“飞舞的精灵”等短语。如果概念组合中是两个同样性质的词，也可以直接并列。例如，将“勇敢”和“坚强”直接组合为“坚强勇敢”。如果概念组合两个词相差很远，则可以仅根据词的属 性进行组合。虽然难以理解，但语义仍是连贯的。例如，将“红色”和“坐下”直接按形容词在前实体词在后组合为“红色的坐下”；将“庙宇”和“飞翔”直接按名词在后组合为“飞翔的庙宇”。In some embodiments, auxiliary words and the like are added to the concept combination to coherent the semantics. For example, a concept corresponds to one or more phrases (or groups of words). For example, combine "graceful" and "jumping" into phrases like "jumping gracefully"; "flying" and "elf" into phrases such as "flying elves." If there are two words of the same nature in the concept combination, they can also be directly juxtaposed. For example, "brave" and "strong" are directly combined as "strong and brave." If the conceptual combination of two words is far apart, the combination can be based only on the properties of the words. Although difficult to understand, the semantics are still coherent. For example, "red" and "sit down" are combined directly with adjectives followed by the former entity word as "red sit down"; "temple" and "fly" are combined directly with nouns as "flying temple".
在一些实施例中，在前后概念组合中增加副词、介词等以连贯语义。前后概念组合之间可能相差较大，因此需要增加副词、介词等以将语音连贯。例如，前面的概念组合是“掌声开始”，后面的概念组合是“优雅地落地”，那么修改后的前后连贯语义可以是“随着掌声开始，优雅地落地”。In some embodiments, adverbs, prepositions, etc. are added to the contextual combination to coherent semantics. There may be a big difference between the before and after concept combinations, so adverbs, prepositions, etc. need to be added to make the pronunciation coherent. For example, the previous concept combination is "applause begins", and the latter concept combination is "elegantly landed", then the modified semantics of contextual coherence can be "beginning with applause, gracefully landed".
在一些实施例中，在前后概念组合中增加少量的实体词以连贯语义。前后概念组合之间可能相差较大，因此需要增加实体词等以将语音连贯。例如，将“红色的坐下”、“飞舞的精灵”和“飞翔的庙宇”进行组合时，可以做出下修改“ 在红色 中坐下”， 看见“飞舞的精灵”和“飞翔的庙宇”；其中，“在”和“中”是增加的介词，而“看见”是增加的实体词。在一些实施例中，增加的助词、介词、副词和实体词应当在序列中持续时间尽可能短，以尽量减少对原有序列定义的反馈的影响。而且，增加的实体词应当尽可能是常见的、习以为常的动作或事物。 In some embodiments, a small number of entity words are added to the contextual concept combination to coherent the semantics. There may be a large difference between the before and after concept combinations, so it is necessary to add entity words to make the speech coherent. For example, when combining "red sit down", "flying spirit" and "flying temple", you can make the following modification "sit down in red" , see "flying spirit" and "flying temple"; where "in" and "in" are added prepositions, and "see" is an added entity word. In some embodiments, the added particles, prepositions, adverbs, and entity words should be as short as possible in the sequence to minimize the impact of feedback on the original sequence definition. Moreover, the added entity words should be as common and habitual actions or things as possible.
在一些实施例中，可以人工校对的方式对于文字进行修改。特别地，可以将文字修改为尽量接近与希望的经验有关的主题。In some embodiments, the text may be modified by manual proofreading. In particular, the text may be modified to be as close as possible to the subject matter related to the desired experience.
在步骤180，根据有连贯语义的内容，调整一个或多个概念在所述时间段内出现的时刻和/或持续的时间。在一些情况下，如果完全按照目标反馈定义的多个概念在所述时间段内出现的时刻和/或持续的时间，一段文字可能听起来节奏比较奇怪，影响大脑中的反馈效果。因此，在一些实施例中，需要调整一个或多个概念在所述时间段内出现的时刻和/或持续的时间，使得整段文字听起来更为自然。然而，调整不能超过预设的范围，否则可能会改变大脑皮层中的反馈，无法达到复制经验的效果。一般而言，起始时刻的调整不超过20ms，而持续时间的调整不超过50ms。这样既有利于调整文字的阅读节奏，也不会 过于偏离目标反馈。At step 180, the timing and/or duration of the occurrence of one or more concepts within the time period is adjusted according to the coherent semantic content. In some cases, a piece of text may sound oddly rhythmic, affecting the feedback effect in the brain, if the timing and/or duration of the multiple concepts defined by the target feedback exactly occurs within the time period. Therefore, in some embodiments, the timing and/or duration of one or more concepts in the time period may need to be adjusted so that the entire text sounds more natural. However, the adjustment should not exceed the preset range, otherwise the feedback in the cerebral cortex may be changed, and the effect of replicating the experience will not be achieved. Generally speaking, the adjustment of the start time does not exceed 20ms, and the adjustment of the duration does not exceed 50ms. This not only helps to adjust the reading rhythm of the text, but also does not deviate too much from the target feedback.
经过在经过调整后，将得到一段确定的序列。序列中包括多个概念，每个概念都有其确定的开始时间和持续时间。这样的文字适于人工或机器朗读或以其他自然方式被大脑感知。After adjustment, a definite sequence will be obtained. The sequence includes multiple concepts, each with its own defined start time and duration. Such text is suitable for human or machine reading or other natural way of being perceived by the brain.
在步骤190，将序列在所述时间长度内以自然方式被大脑感知，以在所述大脑中产生希望的反馈。At step 190, the sequence is perceived by the brain in a natural manner for the length of time to generate the desired feedback in the brain.
将序列可以被大脑以自然方式，如听觉、视觉、触觉的方式感知，从而可以在大脑中产生反馈。例如，在语音输出装置中播放所述序列，或将所述序列以视频的方式呈现，或者以盲文的方式呈现。通过自然的方式，大脑能够在经定义的时刻开始和持续时间内感知到该序列中的多个概念，并在大脑产生希望的反馈。希望的反馈是目标反馈的概念组合序列方式的模拟，从而实现将目标经验从一个人类个体移植到另一个人类个体。The sequence can be perceived by the brain in a natural way, such as auditory, visual, tactile, so that feedback can be generated in the brain. For example, the sequence is played in a speech output device, or presented in a video, or in Braille. In a natural way, the brain is able to perceive multiple concepts in the sequence at a defined moment onset and duration and generate desired feedback in the brain. Feedback of hope is an analog of the sequential way in which the concept of goal feedback is combined, enabling the transplantation of goal experience from one human individual to another.
可以预见地，本发明的方法能够用于治疗或预防心理性疾病或精神疾病的方法，所述心理性疾病包括：凹陷、大凹陷、治疗抗性抑郁症和治疗抗性双相抑郁症、双相性精神障碍、季节性情感障碍、情绪障碍、慢性抑郁症、精神病性抑郁症、产后抑郁症、经前期焦虑症(PMDD)、情境抑郁、非典型抑郁症、躁狂、焦虑症、注意力缺陷障碍(ADD)、具有多动性的注意力缺陷障碍(ADDH)和注意力缺陷/多动性障碍(AD/HD)、双相性和躁狂性病症、强迫症、食欲过盛、月经前期综合征、物质成瘾或滥用、尼古丁成瘾、心理-性功能障碍、和假性球泡症中的一者或多者。所述精神疾病包括：精神分裂症、精神分裂症情感障碍、双相性精神障碍、强迫性精神障碍、帕金森氏精神病、相向违反性精神障碍、Charles Bonnet综合征、自闭症和Tourette氏病中的一种或多种。It is envisioned that the methods of the present invention can be used for the treatment or prevention of psychological disorders or methods of psychiatric disorders including depression, major depression, treatment-resistant depression and treatment-resistant bipolar depression, bipolar depression, bipolar depression Phase disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression, postpartum depression, premenstrual dysphoric disorder (PMDD), situational depression, atypical depression, mania, anxiety, attention deficit Disorders (ADD), Attention Deficit Disorder with Hyperactivity (ADDH) and Attention Deficit/Hyperactivity Disorder (AD/HD), Bipolar and Manic Disorders, Obsessive Compulsive Disorder, Hyperphagia, Premenstrual Syndrome symptoms, substance addiction or abuse, nicotine addiction, psycho-sexual dysfunction, and pseudobulbar disorder. The mental disorders include: schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette's disease one or more of.
根据本发明的另一个方面，本发明还提供了一种在大脑中产生反馈的系统。图10是根据本发明一个实施例的在大脑中产生反馈的系统的原理框图。如图10所示，所述系统包括大脑键盘1和传送装置2，所述大脑键盘1经配置 以生成一个具有一个时间长度的概念序列，其中在所述时间长度的多个时段内包括所述一个或多个概念；所述传送装置2经配置以将所述序列在所述时间长度内发送给受试者的大脑，在受试者的大脑中的产生希望的反馈。According to another aspect of the present invention, the present invention also provides a system for generating feedback in the brain. 10 is a functional block diagram of a system for generating feedback in the brain according to one embodiment of the present invention. As shown in Figure 10, the system includes a brain keyboard 1 and a delivery device 2, the brain keyboard 1 being configured to generate a sequence of concepts having a length of time including the One or more concepts; the delivery device 2 is configured to transmit the sequence to the subject's brain for the time length, where desired feedback is generated.
其中，大脑键盘1包括所述键盘11和处理器12。所述键盘11包括多个按键，至少一个或多个按键对应一个或多个概念。所述键盘11可以为一个物理键盘，如类似于常用的计算机用键盘，其包括各种字母、数字、字符等，也可以是一个虚拟键盘，如在计算机中运行的一个虚拟键盘应用，其显示界面具有多个按键，以供用户输入一个或多个概念。在一个实施例中，所述的键盘还包括具有时间长度的按键。在输入一个概念时，还可以同时输入对应的时间，如开始时刻、持续时间。所述处理器12与键盘11相连接，接收来自所述键盘的按键操作，根据所述按键操作形成具有一个时间长度的概念序列。所述处理器12和传送装置2相连接，传送装置2接收所述处理器12生成的所述概念序列，将其发送给一个用户，使所述概念序列在其时间长度内以自然方式被用户大脑感知，从而在用户大脑中产生希望的反馈。The brain keyboard 1 includes the keyboard 11 and the processor 12 . The keyboard 11 includes a plurality of keys, and at least one or more keys correspond to one or more concepts. The keyboard 11 can be a physical keyboard, such as a common computer keyboard, which includes various letters, numbers, characters, etc., or a virtual keyboard, such as a virtual keyboard application running in the computer, which displays The interface has multiple keys for the user to enter one or more concepts. In one embodiment, the keyboard further includes keys with a time length. When inputting a concept, you can also input the corresponding time, such as start time and duration. The processor 12 is connected to the keyboard 11, receives key operations from the keyboard, and forms a concept sequence with a time length according to the key operations. The processor 12 is connected to a delivery device 2, which receives the sequence of concepts generated by the processor 12 and sends it to a user, so that the sequence of concepts can be used by the user in a natural way over its length of time. Brain perception, resulting in desired feedback in the user's brain.
所述传送装置2可以为传送视觉、听觉或触觉信息的装置，例如为显示器，其将所述概念序列转换为视频信息，使用户大脑感受到视频信息中的概念序列。或者，所述传送装置2为音频播放装置，其将所述概念序列转换为音频信息，尤其是语音信息，使用户大脑感受到音频信息或语音信息中的概念序列。在另一种方式，可以将所述传送装置2制成可由盲人阅读的盲文阅读器，盲人用户通过阅读器可以感知所述的概念序列，从而在其大脑中产生希望的反馈。所述传送装置2中设置有开关，以控制概念序列的发出时间。The transmitting device 2 may be a device transmitting visual, auditory or tactile information, such as a display, which converts the concept sequence into video information, so that the user's brain can perceive the concept sequence in the video information. Alternatively, the transmitting device 2 is an audio playing device, which converts the conceptual sequence into audio information, especially voice information, so that the user's brain can perceive the conceptual sequence in the audio information or voice information. In another way, the delivery device 2 can be made into a Braille reader that can be read by a blind person, through which the blind user can perceive the sequence of concepts, thereby generating the desired feedback in his brain. A switch is provided in the conveying device 2 to control the time when the concept sequence is sent out.
图11是根据本发明另一个实施例的在大脑中产生反馈的系统的原理框图。在图10所示的基础上，本实施例还包括数据中心3和样本处理系统4，数据中心3存储有各种数据，例如大脑反馈样本数据及对应的概念序列，还有各种样本数据，如来自不同人、不同经验的大脑皮层数据。数据中心3包括一个或多 个数据库，如存储有来自各种经验的人的大脑皮层数据及对应的与人相关的参数的数据库、概念模式数据库等。每种经验的人的大脑皮层数据可以是多个人、多种类型的数据。所述数据是通过各种途径，在人们具有某种经验时对其大脑信息进行采集时获得。所述的经验或体验包括但不限于放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽。所述的与人相关的参数包括但不限于语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业等。大脑皮层数据可以是fMRI数据，当然也可以是MRI图像数据、CT图像数据、SEPECT图像数据、NIRS图像数据、fNIRS图像数据、PAI图像数据、脑电图或脑磁图数据等等。11 is a schematic block diagram of a system for generating feedback in the brain according to another embodiment of the present invention. On the basis shown in FIG. 10, this embodiment also includes a data center 3 and a sample processing system 4. The data center 3 stores various data, such as brain feedback sample data and corresponding concept sequences, and various sample data, Such as cerebral cortex data from different people and different experiences. The data center 3 includes one or more databases, such as a database storing cerebral cortex data and corresponding human-related parameters from people with various experiences, a conceptual schema database, and the like. The cerebral cortex data of each type of experience can be multiple people and multiple types of data. The data is obtained through various means, when information is collected from people's brains when they have a certain experience. Said experience or experience includes, but is not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, beauty. The human-related parameters include, but are not limited to, language, age, gender, religious belief, education level, occupation or previous occupation, and the like. The cerebral cortex data can be fMRI data, and of course can also be MRI image data, CT image data, SEPECT image data, NIRS image data, fNIRS image data, PAI image data, EEG or magnetoencephalography data, and so on.
所述样本处理系统4至少包括深度学习神经网络模型模块41和模式识别模型模块42，所述深度学习神经网络模型模块41用以处理来自不同人在相同经验时获得的大脑皮层数据，以得到大脑皮层中激活模式在时间上的变化；所述激活模式在时间上的变化的数据形式根据原始数据的不同而不同。模式识别模型模块42用以将在所述一个时间段内大脑皮层的激活模式的变化与多个概念在大脑皮层中的激活模式进行比较，以得到对应所述大脑皮层中激活模式在时间上的变化的一个或多个概念，并按照概念的出现时刻和持续时间排序得到一个具有时间长度的概念序列。The sample processing system 4 includes at least a deep learning neural network model module 41 and a pattern recognition model module 42, and the deep learning neural network model module 41 is used to process cerebral cortex data obtained from different people in the same experience to obtain the brain. Temporal changes in activation patterns in the cortex; the data format for the temporal changes in activation patterns varies with the original data. The pattern recognition model module 42 is used for comparing the change of the activation pattern of the cerebral cortex in the one time period with the activation pattern of a plurality of concepts in the cerebral cortex, so as to obtain the temporal variation of the activation pattern in the cerebral cortex corresponding to the cerebral cortex. Change one or more concepts, and get a concept sequence with time length sorted by the time of appearance and duration of the concepts.
本实施例中的两个模型模块使用的模型可以是积神经网络CNN模型、深度信念网络DBN模型、递归神经网络RNN模型中的一者或多者的结合，还可以是其他各种类型、采用各种其它算法的模型，本领域的普通技术人员可以参照相关领域的建模方法实现本发明中的所述的深度学习神经网络模型和模式识别模型，由于模型的建立并不是本发明的重点，在此不再赘述。The models used by the two model modules in this embodiment may be a combination of one or more of the product neural network CNN model, the deep belief network DBN model, and the recurrent neural network RNN model, or may be other various types, using Models of various other algorithms, those of ordinary skill in the art can implement the deep learning neural network model and pattern recognition model described in the present invention with reference to modeling methods in related fields, because the establishment of the model is not the focus of the present invention, It is not repeated here.
根据本发明的另一个方面，本发明提供了一种用于治疗或预防心理性疾病或精神疾病的方法，通过利用图10或图11中提供的系统，运用前述在大脑中产生反馈的方法，可以治疗或预防心理性疾病或精神疾病。所述心理性疾病包 括：凹陷、大凹陷、治疗抗性抑郁症和治疗抗性双相抑郁症、双相性精神障碍、季节性情感障碍、情绪障碍、慢性抑郁症、精神病性抑郁症、产后抑郁症、经前期焦虑症(PMDD)、情境抑郁、非典型抑郁症、躁狂、焦虑症、注意力缺陷障碍(ADD)、具有多动性的注意力缺陷障碍(ADDH)和注意力缺陷/多动性障碍(AD/HD)、双相性和躁狂性病症、强迫症、食欲过盛、月经前期综合征、物质成瘾或滥用、尼古丁成瘾、心理-性功能障碍、和假性球泡症中的一者或多者。所述精神疾病包括：精神分裂症、精神分裂症情感障碍、双相性精神障碍、强迫性精神障碍、帕金森氏精神病、相向违反性精神障碍、Charles Bonnet综合征、自闭症和Tourette氏病中的一种或多种。According to another aspect of the present invention, the present invention provides a method for the treatment or prevention of a mental illness or psychiatric disorder by utilizing the aforementioned method of generating feedback in the brain using the system provided in FIG. 10 or FIG. 11 , Mental illness or mental illness can be treated or prevented. The mental disorders include: depression, major depression, treatment-resistant depression and treatment-resistant bipolar depression, bipolar disorder, seasonal affective disorder, mood disorders, chronic depression, psychotic depression, postpartum depression Disorders, Premenstrual Dysphoria (PMDD), Situational Depression, Atypical Depression, Mania, Anxiety Disorders, Attention Deficit Disorder (ADD), Attention Deficit Disorder with Hyperactivity (ADDH), and ADHD Dysactivity disorder (AD/HD), bipolar and manic disorders, obsessive-compulsive disorder, bulimia, premenstrual syndrome, substance addiction or abuse, nicotine addiction, psycho-sexual dysfunction, and pseudobulbar one or more of the symptoms. The mental disorders include: schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's psychosis, contrarian disorder, Charles Bonnet syndrome, autism, and Tourette's disease one or more of.
根据本发明的另一个方面，提出了一种在大脑中产生反馈的系统，如图12所示，其包括生成装置1a和传送装置2a，所述生成装置1a经配置以生成一个具有一个时间长度的概念序列，其中在所述时间长度的多个时段内包括所述一个或多个概念；所述传送装置2a经配置以将所述序列在所述时间长度内以自然的方式被受试者的大脑所感知，在受试者的大脑中的产生希望的反馈。在一些实施例中，生产装置1a包括目标反馈确定装置11a以及目标反馈分析装置12a。所述目标反馈确定装置11a根据受试者要获得的经验确定目标反馈。所述的经验包括但不限于放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽等。其中，在一些实施例中，生产装置1a中包括目标反馈数据库16a，其包括了与前述任一种经验对应的目标反馈数据。例如，对于深度放松的经验，目标反馈数据库16a中存储有宗教虔诚信徒在进行完冥想、祈祷等宗教活动后大脑中形成的反馈数据。该反馈数据为通过脑信号探测技术记录的在一段时间内大脑皮层中形成的反馈数据。又例如，对于自信的经验，目标反馈数据库16a记录有在成绩优秀的运动员在参加自己擅长的比赛项目之前一段时间内在大脑中形成的反馈数据。所述反馈数据包括但不限于MRI图像数据、fMRI图像数据、CT图像数据、SEPECT图像数据和PAI图像数据、fNIRS图 像数据等等。According to another aspect of the present invention, a system for generating feedback in the brain is proposed, as shown in FIG. 12, comprising generating means 1a and transmitting means 2a, the generating means 1a being configured to generate a feedback having a length of time a sequence of concepts in which the one or more concepts are included in a plurality of periods of the time length; the delivery device 2a is configured to transmit the sequence to the subject in a natural manner during the time length The desired feedback in the subject's brain, as perceived by the brain. In some embodiments, the production device 1a includes a target feedback determination device 11a and a target feedback analysis device 12a. The target feedback determining means 11a determines the target feedback according to the experience to be obtained by the subject. Such experiences include, but are not limited to, relaxation, calm, confidence, joy, contentment, bravery, health, excitement, success, beauty, and the like. Among them, in some embodiments, the production device 1a includes a target feedback database 16a, which includes target feedback data corresponding to any of the foregoing experiences. For example, for the experience of deep relaxation, the target feedback database 16a stores the feedback data formed in the brain of religious believers after performing religious activities such as meditation and prayer. The feedback data is the feedback data formed in the cerebral cortex for a period of time recorded by the brain signal detection technology. As another example, for the experience of self-confidence, the target feedback database 16a records the feedback data formed in the brain of an outstanding athlete for a period of time before participating in the competition event he is good at. The feedback data includes, but is not limited to, MRI image data, fMRI image data, CT image data, SEPECT image data and PAI image data, fNIRS image data, and the like.
在一些实施例中，目标反馈确定装置11a包括目标反馈深度学习神经网络模型。该神经网络模型根据训练的结果，在输入个体(受试者)的与人有关的参数和经验，输出与该个体的该经验最为匹配的目标反馈。其中，所述的与人有关的参数包括个体的语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业作，所述的经验包括前述的放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽等的任一种。In some embodiments, the target feedback determination device 11a includes a target feedback deep learning neural network model. The neural network model inputs human-related parameters and experience of an individual (subject) according to the training result, and outputs target feedback that best matches the experience of the individual. Wherein, the people-related parameters include the individual's language, age, gender, religious belief, education level, occupation or previous occupation, and the experience includes the aforementioned relaxation, calmness, self-confidence, pleasure, satisfaction, courage, Any of health, excitement, success, beauty, etc.
在一些实施例中，目标反馈分析装置12a包括目标反馈时间切片模块120a和激活模式关联分析模块121a。目标反馈确定装置11a在得到目标反馈后，将其发送给所述的目标反馈分析装置12a。目标反馈分析装置12a的目标反馈时间切片模块120a获取目标反馈数据，并对其按照一定的时间段进行切片以得到多个时间段相同的时间切片，如图6所示。其中，时间切片的时间段不能过大，也不能过小，在一个实施例中，10-50ms较佳。激活模式关联分析模块121a对每一个时间切片进行激活模式的关联分析，例如，确定每个时间切片中大脑皮层的激活模式，所述激活模式包括激活部分的空间分布及信号强度，而后在再计算相邻时间切片中大脑皮层的激活模式是否相互独立。例如比较两个时间切片中大脑皮层的激活部分的空间分布是否相同或其差异是否在允许的范围内，再比较对应激活部分的信号强度，在激活部分的空间分布的差异和信号强度的差异都在允许的范围内时，确定两者非独立，否则二者独立。经过激活模式的关联分析后，得到对应的多个时间切片，每一个时间切片对应一个激活模式，如图7所示。In some embodiments, the target feedback analysis device 12a includes a target feedback time slicing module 120a and an activation mode correlation analysis module 121a. After obtaining the target feedback, the target feedback determination device 11a sends it to the target feedback analysis device 12a. The target feedback time slice module 120a of the target feedback analysis device 12a acquires the target feedback data, and slices it according to a certain time period to obtain time slices with the same time period, as shown in FIG. 6 . Wherein, the time period of the time slice cannot be too large or too small, and in one embodiment, 10-50 ms is better. The activation pattern correlation analysis module 121a performs correlation analysis of the activation pattern for each time slice, for example, determines the activation pattern of the cerebral cortex in each time slice, the activation pattern includes the spatial distribution of the activated part and the signal intensity, and then recalculates Whether the activation patterns of the cerebral cortex in adjacent time slices are independent of each other. For example, compare whether the spatial distribution of the activation part of the cerebral cortex in two time slices is the same or whether the difference is within the allowable range, and then compare the signal intensity of the corresponding activation part. The difference in the spatial distribution of the activation part and the difference in signal intensity are both Within the allowable range, it is determined that the two are not independent, otherwise the two are independent. After the correlation analysis of the activation patterns, a plurality of corresponding time slices are obtained, and each time slice corresponds to an activation pattern, as shown in FIG. 7 .
在一些实施例中，目标反馈分析装置12a包括概念组合识别模块122a。在一些实施例中，生产装置1a中包括有概念模式数据库17a，其中存储了由一个或多个概念组合而在大脑中形成的反馈数据，即对应于大脑皮层中激活区域的分布及其信号强度。概念组合识别模块122a中包括概念识别神经网络模型。 该神经网络模型根据训练的结果，输入目标反馈的一个激活模式，如图7中的一个时间切片对应的数据，则输出与该激活模式匹配的一个或多个概念组合。In some embodiments, the target feedback analysis device 12a includes a concept combination identification module 122a. In some embodiments, the production device 1a includes a conceptual pattern database 17a, which stores feedback data formed in the brain by the combination of one or more concepts, ie, corresponding to the distribution of activation areas in the cerebral cortex and their signal strengths . The concept combination recognition module 122a includes a concept recognition neural network model. According to the training result, the neural network model inputs an activation pattern of the target feedback, such as data corresponding to a time slice in Figure 7, and outputs one or more concept combinations matching the activation pattern.
在一些实施例中，生产装置1a进一步包括语义模块13a，其经配置以对概念序列进行修改，以形成连贯语义。概念组合识别模块122a得到了一个按时间排序的概念组合，每个概念具有时间属性，即起如时间和持续时间，为了更利于大脑接受，达到预期的效果，在概念组合中增加额外的词，如助词、介词、副词、实词等，以将以连贯所述概念组合的语义。例如，在概念“聆听”“潺潺”“流水”中增加“请”、“的”从而得到具有连贯语义的文字“请聆听潺潺的流水”。在增加了额外的词汇之后调整多个概念在所述时间段内出现的时刻和/或持续的时间。例如，将“请”增加到“聆听”的概念中，并将“聆听”的起始时间延后，使调整后的“请聆听”的持续时间段与调整前“聆听”的持续时间段相同。经过语义模块13a对概念序列的处理得到具有连贯语义的文字。In some embodiments, the production device 1a further includes a semantic module 13a configured to modify the sequence of concepts to form coherent semantics. The concept combination identification module 122a obtains a concept combination sorted by time, and each concept has time attributes, such as time and duration. Such as auxiliary words, prepositions, adverbs, content words, etc., to combine the semantics of the concepts described in coherence. For example, adding "please" and "de" to the concepts "listen", "gurgling" and "flowing water" to obtain the text "please listen to the gurgling water" with coherent semantics. The timing and/or duration of the multiple concepts' occurrence within the time period is adjusted after additional vocabulary has been added. For example, adding "please" to the concept of "listening" and delaying the start of "listening" so that the duration of the adjusted "please listen" is the same as the duration of the pre-adjusted "listening" . After the semantic module 13a processes the concept sequence, a text with coherent semantics is obtained.
在一些实施例中，传送装置2a包括针对听觉、视觉或触觉中一个或多个，以使大脑以自然方式感知该序列的一个或多个装置或其组合，对应地，所述系统还包括有转换装置3a。如图13所示的系统框图，转换装置3a包括音频转换模块30a、视频转换模块31a和盲文转换模块32a等。音频转换模块30a用以将代表一种经验的目标反馈的、具有连贯语义的文字转换为相应的音频信息。例如，由人工或机器朗读发出的语音信息。进一步地，还可以根据希望的经验，为语音信息配置对应的背景音乐，例如，中外古典音乐、各种自然界拟声等。视频转换模块31a根据代表一种经验的目标反馈的、具有连贯语义的文字在素材库在选择与其对应的一段视频或多段视频，并将多段视频连接成一个整段视频图像，并根据传送装置的制式，存储成相应的格式。所述盲文转换模块32a可将代表一种经验的目标反馈的、具有连贯语义的文字转换成可输出的盲文。当然，上述的转换装置3a也可以作为生产装置1a的一个转换模块而集成在生产装置1a中。In some embodiments, the delivery device 2a includes one or more devices for hearing, vision, or touch, so that the brain perceives the sequence in a natural way, or a combination thereof, correspondingly, the system further includes: Conversion device 3a. As shown in the system block diagram of FIG. 13, the conversion device 3a includes an audio conversion module 30a, a video conversion module 31a, a Braille conversion module 32a, and the like. The audio conversion module 30a is used to convert the text with coherent semantics representing the target feedback of an experience into corresponding audio information. For example, voice messages read aloud by humans or machines. Further, corresponding background music can also be configured for the voice information according to the desired experience, for example, Chinese and foreign classical music, various natural onomatopoeia, and the like. The video conversion module 31a selects a corresponding piece of video or multiple pieces of video in the material library according to the text with coherent semantics fed back by the target representing an experience, and connects the multiple pieces of video into a whole piece of video image, and according to the transmission device format, and store it in the corresponding format. The Braille conversion module 32a can convert text with coherent semantics representing target feedback of an experience into outputable Braille. Of course, the above-mentioned conversion device 3a can also be integrated in the production device 1a as a conversion module of the production device 1a.
传送装置2a可以有多种多样，例如，针对听觉的音频播放装置20a包括但不限于扬声器、耳机、音频播放机等等；其中，针对视觉的视频播放装置21a包括但不限于各种显示器，如桌上型计算机显示器、膝上型计算机显示器；各种移动终端显示屏，如手机显示屏、平板显示屏等；各种大屏幕显示器；针对触觉的盲文阅读器22a装置包括但不限于用于打印盲文的打印机、电子盲文阅读器等。The transmission device 2a can be various, for example, the audio playback device 20a for hearing includes but is not limited to speakers, earphones, audio players, etc.; wherein, the video playback device 21a for vision includes but is not limited to various displays, such as Desktop computer monitors, laptop computer monitors; various mobile terminal displays, such as mobile phone displays, flat panel displays, etc.; various large-screen displays; Braille printers, electronic Braille readers, etc.
在一些实施例中，传送装置2a还可以是兼具音频、视频播放的装置，例如VR/AR装置，包括但不限于VR/AR眼镜、VR/AR头显等。In some embodiments, the transmission device 2a may also be a device that has both audio and video playback, such as a VR/AR device, including but not limited to VR/AR glasses, VR/AR head-mounted displays, and the like.
传送装置2a还可以包括一些专业与非专业的视听室、影棚。例如，视听室可以是包括音频播放设备、视频播放设备、信号源设备等的专业或家用视听室，生产装置1a将转换好的音视频信息发送到视频室的信号源设备中，则可以在视听室中使受试者以听觉、视觉的方式感知所述序列中的概念，从而在其大脑中产生希望的反馈，并获得希望的经验。影棚可以是包括相机(如全幅机或者数码后背)、镜头、灯光、幕布、背景道具等的专业或非专业影棚。在影棚，通过演员表演的方式，使受试者以听觉、视觉的方式感知所述序列中的概念，从而在其大脑中产生希望的反馈，并获得希望的经验。The transmission device 2a may also include some professional and non-professional audio-visual rooms and studios. For example, the audio-visual room may be a professional or home audio-visual room including audio playback equipment, video playback equipment, signal source equipment, etc. The production device 1a sends the converted audio and video information to the signal source equipment of the video room, then the audio-visual room can be used in the audio-visual equipment. The room causes the subject to perceive the concepts in the sequence auditorily and visually, thereby generating the desired feedback in their brains and obtaining the desired experience. Studios can be professional or non-professional studios including cameras (such as full-frame cameras or digital backs), lenses, lights, curtains, background props, etc. In the studio, by means of actors' performances, subjects are allowed to perceive the concepts in the sequence in auditory and visual ways, so as to generate desired feedback in their brains and obtain desired experiences.
在一些实施例中，在大脑中产生反馈的系统进一步包括存储装置4a，如图14所示的系统框图。所述存储装置4a经配置以存储大脑反馈样本数据及对应的概念序列，所述存储装置可以是本地存储器、本地数据库和云数据库中的一种或多种。在一个实施例中，存储大脑反馈样本数据及对应的概念序列、前述的目标反馈数据库、概念模式数据库等全部位于云数据库中，生产装置1a在需要时，通过现有的任意一种通信机制从云数据库中获取其所需的数据，可以节省本地存储空间。在一些具体实施例中，可以将生产装置1a中的部分模块，如前述的目标反馈分析装置、转换装置等放在云中，利用强大的云计算通力以获得所需的、可由具体传送装置执行的音、视频信息。In some embodiments, the system for generating feedback in the brain further includes a storage device 4a, as shown in the system block diagram of FIG. 14 . The storage device 4a is configured to store brain feedback sample data and corresponding conceptual sequences, which may be one or more of local memory, local database, and cloud database. In one embodiment, the stored brain feedback sample data and the corresponding concept sequence, the aforementioned target feedback database, concept pattern database, etc. are all located in the cloud database, and the production device 1a, when needed, uses any existing communication mechanism to send data from Obtaining the required data from the cloud database can save local storage space. In some specific embodiments, some modules in the production device 1a, such as the aforementioned target feedback analysis device, conversion device, etc., can be placed in the cloud, and the powerful cloud computing can be used to obtain the required modules, which can be executed by the specific transmission device. audio and video information.
大脑中产生反馈的系统还包括样本处理装置5a，其经配置以利用来自于不同人在相同经验时获得的大脑皮层数据作为样本，获得对应的大脑反馈样本数据及对应的概念序列。其得到的大脑反馈样本数据及对应的概念序列存储在存储装置4a中。在一些实施例中，样本处理系统5a包括深度学习神经网络模型模块51a和模式识别模型模块52a，所述深度学习神经网络模型模块51a经配置以处理来自不同人在相同经验时获得的大脑皮层数据，以得到大脑皮层中激活模式在时间上的变化；所述模式识别模型模块52a将在所述一个时间段内大脑皮层的激活模式的变化与多个概念在大脑皮层中的激活模式进行比较，以得到对应所述大脑皮层中激活模式在时间上的变化的一个或多个概念，并按照概念的出现时刻和持续时间排序得到一个具有时间长度的概念序列。The system for generating feedback in the brain also includes a sample processing device 5a configured to obtain corresponding brain feedback sample data and corresponding concept sequences using cerebral cortex data obtained from different people during the same experience as samples. The obtained brain feedback sample data and the corresponding concept sequence are stored in the storage device 4a. In some embodiments, the sample processing system 5a includes a deep learning neural network model module 51a and a pattern recognition model module 52a, the deep learning neural network model module 51a being configured to process cerebral cortex data obtained from different people while having the same experience , to obtain the temporal change of the activation pattern in the cerebral cortex; the pattern recognition model module 52a compares the change in the activation pattern of the cerebral cortex with the activation pattern of multiple concepts in the cerebral cortex in the one time period, To obtain one or more concepts corresponding to the temporal changes of the activation patterns in the cerebral cortex, and to obtain a concept sequence with a time length by sorting the concept's appearance time and duration.
根据本发明的另一个方面，提出了一种大脑键盘，如图10中所示，其包括键盘11和处理器12，所述键盘包括多个按键，至少一个或多个按键对应一个或多个概念；处理器以接收来自所述键盘的按键操作形成具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括所述一个或多个概念；其中，所述序列在所述时间长度内以自然方式被大脑感知在所述大脑中产生希望的反馈。According to another aspect of the present invention, a brain keyboard is proposed, as shown in FIG. 10, which includes a keyboard 11 and a processor 12, the keyboard includes a plurality of keys, at least one or more keys correspond to one or more keys a concept; the processor forms a sequence having a time length by receiving key operations from the keyboard, wherein the sequence includes the one or more concepts for a plurality of periods of the time length; wherein the sequence is Being perceived by the brain in a natural manner for that length of time produces a desired feedback in the brain.
下面通过具体的实施例来说明本发明的应用效果。The application effects of the present invention will be described below through specific embodiments.
应用实施例一Application Example 1
通过以佛教信徒冥想后的大脑反馈作为“深度放松”经验的目标反馈，利用扬声器向受试者以语音的方式重复播放以本发明的方式获得的“深度放松”的序列，时长为13分钟。以受试者深度放松评估指标来确定是否形成了希望的反馈；其中评估指标包括身体疲劳恢复能力、睡眠质量及脑波呈现的深度放松占比。By taking the brain feedback of Buddhist believers after meditation as the target feedback of the "deep relaxation" experience, the speaker repeatedly played the sequence of "deep relaxation" obtained by the method of the present invention to the subjects by voice for 13 minutes. The subject's deep relaxation evaluation index was used to determine whether the desired feedback was formed; the evaluation index included physical fatigue recovery ability, sleep quality and the proportion of deep relaxation presented by brain waves.
共进行了两轮测试：A total of two rounds of testing were conducted:
第一轮测试：First round of testing:
以音频的方式，每天在一个固定时间直播深度放松概念序列13分钟，连续直播7天，并向受试者发放调查问卷，用以评估其身体疲劳恢复能力。在7天结束后共收获了120份有效调查问卷。受试者的男女性别比例为5：1；77％的受试者为在30-50岁之间的成年人；本科学历以上学历占40％，其中硕士学历6名，博士学历2名；一线城市用户占比21.67％。In the form of audio, the deep relaxation concept sequence was broadcast live at a fixed time for 13 minutes every day for 7 consecutive days, and questionnaires were distributed to the subjects to evaluate their physical fatigue recovery ability. A total of 120 valid questionnaires were harvested after 7 days. The gender ratio of the subjects is 5:1; 77% of the subjects are adults between the ages of 30 and 50; 40% have a bachelor's degree or above, of which 6 have master's degrees and 2 have doctoral degrees; Urban users accounted for 21.67%.
第二轮测试：Second round of testing:
选择42人进组参与测试，测试时间共计14天。32名进入目标测试组；其中，17人每天经大约40分钟(连续播放三次)的深度放松序列音频进行测试，另外的15人每天大约80分钟(连续播放六次)的深度放松序列音频进行测试。另有10名有经验的冥想者作为参照组。参照组的10人用自己习惯的方法每天进行40分钟冥想放松。A total of 42 people were selected to participate in the test, and the test time was 14 days. Thirty-two people entered the target test group; 17 of them were tested daily with about 40 minutes of deep relaxation sequence audio (played three times in a row), and another 15 people were tested with deep relaxation sequence audio for about 80 minutes a day (played six times in a row) . Another 10 experienced meditators served as a reference group. The 10 people in the reference group used their own methods to meditate and relax for 40 minutes a day.
在第二轮测试期间，42名参与者每天用蜗牛睡眠软件记录睡眠数据，并且每天在受试后使用Brainlink Lite智能头环测试了4分钟的脑电波。如本领域技术人员所了解的，蜗牛睡眠软件是赛博龙科技(北京)有限公司生产的一款智能睡眠监测软件。该软件用深度学习算法来学习由专业睡眠监测设备PSG(多导睡眠图仪)测量，其测量数据的准确性和正确性趋近于科研和临床的测量结果。Brainlink Lite是智能头环深圳宏智力科技有限公司生产的脑电波EEG采集智能设备，该设备是一种512采样率的单通道头带，带有前额3个干电极。During the second round of testing, 42 participants recorded sleep data with Snail Sleep software every day, and tested their brainwaves for 4 minutes every day after the test using the Brainlink Lite smart headband. As known by those skilled in the art, Snail Sleep Software is an intelligent sleep monitoring software produced by Cyberron Technology (Beijing) Co., Ltd. The software uses deep learning algorithms to learn the measurement by professional sleep monitoring equipment PSG (polysomnography), and the accuracy and correctness of its measurement data are close to the measurement results of scientific research and clinical practice. Brainlink Lite is a smart headband EEG acquisition smart device produced by Shenzhen Hongzhi Technology Co., Ltd. The device is a single-channel headband with a sampling rate of 512, with 3 dry electrodes on the forehead.
身体疲劳恢复能力指标：Physical fatigue recovery ability index:
在第一轮测试的问卷中，按照疲劳恢复能力从1到10级的10个等级，请受试者对受试前和受试第七天这两个时间段的自身疲劳恢复能力进行评估打分。经过对120份数据计算得到的平均分进行统计，如图15所示，是根据本发明一个实施例的身体疲劳恢复能力评估平均分数示意图。从受试前的平均5.53分上升到受试第七天的平均7.41分，提升了34.00％。In the questionnaire of the first round of testing, the subjects were asked to evaluate and score their own fatigue recovery ability before the test and on the seventh day of the test according to 10 levels of fatigue recovery ability from 1 to 10. . After statistic on the average score calculated from 120 pieces of data, as shown in FIG. 15 , it is a schematic diagram of the average score of physical fatigue recovery ability evaluation according to an embodiment of the present invention. From the average score of 5.53 before the test to the average score of 7.41 on the seventh day of the test, an increase of 34.00%.
在第二轮的14天的测试过程中，共进行了四次统计，图16是测试组在4次统计时对身体疲劳恢复能力的评估平均分数示意图。从图中可见，32名测试组成员在第一天测试后，身体疲劳恢复能力从之前的平均4.50分上升到5.19分，提升了15.33％；7天后的身体疲劳恢复能力从之前的4.50分上升到6.56分，提升了45.78％；14天后的身体疲劳恢复能力，从之前的4.50分上升到7.63分，提升了69.56％。During the 14-day test in the second round, a total of four statistics were performed. Figure 16 is a schematic diagram of the average score of the test group's assessment of physical fatigue recovery ability during the four statistics. It can be seen from the figure that after the first day of testing, the physical fatigue recovery ability of the 32 test group members increased from the previous average of 4.50 points to 5.19 points, an increase of 15.33%; the physical fatigue recovery ability after 7 days increased from the previous 4.50 points To 6.56 points, an increase of 45.78%; the physical fatigue recovery ability after 14 days, from the previous 4.50 points to 7.63 points, an increase of 69.56%.
图17是第二轮测试中参照组在4次统计时对身体疲劳恢复能力的评估平均分。第一天冥想后比前一天的5.2上升到5.5，提升了5.77％；连续7天后，身体疲劳恢复能力从之前的5.2上升到6.1，提升了17.31％；连续14天后，身体疲劳恢复能力从之前的5.2上升到6.8，提升了30.77％。Figure 17 is the average score of the evaluation of the physical fatigue recovery ability of the reference group in the second round of testing at 4 counts. After the first day of meditation, it rose to 5.5 from 5.2 on the previous day, an increase of 5.77%; after 7 consecutive days, the physical fatigue recovery ability increased from the previous 5.2 to 6.1, an increase of 17.31%; after 14 consecutive days, the physical fatigue recovery ability increased from before The 5.2 rose to 6.8, an increase of 30.77%.
通过对比目标测试组与参照组的数据可见，目标测试组对身体疲劳的恢复能力提高更加显著。由于目标测试组的个体并没有冥想深度放松的经验，能够通过本发明的来自冥想个体的深度放松的序列就产生超过具有冥想经验者的身体疲劳恢复数据是非常出人意料的。By comparing the data of the target test group and the reference group, it can be seen that the recovery ability of the target test group to physical fatigue is more significantly improved. Since the subjects of the target test group have no experience of meditating for deep relaxation, it is very unexpected that the sequence of deep relaxation from meditating subjects of the present invention can generate more physical fatigue recovery data than those with meditative experience.
深度放松占比指标：Proportion of deep relaxation indicators:
图18是根据目标测试组的脑波数据计算得到的脑波平均深度放松程度占比值示意图。测试前一天的脑波平均深度放松占比为7.53％；第一天测试后的脑波平均深度放松占比为10.75％；第7天的为11.66％。脑波的深度放松占比在整体上呈现随时间越来越高的趋势。FIG. 18 is a schematic diagram of the ratio of the average deep relaxation degree of the brain waves calculated according to the brain wave data of the target test group. The average deep relaxation ratio of the brainwaves on the day before the test was 7.53%; the average deep relaxation ratio of the brainwaves after the test on the first day was 10.75%; on the seventh day, it was 11.66%. The proportion of deep relaxation in brainwaves showed an increasing trend over time as a whole.
图19是根参照组的据脑波数据计算得到的脑波平均深度放松程度占比值。第一天冥想前和脑波的平均深度放松占比是1.50％，第一天冥想后，平均深度放松占比是4.20％；7天后为1.9％，并没有呈现递增趋势。FIG. 19 is the ratio of the average deep relaxation degree of the brain wave calculated according to the brain wave data of the reference group. The average proportion of deep relaxation before the first day of meditation and brain waves was 1.50%, after the first day of meditation, the average proportion of deep relaxation was 4.20%; after 7 days, it was 1.9%, and there was no increasing trend.
通过对比两组的脑波数据可见，经过本实施例的受试者的大脑放松程度更好，效果更佳。脑波数据是反映深度放松最为直接的数据。应用本发明的方法 后获得的脑波数据反映出冥想深度放松经验的成功移植。而且，递增的趋势说明随着多次应用本发明的方法在大脑中形成反馈，这种经验越来越成为受试者本身的一种经验。By comparing the brain wave data of the two groups, it can be seen that the subjects who have undergone this example have better brain relaxation and better effects. Brain wave data is the most direct data reflecting deep relaxation. The brain wave data obtained after applying the method of the present invention reflects the successful transplantation of the deep relaxation experience of meditation. Moreover, the increasing trend indicates that as multiple applications of the method of the present invention generate feedback in the brain, this experience is increasingly becoming an experience of the subject itself.
睡眠质量指标：Sleep quality indicators:
图20是第一轮测试中120人睡眠质量评分值示意图。7天睡眠质量评分从之前的平均5.8分上升到平均7.58分，提升了30.69％。Figure 20 is a schematic diagram of the sleep quality scores of 120 people in the first round of tests. The 7-day sleep quality score increased from the previous average of 5.8 to an average of 7.58, an increase of 30.69%.
图21是是第二轮测试中目标测试组的32人睡眠质量评分值示意图。开始测试前一天的睡眠质量平均分为5.25分，第一天测试后提升到5.88分，提升了12.00％；7天后提升到平均6.75分，提升了28.57％；14天后提升到平均7.5分，提升了42.86％。目标测试组的睡眠质量随着时间逐渐递增的趋势。FIG. 21 is a schematic diagram of the sleep quality score values of 32 people in the target test group in the second round of testing. The average score of sleep quality on the day before the test was 5.25 points. After the first day of testing, it increased to 5.88 points, an increase of 12.00%; after 7 days, it increased to an average of 6.75 points, an increase of 28.57%; after 14 days, it increased to an average of 7.5 points, an increase of 28.57%. up 42.86%. The sleep quality of the target test group gradually increased over time.
图22是是第二轮测试中参照组的10人睡眠质量评分值示意图。开始冥想前睡眠质量平均分为6.2分，第一天冥想后提升到6.5分，提升了4.84％；7天后提升到平均6.5分，提升了16.13％；14天后提升到平均7.4分，提升了19.35％。Figure 22 is a schematic diagram of the sleep quality score values of 10 people in the reference group in the second round of testing. The average score of sleep quality before meditating was 6.2 points. After the first day of meditation, it increased to 6.5 points, an increase of 4.84%; after 7 days, it increased to an average of 6.5 points, an increase of 16.13%; after 14 days, it increased to an average of 7.4 points, an increase of 19.35 points. %.
通过对比目标测试组与参照组的睡眠质量评分值可见，采用本发明所述方法的目标测试组更好地提高了睡眠质量。睡眠质量一方面反映了深度放松的程度，另一方面也反映了经验移植后对受试者整体产生的改变。由于睡眠问题存在的广泛性，受试者存在睡眠问题是非常普遍的，也是愿意参与本测试的原因之一。本方法的应用效果反映出，这样的经验移植并不是短期的和易逝的，而是能够对受试者整体产生改变。这与受试者本身获得类似经验几乎是相同的。进一步地，这样的整体改变对于治疗和预防受试者的心理性疾病或者精神疾病的效果也是明确的。By comparing the sleep quality score values of the target test group and the reference group, it can be seen that the target test group using the method of the present invention can better improve the sleep quality. On the one hand, sleep quality reflects the degree of deep relaxation, and on the other hand, it also reflects the overall changes in the subjects after experience transplantation. Due to the widespread nature of sleep problems, it is very common for subjects to have sleep problems, which is one of the reasons for their willingness to participate in this test. The application effect of this method reflects that such experience transplantation is not short-term and fleeting, but can produce changes to the subject as a whole. This is almost the same as the subjects themselves getting a similar experience. Further, the effect of such a global change on the treatment and prevention of a psychological disorder or psychiatric disorder in a subject is also clear.
应用实施例二Application Example 2
通过以飞行员倾听机舱中飞机起飞录音后的大脑反馈作为“平静”经验的目标反馈，利用扬声器向受试者以语音的方式发送“平静”经验序列，重复播 放序列时长为15分钟。希望的反馈体现为受试者能够减少焦虑、提高内心的平静程度，减少身体问题。利用STAI斯皮尔伯格焦虑量表的反向计分值作为平静安心程度对应的评估指标；利用对身体问题的严重程度评分(反向计分)和身份问题的改山评分(正向计分)值作为身体问题的指标。By taking the pilot's brain feedback after listening to the recording of the plane taking off in the cabin as the target feedback of the "calm" experience, the "calm" experience sequence was sent to the subjects in the form of voice through the speaker, and the duration of the sequence was repeated for 15 minutes. The hopeful feedback is that subjects are able to reduce anxiety, increase inner peace, and reduce physical problems. Use the reverse score of the STAI Spielberg Anxiety Scale as an evaluation index corresponding to the level of calmness and reassurance; use the severity score of physical problems (reverse score) and the change score of identity problems (forward score) ) value as an indicator of physical problems.
共进行了两轮测试：A total of two rounds of testing were conducted:
第一轮测试：First round of testing:
以音频的方式，每天在一个固定时间播放平静概念序列15分钟，连续直播7天，并向120名受试者发放调查问卷，用以评估其焦虑程度。In the form of audio, the calm concept sequence was played at a fixed time for 15 minutes every day for 7 consecutive days, and a questionnaire was distributed to 120 subjects to assess their anxiety level.
第二轮测试：Second round of testing:
选择42人进组参与测试，测试时间共计14天。32名进入目标测试组；其中，17人每天经大约45分钟(连续播放三次)的平静概念序列音频进行测试，另外的15人每天大约90分钟(连续播放六次)的平静概念序列音频进行测试。另有10名有经验的冥想者作为参照组。参照组的10人用自己习惯的方法每天进行40分钟冥想放松。A total of 42 people were selected to participate in the test, and the test time was 14 days. Thirty-two entered the target test group; of these, 17 were tested daily with approximately 45 minutes of audio of the calm concept sequence (played three times in a row), and another 15 were tested with approximately 90 minutes of audio of the calm concept sequence each day (played six times in a row). . Another 10 experienced meditators served as a reference group. The 10 people in the reference group used their own methods to meditate and relax for 40 minutes a day.
平静安心程度指标：Peace of mind indicator:
图23是第一轮测试中，120名受试者的STAI斯皮尔伯格焦虑量表反向计分平均值示意图。在经过连续7天的测试后，平静安心程度从之前的平均2.63分变为1.84分，改良了30.04％。Figure 23 is a schematic diagram of the average reverse score of the STAI Spielberg Anxiety Scale for 120 subjects in the first round of testing. After 7 consecutive days of testing, the level of calm and reassurance changed from the previous average of 2.63 to 1.84, an improvement of 30.04%.
图24是第二轮测试中，32名目标测试组的受试者的STAI斯皮尔伯格焦虑量表反向计分平均值。第一天测试后，平静安心程度从之前的2.53变为2.47，改良了2.37％；7天后，平静安心程度从之前的平均2.53分变为2.06分，改良了18.58％；14天后，平静安心程度从之前的平均2.53分变为1.70分，改良了32.81％。平静安心程度呈现随时间越来越高的趋势。FIG. 24 is the average reverse score of the STAI Spielberg Anxiety Scale for the subjects of the 32 target test groups in the second round of testing. After the first day of testing, the level of peace of mind changed from 2.53 to 2.47, an improvement of 2.37%; after 7 days, the level of peace of mind changed from the previous average of 2.53 to 2.06, an improvement of 18.58%; after 14 days, the level of peace of mind From the previous average of 2.53 points to 1.70 points, an improvement of 32.81%. The level of calm and reassurance showed an increasing trend over time.
图25是第二轮测试中，10名参照组的受试者的STAI斯皮尔伯格焦虑量 表反向计分平均值。参照组的受试者从冥想前一天的2.6变为2.5，改良了3.85％；7天后，平静安心程度从2.6变为2.4，改良了7.69％；14天后，平静安心程度从2.6变为2.3，改良了11.54％。Figure 25 is the average of the reverse scores of the STAI Spielberg Anxiety Scale for 10 subjects in the reference group in the second round of testing. Subjects in the reference group changed from 2.6 to 2.5 the day before meditation, an improvement of 3.85%; after 7 days, the level of peace of mind changed from 2.6 to 2.4, an improvement of 7.69%; after 14 days, the level of peace of mind changed from 2.6 to 2.3, Improved by 11.54%.
通过对比，虽然参照组的平静安心程度也呈现随时间逐渐增高的趋势，但同期比较来看，目标测试组平静安心程度的提升的幅度高于参照组，其平静和减少焦虑的效果更好。Through comparison, although the level of calm and reassurance in the reference group also showed a trend of increasing gradually over time, compared with the same period, the improvement of the level of calm and reassurance in the target test group was higher than that in the reference group, and the effect of calming and reducing anxiety was better.
身体问题指标：Physical Problem Indicators:
图26是第一轮测试中，120名受试者的身体问题评估表的反向计分平均值示意图。在第一轮测试中，120名受试者在测试完，对身体问题的严重程度(反向计分)评分从之前的7.32分变为到4.31分，改善了41.12％。Figure 26 is a schematic diagram of the reverse scoring average of the physical problem assessment form for 120 subjects in the first round of testing. In the first round of the test, 120 subjects completed the test, and the severity of the physical problem (reverse scoring) score changed from the previous 7.32 points to 4.31 points, an improvement of 41.12%.
图27是第二轮测试中，32名目标测试组受试者的身体问题改善评估表的正向计分平均值示意图。第二轮测试中，目标测试组中的32名受试者的身体问题改善程度，从开始前一天的3.90分，到第一天测试后上升到4.56，提升了16.92％；7天后，上升到5.78，提升了48.21％；14天后，上升到了7.59，提升了94.12％。FIG. 27 is a schematic diagram of the average positive score of the physical problem improvement assessment form of the 32 subjects in the target test group in the second round of testing. In the second round of testing, the physical problems of the 32 subjects in the target test group improved from 3.90 points on the day before the start to 4.56 points on the first day of testing, an increase of 16.92%; after 7 days, it rose to 4.56 points. 5.78, an increase of 48.21%; 14 days later, it rose to 7.59, an increase of 94.12%.
图28是第二轮测试中，10名参照组受试者的身体问题改善评估表的正向计分平均值示意图。参照组受试者的身体问题改善程度从开始前一天的4.8分，在第一天上升到5.4，提升了12.50％；7天后，上升到5.5，提升了14.58％；14天后，上升到6.4，提升了33.33％。Figure 28 is a schematic diagram of the average positive score of the physical problem improvement assessment form for 10 subjects in the reference group in the second round of testing. The improvement of the physical problems of the subjects in the reference group increased from 4.8 points on the day before the start to 5.4 on the first day, an increase of 12.50%; after 7 days, it increased to 5.5, an increase of 14.58%; after 14 days, it increased to 6.4, An increase of 33.33%.
通过对比可见，本发明提供的方法可以更加有效地改善身体问题。对于焦虑心理以及由于焦虑引发的身体问题的明显改善说明了本发明的方法在治疗心理性疾病和精神疾病方面的高效性。而且，无需任何药物辅助，也不会产生任何的副作用，本发明的方法当然也可以更好地应用于预防这些问题的发生。当然，如果与其他药物联合使用，本发明的方法预计能够产生更好的治疗和预 防效果。By comparison, it can be seen that the method provided by the present invention can more effectively improve physical problems. The marked improvement in anxiety, as well as physical problems due to anxiety, illustrates the high efficacy of the methods of the present invention in the treatment of psychological and psychiatric disorders. Moreover, without any drug assistance and without any side effects, the method of the present invention can of course be better applied to prevent the occurrence of these problems. Of course, the methods of the present invention are expected to produce better therapeutic and prophylactic effects if used in combination with other drugs.
上述实施例仅供说明本发明之用，而并非是对本发明的限制，有关技术领域的普通技术人员，在不脱离本发明范围的情况下，还可以做出各种变化和变型，因此，所有等同的技术方案也应属于本发明公开的范畴。The above-mentioned embodiments are only for the purpose of illustrating the present invention, rather than limiting the present invention. Those of ordinary skill in the relevant technical field can also make various changes and modifications without departing from the scope of the present invention. Therefore, all Equivalent technical solutions should also belong to the scope of the disclosure of the present invention.

Claims (41)

  1. 一种在大脑中产生反馈的方法，包括：A method of generating feedback in the brain, including:
    确定具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括一个或多个概念(concept)；以及determining a sequence having a length of time, wherein the sequence includes one or more concepts over a plurality of time periods of the length of time; and
    将所述序列在所述时间长度内以自然方式被大脑感知，以在所述大脑中产生希望的反馈(desired feedback)。The sequence is perceived by the brain in a natural way for the length of time to generate desired feedback in the brain.
  2. 根据权利要求1所述的方法，其中所述激活模式包括大脑皮层中激活部分的空间分布。The method of claim 1, wherein the activation pattern comprises a spatial distribution of activated portions in the cerebral cortex.
  3. 根据权利要求2所述的方法，其中所述希望的反馈包括大脑皮层中激活模式在时间上的变化。3. The method of claim 2, wherein the desired feedback comprises temporal changes in activation patterns in the cerebral cortex.
  4. 根据权利要求1所述的方法，其中所述希望的反馈模拟另一个大脑皮层中激活模式在时间上的变化。The method of claim 1, wherein the desired feedback simulates temporal changes in activation patterns in another cerebral cortex.
  5. 根据权利要求4所述的方法，其中作为模拟对象的另一个大脑皮层中激活模式在时间上的变化代表一种经验。5. The method of claim 4, wherein temporal changes in activation patterns in another cerebral cortex that is the subject of the simulation represent an experience.
  6. 根据权利要求4所述的方法，其中作为模拟对象的另一个大脑皮层中激活模式在时间上的变化来自一个深度学习神经网络模型。5. The method of claim 4, wherein the temporal change in the activation pattern in the other cerebral cortex that is the subject of the simulation comes from a deep learning neural network model.
  7. 根据权利要求6所述的方法，其中进一步包括：通过所述深度学习神经网络模型处理来自不同人在相同经验时获得的大脑皮层数据得到其激活模式在时间上的变化。The method according to claim 6, further comprising: processing cerebral cortex data obtained from different people during the same experience through the deep learning neural network model to obtain temporal changes in activation patterns thereof.
  8. 根据权利要求7所述的方法，其中所述大脑皮层数据为fMRI数据、MRI数据、CT数据、SEPECT数据、NIRS数据、fNIRS数据、PAI数据中的一者或多者。The method of claim 7, wherein the cerebral cortex data is one or more of fMRI data, MRI data, CT data, SEPECT data, NIRS data, fNIRS data, PAI data.
  9. 根据权利要求6所述的方法，其中另一个大脑皮层中激活模式在时间上的变化和希望的经验以及与人有关的参数相对应。7. The method of claim 6, wherein temporal changes in activation patterns in the other cerebral cortex correspond to desired experiences and human-related parameters.
  10. 根据权利要求9所述的方法，其中所述与人有关的参数包括：语言、年龄、性别、宗教信仰、教育程度、职业或曾经职业中的一者或多者。10. The method of claim 9, wherein the person-related parameters include one or more of language, age, gender, religion, education, occupation, or previous occupation.
  11. 根据权利要求9所述的方法，其中所述希望的经验包括：放松、平静、自信、愉悦、满足、勇敢、健康、兴奋、成功、美丽中的一者或多者。9. The method of claim 9, wherein the desired experience comprises one or more of relaxation, calm, confidence, joy, satisfaction, bravery, health, excitement, success, beauty.
  12. 根据权利要求4所述的方法，进一步包括：作为模拟对象的另一个大脑皮层中激活模式在时间上的变化划分为多个时间段。The method according to claim 4, further comprising: dividing the temporal change of the activation pattern in another cerebral cortex as the simulation object into a plurality of time periods.
  13. 根据权利要求12所述的方法，其中在所述多个时间段之间大脑皮层的激活模式是相互独立的。13. The method of claim 12, wherein the activation patterns of the cerebral cortex between the plurality of time periods are independent of each other.
  14. 根据权利要求12所述的方法，其中在所述一个时间段内大脑皮层的激活模式对应于一个或多个概念。13. The method of claim 12, wherein the activation pattern of the cerebral cortex during the one time period corresponds to one or more concepts.
  15. 根据权利要求14所述的方法，进一步包括：基于在所述一个时间段内大脑皮层的激活模式的变化，获取所述一个或多个概念在所述时间段内出现的时刻和持续的时间。15. The method of claim 14, further comprising: based on changes in the activation pattern of the cerebral cortex in the one time period, obtaining the time and duration of the one or more concepts in the time period.
  16. 根据权利要求15所述的方法，进一步包括：通过将在所述一个时间段内大脑皮层的激活模式的变化与多个概念在大脑皮层中的激活模式进行比较。16. The method of claim 15, further comprising: by comparing the change in the activation pattern of the cerebral cortex over the one time period with the activation pattern of the plurality of concepts in the cerebral cortex.
  17. 根据权利要求16所述的方法，进一步包括：通过模式识别模型识确定在所述一个时间段内大脑皮层的激活模式对应的一个或多个概念。The method according to claim 16, further comprising: identifying one or more concepts corresponding to the activation pattern of the cerebral cortex in the one time period through a pattern recognition model.
  18. 根据权利要求17所述的方法，进一步包括：所述模式识别模型为卷积神经网络CNN模型、深度信念网络DBN模型、递归神经网络RNN模型中的一者或多者的结合。The method according to claim 17, further comprising: the pattern recognition model is a combination of one or more of a convolutional neural network (CNN) model, a deep belief network (DBN) model, and a recurrent neural network (RNN) model.
  19. 根据权利要求15所述的方法，进一步包括：根据所述一个或多个概念在所述时间段内出现的时刻和持续的时间形成一段文字。16. The method of claim 15, further comprising forming a text based on the time and duration of the occurrence of the one or more concepts within the time period.
  20. 根据权利要求19所述的方法，进一步包括：在所述一段文字中增加文字以形成有连贯语义的内容。20. The method of claim 19, further comprising adding text to the segment of text to form coherent semantic content.
  21. 根据权利要求20所述的方法，进一步包括：根据有连贯语义的内容，调整一个或多个概念在所述时间段内出现的时刻和/或持续的时间。21. The method of claim 20, further comprising: adjusting the timing and/or duration of the occurrence of one or more concepts within the time period according to the coherent semantic content.
  22. 根据权利要求21所述的方法，其中预设的范围内调整一个或多个概 念在所述时间段内出现的时刻和/或持续的时间。21. The method of claim 21, wherein the timing and/or duration of the occurrence of one or more concepts within the time period is adjusted within a preset range.
  23. 根据权利要求1所述的方法，其中所述概念对应于一个或多个短语、词、或者语素。The method of claim 1, wherein the concept corresponds to one or more phrases, words, or morphemes.
  24. 根据权利要求23所述的方法，其中不同语言的同一短语、词、或者语素对应于同一个概念或者不同概念。24. The method of claim 23, wherein the same phrase, word, or morpheme in different languages corresponds to the same concept or different concepts.
  25. 根据权利要求1所述的方法，其中所述自然方式为听觉方式。The method of claim 1, wherein the natural manner is an auditory manner.
  26. 根据权利要求1所述的方法，其中所述自然方式为视觉方式。The method of claim 1, wherein the natural manner is a visual manner.
  27. 根据权利要求1所述的方法，其中，将所述序列在所述时间长度内以自然方式、以预置的时间间隔和预置次数被大脑重复感知。The method of claim 1, wherein the sequence is repeatedly perceived by the brain in a natural manner, at a preset time interval and a preset number of times within the time length.
  28. 一种用于治疗或预防心理性疾病或精神疾病的方法，其包括如权利要求1-26任一所述的在大脑中产生反馈的方法。A method for the treatment or prevention of a psychological or psychiatric disorder comprising the method of generating feedback in the brain as claimed in any one of claims 1-26.
  29. 如权利要求28所述的方法，所述心理性疾病包括：凹陷、大凹陷、治疗抗性抑郁症和治疗抗性双相抑郁症、双相性精神障碍、季节性情感障碍、情绪障碍、慢性抑郁症、精神病性抑郁症、产后抑郁症、经前期焦虑症(PMDD)、情境抑郁、非典型抑郁症、躁狂、焦虑症、注意力缺陷障碍(ADD)、具有多动性的注意力缺陷障碍(ADDH)和注意力缺陷/多动性障碍(AD/HD)、双相性和躁狂性病症、强迫症、食欲过盛、月经前期综合征、物质成瘾或滥用、尼古丁成瘾、心理-性功能障碍、和假性球泡症中的一者或多者。29. The method of claim 28, wherein the psychological disorders include depression, major depression, treatment-resistant depression and treatment-resistant bipolar depression, bipolar disorder, seasonal affective disorder, mood disorders, chronic depression psychotic depression, postpartum depression, premenstrual dysphoric disorder (PMDD), situational depression, atypical depression, mania, anxiety, attention deficit disorder (ADD), attention deficit disorder with hyperactivity (ADDH) and attention deficit/hyperactivity disorder (AD/HD), bipolar and manic disorders, obsessive-compulsive disorder, bulimia, premenstrual syndrome, substance addiction or abuse, nicotine addiction, psycho- One or more of sexual dysfunction, and pseudobulbar disease.
  30. 如权利要求28所述的方法，所述精神疾病包括：精神分裂症、精神分裂症情感障碍、双相性精神障碍、强迫性精神障碍、帕金森氏精神病、相向违反性精神障碍、Charles Bonnet综合征、自闭症和Tourette氏病中的一种或多种。The method of claim 28, wherein the mental illness comprises: schizophrenia, schizoaffective disorder, bipolar disorder, obsessive-compulsive disorder, Parkinson's disease, contrarian disorder, Charles Bonnet syndrome , autism and one or more of Tourette's disease.
  31. 一种在大脑中产生反馈的系统，包括：A system that produces feedback in the brain, including:
    生成装置，经配置以生成一个具有一个时间长度的概念序列，其中在所述时间长度的多个时段内包括所述一个或多个概念；和generating means configured to generate a sequence of concepts having a length of time, wherein the one or more concepts are included in a plurality of time periods of the length of time; and
    传送装置，经配置以将所述序列在所述时间长度内以自然的方式被受试者的大脑所感知，在受试者的大脑中的产生希望的反馈。A delivery device configured to cause the sequence to be perceived by the subject's brain in a natural manner for the time length, producing desired feedback in the subject's brain.
  32. 根据权利要求31所述的系统，进一步包括转换装置，经配置以将所述序列转换为音频信息或视频信息。31. The system of claim 31, further comprising a conversion device configured to convert the sequence to audio information or video information.
  33. 根据权利要求31所述的系统，进一步包括存储装置，经配置以存储大脑反馈样本数据及对应的概念序列。The system of claim 31 , further comprising a storage device configured to store brain feedback sample data and corresponding concept sequences.
  34. 根据权利要求33所述的系统，其中还包括样本处理装置，其经配置以利用来自于不同人在相同经验时获得的大脑皮层数据作为样本，获得对应的大脑反馈样本数据及对应的概念序列。34. The system of claim 33, further comprising a sample processing device configured to obtain corresponding brain feedback sample data and corresponding concept sequences using cerebral cortex data obtained from different people during the same experience as samples.
  35. 根据权利要求34所述的系统，其中所述样本处理装置包括：The system of claim 34, wherein the sample processing device comprises:
    深度学习神经网络模型模块，经配置以处理来自不同人在相同经验时获得的大脑皮层数据，以得到大脑皮层中激活模式在时间上的变化；以及a deep learning neural network model module configured to process cerebral cortical data obtained from different people during the same experience to obtain temporal changes in activation patterns in the cerebral cortex; and
    模式识别模型模块，经配置以将在所述一个时间段内大脑皮层的激活模式的变化与多个概念在大脑皮层中的激活模式进行比较，以得到对应所述大脑皮层中激活模式在时间上的变化的一个或多个概念，并按照概念的出现时刻和持续时间排序得到一个具有时间长度的概念序列。a pattern recognition model module configured to compare changes in activation patterns in the cerebral cortex over the one time period with activation patterns in the cerebral cortex for a plurality of concepts to obtain temporally corresponding activation patterns in the cerebral cortex change one or more concepts, and sort them according to the time of appearance and duration of the concepts to obtain a concept sequence with a time length.
  36. 根据权利要求31所述的系统，其中所述传送装置还包括概念转换模块，经配置以将所述概念序列转换为文字、视频信息和音频信息中的一种或多种。31. The system of claim 31, wherein the transmitting device further comprises a concept conversion module configured to convert the sequence of concepts into one or more of text, video information, and audio information.
  37. 根据权利要求31所述的系统，其中所述生产装置包括：The system of claim 31, wherein the production device comprises:
    目标反馈确定装置，经配置以根据受试者要获得的经验确定目标反馈；以及target feedback determination means configured to determine target feedback based on experience to be obtained by the subject; and
    目标反馈分析装置12a，经配置以分析所述目标反馈以获得所述具有一个时间长度的概念序列。A target feedback analysis device 12a configured to analyze the target feedback to obtain the sequence of concepts having a length of time.
  38. 根据权利要求37所述的系统，其中所述目标反馈确定装置包括目标反馈深度学习神经网络模型，其根据输入的与个体相关的与人相关的参数和经 验输出与该个体的该经验最为匹配的目标反馈。38. The system of claim 37, wherein the target feedback determining means comprises a target feedback deep learning neural network model that outputs a best match to the experience of the individual based on input human-related parameters and experience. target feedback.
  39. 根据权利要求37所述的系统，所述目标反馈分析装置包括：The system of claim 37, the target feedback analysis means comprising:
    目标反馈时间切片模块，经配置以对目标反馈数据按照一定的时间段进行切片以得到多个时间切片；a target feedback time slice module, configured to slice the target feedback data according to a certain time period to obtain a plurality of time slices;
    激活模式关联分析模块，经配置以对所述多个时间切片进行激活模式的关联分析，以得到多个按时间排序的激活模式；以及an activation pattern correlation analysis module configured to perform correlation analysis of activation patterns on the plurality of time slices to obtain a plurality of time-ordered activation patterns; and
    概念组合识别模块，经配置以为所述多个激活模式匹配与其对应的一个概念或多个概念组合。A concept combination identification module configured to match the plurality of activation patterns with a concept or combination of concepts corresponding thereto.
  40. 根据权利要求31所述的系统，所述传送装置包括以下装置中的一种或多种组合：音频播放装置、视频播放装置、盲文阅读器、VR/AR装置、视听室和影棚。31. The system of claim 31, the delivery device comprising one or more of the following: audio playback device, video playback device, Braille reader, VR/AR device, audio-visual room, and studio.
  41. 一种大脑键盘，包括：A brain keyboard comprising:
    键盘，所述键盘包括多个按键，至少一个或多个按键对应一个或多个概念；以及a keyboard comprising a plurality of keys, at least one or more of the keys corresponding to one or more concepts; and
    处理器，其经配置以接收来自所述键盘的按键操作形成具有一个时间长度的序列，其中在所述时间长度的多个时段内所述序列包括所述一个或多个概念；a processor configured to receive key operations from the keyboard to form a sequence having a length of time, wherein the sequence includes the one or more concepts for a plurality of periods of the length of time;
    其中，所述序列在所述时间长度内以自然方式被大脑感知在所述大脑中产生希望的反馈。wherein the sequence is perceived by the brain in a natural manner over the length of time resulting in a desired feedback in the brain.
PCT/CN2021/075967 2021-02-08 2021-02-08 Method, system and brain keyboard for generating feedback in brain WO2022165832A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2021/075967 WO2022165832A1 (en) 2021-02-08 2021-02-08 Method, system and brain keyboard for generating feedback in brain
CN202180016100.1A CN115335102A (en) 2021-02-08 2021-02-08 Method and system for generating feedback in brain and brain keyboard

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/075967 WO2022165832A1 (en) 2021-02-08 2021-02-08 Method, system and brain keyboard for generating feedback in brain

Publications (1)

Publication Number Publication Date
WO2022165832A1 true WO2022165832A1 (en) 2022-08-11

Family

ID=82741968

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/075967 WO2022165832A1 (en) 2021-02-08 2021-02-08 Method, system and brain keyboard for generating feedback in brain

Country Status (2)

Country Link
CN (1) CN115335102A (en)
WO (1) WO2022165832A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104871160A (en) * 2012-09-28 2015-08-26 加利福尼亚大学董事会 Systems and methods for sensory and cognitive profiling
CN108697889A (en) * 2015-11-24 2018-10-23 麻省理工学院 For preventing, mitigating and/or treating dull-witted system and method
WO2018204119A1 (en) * 2017-05-03 2018-11-08 Hrl Laboratories, Llc Method and apparatus to determine optimal brain stimulation to induce desired behavior
CN111068159A (en) * 2019-12-27 2020-04-28 兰州大学 Music feedback depression mood adjusting system based on electroencephalogram signals

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014075029A1 (en) * 2012-11-10 2014-05-15 The Regents Of The University Of California Systems and methods for evaluation of neuropathologies
CN109992113B (en) * 2019-04-09 2020-05-15 燕山大学 MI-BCI system based on multi-scene induction and control method thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104871160A (en) * 2012-09-28 2015-08-26 加利福尼亚大学董事会 Systems and methods for sensory and cognitive profiling
CN108697889A (en) * 2015-11-24 2018-10-23 麻省理工学院 For preventing, mitigating and/or treating dull-witted system and method
WO2018204119A1 (en) * 2017-05-03 2018-11-08 Hrl Laboratories, Llc Method and apparatus to determine optimal brain stimulation to induce desired behavior
CN111068159A (en) * 2019-12-27 2020-04-28 兰州大学 Music feedback depression mood adjusting system based on electroencephalogram signals

Also Published As

Publication number Publication date
CN115335102A (en) 2022-11-11

Similar Documents

Publication Publication Date Title
Alakus et al. Database for an emotion recognition system based on EEG signals and various computer games–GAMEEMO
KR102179983B1 (en) Systems and methods for analyzing brain activity and their applications
Näätänen et al. ‘Primitive intelligence’in the auditory cortex
Porcelli et al. Mirroring activity in the brain and movement determinant in the Rorschach test
Littlefield Instrumental intimacy: EEG wearables and neuroscientific control
KR102388596B1 (en) device and method that collect signals related to the brain in an active state and provide digital therapeutics information
CN116578731B (en) Multimedia information processing method, system, computer device and storage medium
Ramachandra et al. The role of mirror neurons in processing vocal emotions: evidence from psychophysiological data
Hunt A neurophenomenological description of the Guided Imagery and Music experience
Markovinović et al. Design considerations for the auditory brain computer interface speller
WO2022165832A1 (en) Method, system and brain keyboard for generating feedback in brain
Zhang et al. The use of deep learning-based gesture interactive robot in the treatment of autistic children under music perception education
Horowitz et al. Targeted dreaming increases waking creativity
Fachner et al. Ethno Therapy, Music and Trance: an EEG investigation into a sound-trance induction
Li et al. Affective virtual reality gaming for autism
Sun Study on Antidepressant Emotion Regulation Based on Feedback Analysis of Music Therapy with Brain-Computer Interface
Ignacio et al. EFFECTS OF BIOPHILIC RESTORATIVE EXPERIENCES ON DESIGNERS'BODIES, BRAINS, AND MINDS
Steffert Real-time electroencephalogram sonification for neurofeedback
Chen The Evaluation and Application of Bio-Emotion Estimation Methods
Poublan-Couzardot Neurocomputational mechanisms of meditative practice: investigating an intensive mindfulness meditation retreat within a predictive processing Bayesian framework
Swart Mechanisms and outcomes of Autonomous Sensory Meridian Response
Salehzadeh A framework to measure human behaviour whilst reading
Zhou Language comprehension grounded in perceptual experiences: an investigation on text and speech perception
Singer Time as construct and implicit coding space. A neurobiological perspective
Ergin Multi-Channel EEG-Based Emotion Classification Using Deep Neural Networks

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21923842

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21923842

Country of ref document: EP

Kind code of ref document: A1