KR102288267B1 - AI(Artificial Intelligence) BASED METHOD OF PROVIDING BRAIN INFORMATION - Google Patents

Abstract

A method for providing brain information based on artificial intelligence, according to an aspect of the present invention, comprises: a first step of in which a brain signal acquiring unit of a brain signal measuring unit emits near-infrared rays to the brain of a user and detects light which has passed through the cerebral cortex of the brain; a second step of determining, by a brain signal processing unit of the brain signal measuring unit, the degree of hemoglobin oxygenation in the cerebral blood flow of the user based on the detected light; a third step of extracting, by a brain signal analysis unit of the brain signal measuring unit, at least one brain activation region from among a plurality of regions of the user brain based on the determined degree of hemoglobin oxygenation; and a fourth step of determining, by a diagnostic unit, a brain state of the user based on the at least one brain activation region continuously extracted for a certain period of time. A signal collection operation of the brain signal acquiring unit of the first step can be performed while the user is moving. According to the present invention, the user brain state can be determined more precisely.

Description

AI-based brain information provision method {AI (Artificial Intelligence) BASED METHOD OF PROVIDING BRAIN INFORMATION}

The present invention relates to an artificial intelligence-based brain information providing method, and more particularly, an artificial intelligence-based brain capable of collecting brain-related signals while the user is active, and judging the user's brain state based on this and providing feedback How to provide information.

Depending on our behavior and environment, the brain changes circuits, creates new functions, or creates cells to grow. And how we think changes the function of brain regions.

Among the brain regions, the hippocampus, which is responsible for learning and memory, plays a role in producing new neurons (neurogenesis) and is the region where neurogenesis occurs most actively. After Hippocampus, the area where nerve cell production and neurogenesis are active is the olfactory-related area, and when a new smell is smelled, the brain creates new brain cells that differentiate the smell.

However, all neurofibrins, including brain cells, do not survive long-term, but die over time. Since the annihilation of cells means aging or degeneration, the production and neurodevelopment of cells should be active rather than extinction. Therefore, in order to activate neurofibrin, especially brain cells, it is very important to find factors affecting the speed of neurogenesis and the survival period of neurons.

When activation refers to activating brain cell production, neurogenesis, speed of neurogenesis, and survival period, exercise, food, stimulants, games, music, and meditation have been used as methods to activate the brain. . However, recently, the above factors have been discovered by neuroscientists, and methods for improving brain function by activating the brain by applying the discovered factors are being developed.

On the other hand, in the case of examining or diagnosing brain function, computer tomography (CT), magnetic resonance imaging (MRI), proton emission tomography (PET), electroencephalograph (EEG) ), magnetoencephalography (MEG), and functional magnetic resonance imaging (FMRI) are being used.

1. Republic of Korea Patent Registration No. 10-1754291 (Announced on July 6, 2017) 2. Korean Patent Publication No. 10-2016-0058812 (published on May 25, 2016) 3. Republic of Korea Patent Registration No. 10-1768393 (Announced on August 17, 2017) 4. Republic of Korea Patent Registration No. 10-1295187 (Announced on August 9, 2013)

The present invention intends to propose a method for providing brain information based on artificial intelligence in order to solve the above-mentioned conventional problems.

Specifically, the present invention aims to provide an artificial intelligence-based brain information providing method capable of collecting brain-related signals while the user is active, and judging the user's brain state based on this and providing feedback.

In addition, the present invention irradiates near-infrared rays to the user's brain, a brain signal acquisition unit that detects light that has passed through the cerebral cortex of the brain, and determines the degree of oxygenation of hemoglobin in the user's cerebral bloodstream based on the detected light Continuously with a brain signal measuring unit including a brain signal processing unit and a brain signal analyzing unit for extracting at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation and for a predetermined period of time An object of the present invention is to propose a system and method including a diagnosis unit for determining the user's brain state based on the extracted at least one brain activation region.

In addition, an object of the present invention is to provide a system and method including a brain signal stimulation unit for stimulating a user's brain for a signal collection operation of the brain signal measuring unit.

In addition, the present invention provides an auditory information collection unit that collects the user's auditory information, a gait information collection unit that collects the user's gait information, a stress information collection unit that collects the user's stress information, and the user's electrocardiogram information. Electrocardiogram information collection unit, sleep information collection unit that collects user sleep information, intensive information collection unit that collects user's intensive information, and user's face landmark masking data are collected based on user's eye tracking The user's brain state is further utilized by a non-contact emotion change collecting unit that collects the user's emotional change information in a contact manner, and a non-contact emotional change collecting unit that collects the user's emotional change information based on the user's voice change It is intended to provide a system and method for precise judgment.

In addition, the present invention additionally uses an image collection unit that collects user-related medical images, and uses an artificial brain activation region, user-related body information, and user-related medical images together to more precisely determine the user's brain state. An object of the present invention is to provide an apparatus and method for providing intelligence-based brain information.

In addition, the present invention responds to the user's brain condition determined by the diagnosis unit and the diagnosis unit that can determine the user's dementia, Parkinson's disease, stroke, epilepsy, brain diseases including brain tumors and developmental disorders, brain condition improvement It is intended to propose a system and method including a management unit that provides information for

In addition, according to the present invention, by applying a preset criterion, based on the user's brain state, it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period of time In this case, in order to improve the user's brain state, it is intended to provide a system and method that further utilizes a brain signal stimulation unit for stimulating the user's brain.

On the other hand, the technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned are clearly to those of ordinary skill in the art to which the present invention belongs from the description below. can be understood

In an artificial intelligence-based brain information providing method of an aspect of the present invention for achieving the above technical problem, the brain signal acquisition unit of the brain signal measurement unit irradiates near-infrared rays to the user's brain, and receives the light that has passed through the cerebral cortex of the brain. a first step of detecting; a second step of determining, by the brain signal processing unit of the brain signal measuring unit, the degree of oxygenation of hemoglobin in the user's cerebral blood flow based on the detected light; a third step of extracting, by the brain signal analysis unit of the brain signal measurement unit, at least one activated brain region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation; and a fourth step of determining the user's brain state based on the at least one brain activation region continuously extracted for a predetermined period by the diagnostic unit; , may be performed while the user is moving.

Also, in the second step, the degree of hemoglobin oxygenation can be determined by extracting the concentration of oxidized hemoglobin (Oxy Hb) and reduced hemoglobin concentration (Deoxy Hb) in the cerebral bloodstream of the user based on the detected light. .

In addition, in the at least one brain activation region, oxygen transported by oxidized hemoglobin in the cerebral bloodstream is consumed, so that the oxidized hemoglobin concentration (Oxy Hb) decreases, and in response to the decrease in the oxidized hemoglobin concentration (Oxy Hb) The reduced hemoglobin concentration (Deoxy Hb) may be increased.

In addition, the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) have optical characteristics that change in a visible light region and a near-infrared light region, and in the first step, the brain signal measuring unit, the near-infrared irradiation The signal may be collected through functional Near-Infrared Spectroscopy (fNIRS).

In addition, in the third step, the brain signal analyzer divides the user's brain region into a plurality, and determines whether each of the divided brain regions is changed to the brain activation region in a predetermined time unit or whether the brain is activated. It may be determined whether the region is changed to an inactive region.

Also, in the fourth step, the diagnosis unit may determine the brain state of the user by using changes in brain activation of each of the plurality of divided brain regions based on the predetermined time unit.

In addition, between the third step and the fourth step, the display unit of the brain signal measuring unit, displaying the brain activation change of each of the plurality of divided brain regions over time; may further include.

Also, in the first step, the brain signal measuring unit may be in the form of a full cover covering all of the user's head or in the form of a front part in close contact with the front part of the user's head.

In addition, before the first step, for the signal collection operation of the brain signal measuring unit, the brain signal stimulation unit for stimulating the user's brain 0.5 step; may further include.

In addition, in step 0.5, the brain activation device of the brain signal stimulation unit intensively strengthens the activity of a pre-targeted brain region within a certain space, and multi-dimensional environmental stimulation is applied to activate the function of the brain region. adding (a) step; and (b) controlling, by the management server of the brain signal stimulation unit, the driving of the brain activation device to apply a multi-dimensional environmental stimulus to the targeted brain region in the space.

In addition, the space is divided into a plurality of spaces, and the step (a) may include: implementing media art corresponding to a concept that the media unit intends to produce in each of the plurality of spaces; outputting sound by the sound unit; illuminating or illuminating the light; and transmitting the fragrance to the user by the fragrance delivery unit; may include.

In addition, in step (b), the management server includes a database in which a plurality of media information and sound information corresponding to the concept of each of the plurality of spaces, and a program for controlling the operation of the brain activation device are stored, Select one or more of a plurality of media information, sound files and programs corresponding to the concept of each space from among various information and programs stored in the database, and control the driving of the brain activation device to change according to the selected information and program there is.

In addition, the media unit includes a plurality of media modules for implementing media art corresponding to the concept of each space, and each media module includes a projector or a display panel for projecting media art including a photo or a moving image, The sound unit may include a plurality of sound modules for outputting sound corresponding to the concept of each space, and each sound module may include a speaker and an amplifier for outputting sound including music or natural white noise.

In addition, the lighting unit includes a plurality of lighting modules that irradiate or illuminate light corresponding to the concept of each space, and each lighting module includes one or more of a laser generator irradiating a laser, an LED irradiating light, and neon lighting. is configured by combining, the fragrance delivery unit includes a plurality of delivery modules for delivering the fragrance of a natural object disposed in each space to the user, and each delivery module is configured to deliver a fragrance or a fragrance of a natural object disposed in each space. It may include a blower for blowing air to deliver the imitated fragrance to the user inside each space.

In addition, after step (b), the manager terminal for controlling the operation of the brain activation device based on the manager's operation, a plurality of information files stored in the database of the management server for each module installed in each space and The method may further include (c) selecting one or more programs and controlling to change media, sound, lighting, etc. periodically or irregularly according to the selected information file and program.

In addition, after step (b), the user terminal capable of communicating with the management server and inputting the user's operation command inputs user information including age, gender, taste, and psychological state from the user, (c) transmitting the input user information to the management server; and (d) controlling the management server to determine the user's age, gender, taste, and psychological state according to the received user information, and to provide media, sound, and lighting corresponding to the personalized concept based on the determination result ; may be further included.

In addition, between the third step and the fourth step, the step 3-5 in which the user data collection unit collects body information related to the user; may further include.

In addition, in step 3-5, the auditory information collecting unit of the user data collecting unit collects the user's auditory information, and in the fourth step, the diagnosis unit, the hearing information of the user based on the collected auditory information The degree of damage is determined, and in the fourth step, the diagnosis unit may determine the brain state of the user by using the at least one brain activation region and the degree of hearing damage of the user together.

In addition, the hearing loss of the user is due to hearing loss, and the diagnosis unit additionally uses the user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of hearing loss of the user, and the auditory information collection unit-related information. It is possible to determine the user's brain state.

In addition, the hearing impairment of the user is due to tinnitus, and the diagnosis unit additionally uses the user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of tinnitus of the user, and the auditory information collection unit-related information. It is possible to determine the user's brain state.

In addition, in the first step, the signal collecting operation of the brain signal measuring unit may be performed while the user moves while the user's hearing damage occurs.

In addition, in step 3-5, the user data collection unit, a gait information collection unit for collecting the user's gait information, a stress information collection unit for collecting the user's stress information, and the user's electrocardiogram information and at least one of an electrocardiogram information collecting unit collecting the user's sleep information, a sleep information collecting unit collecting the user's sleep information, and an intensive information collecting unit collecting the user's intensive information, in the fourth step, the diagnosis unit, Based on the information collected by the user data collection unit, at least one of a gait pattern, a stress level, an electrocardiogram change, a sleep state, and a change in concentration of the user is determined based on the information collected by the user data collection unit, and in the fourth step, the diagnosis unit comprises: The user's brain state may be determined by using at least one of the user's gait pattern, stress level, electrocardiogram change, sleep state, and concentration change together with the at least one brain activation region.

In addition, the signal collection operation of the brain signal measuring unit may include a state in which the user's gait changes, a state in which the user's stress changes, a state in which the user's electrocardiogram changes, and a state in which the user's sleep condition changes. and in a situation in which at least one of a state in which the user's concentration changes is applied, may be performed while the user moves.

In addition, in step 3-5, the user data collection unit collects the user's face landmark masking data based on the user's eye tracking and collects the user's emotional change information in a non-contact manner and a non-contact emotion change collecting unit, wherein in the fourth step, the diagnosis unit may determine the user's brain state by using the at least one brain activation region and the user's emotional change information together. .

In addition, in step 3-5, the user data collection unit, a non-contact emotion change collection unit for collecting the emotion change information of the user based on the change in the user's voice; further comprising, in the fourth step, The diagnosis unit may determine the user's brain state by using the at least one brain activation region and the user's emotion change information together.

In addition, between step 3.5 and step 4, a step 3.7 of collecting a medical image related to the user by an image collecting unit; and, in the fourth step, the diagnosis unit may determine the brain state of the user by using the at least one brain activation region, body information related to the user, and a medical image related to the user together. .

In addition, the image collecting unit of step 3.7 may include: an MRI image obtaining unit collecting an MRI image related to the user; a CT image acquisition unit for collecting CT images related to the user; and an fMRI image acquisition unit that collects fMRI images related to the user.

Also, in the fourth step, the diagnosis unit may determine the user's brain disease, and the brain disease may include dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability.

In addition, after the fourth step, a fifth step of providing, by the management unit, information for improving the user's brain condition in response to the brain condition of the user determined by the diagnosis unit; may be further included.

In addition, if it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period based on the user's brain state by applying a preset criterion, the The fifth step may further include, under the control of the manager, stimulating the user's brain by the brain signal stimulation unit to improve the user's brain condition.

As described above, the AI-based brain information providing method according to the present invention may collect brain-related signals while the user is active, determine the user's brain state based on this, and provide feedback.

In addition, the present invention irradiates near-infrared rays to the user's brain, a brain signal acquisition unit that detects light that has passed through the cerebral cortex of the brain, and determines the degree of oxygenation of hemoglobin in the user's cerebral bloodstream based on the detected light Continuously with a brain signal measuring unit including a brain signal processing unit and a brain signal analyzing unit for extracting at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation and for a predetermined period of time It is possible to provide a system and method including a diagnostic unit for determining the user's brain state based on the extracted at least one brain activation region.

In addition, the present invention may provide a system and method including a brain signal stimulation unit for stimulating a user's brain for a signal collection operation of the brain signal measuring unit.

In addition, the present invention provides an auditory information collection unit that collects the user's auditory information, a gait information collection unit that collects the user's gait information, a stress information collection unit that collects the user's stress information, and the user's electrocardiogram information. Electrocardiogram information collection unit, sleep information collection unit that collects user sleep information, intensive information collection unit that collects user's intensive information, and user's face landmark masking data are collected based on user's eye tracking The user's brain state is further utilized by a non-contact emotion change collecting unit that collects the user's emotional change information in a contact manner, and a non-contact emotional change collecting unit that collects the user's emotional change information based on the user's voice change It is possible to provide a system and method for precisely judging.

In addition, the present invention additionally uses an image collection unit that collects user-related medical images, and uses an artificial brain activation region, user-related body information, and user-related medical images together to more precisely determine the user's brain state. It is possible to provide an apparatus for providing intelligence-based brain information and a control method.

In addition, the present invention responds to the user's brain condition determined by the diagnosis unit and the diagnosis unit that can determine the user's dementia, Parkinson's disease, stroke, epilepsy, brain diseases including brain tumors and developmental disorders, brain condition improvement It is possible to provide a system and method including a management unit that provides information for

In addition, according to the present invention, by applying a preset criterion, based on the user's brain state, it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period of time In this case, in order to improve the user's brain condition, a system and method further utilizing a brain signal stimulation unit for stimulating the user's brain may be provided.

The present invention can be used as a result of a study on the mechanism of degenerative brain disease. Through brain imaging and cognitive ability measurement experiment design, it is possible to develop and utilize a non-invasive diagnosis and treatment monitoring system, and through brain structure imaging and brain function imaging experiments, the brain It is possible to prepare and share the basis for building an image database, and it becomes possible to derive and propose an optimal treatment technique by identifying the treatment mechanism based on deep learning analysis.

In addition, the present invention can be utilized as an early diagnosis platform for degenerative brain disease, and can be utilized for early diagnosis through the development of a non-invasive dementia diagnosis platform through retina imaging, ophthalmic disease and eye-tracking, etc., and real-time diagnosis without time and space limitations It is possible to improve the quality of life through

In addition, the present invention can be used as an integrated customized medical device for phototherapy and diagnosis of degenerative brain disease, and it is possible to implement a prototype for commercialization of a treatment platform for degenerative brain disease, and based on the research results, it is possible to commercialize medical device products by collaborating with companies In addition, it is possible to maximize the utilization of therapeutic medical devices by linking with local dementia centers under the 'National Dementia Responsibility System'.

Furthermore, according to the present invention, it is possible to prepare standard guidelines for each field of data construction for AI learning, to secure high-quality AI data sets through step-by-step quality verification, to provide basic data essential for AI technology development, and to follow the development of AI technology in the private sector. It is possible to create a virtuous cycle ecosystem that expands and opens data autonomously.

In addition, according to the present invention, it is possible to lead the field of brain science by identifying the relationship between degenerative brain disease and treatment for hearing loss, and to secure competitiveness in the field of advanced medical image analysis through deep learning techniques for analyzing brain image data.

In addition, according to the present invention, it is possible to secure new and original technology for an effective non-invasive method for early diagnosis of degenerative brain disease, and development of a non-invasive brain disease diagnosis and treatment progress monitoring system based on brain imaging and cognitive ability experiments is possible. possible.

In addition, according to the present invention, it is possible to contribute to solving related social problems through the effects of reducing dementia disease-related mortality, increasing the treatment rate, and reducing treatment costs, and contributing to productivity improvement due to the increase in social participation of the elderly.

In addition, according to the present invention, in the age of low fertility, if you have a brain disease or a brain disease symptom is suspected due to an increase in brain diseases such as epilepsy, stroke, brain tumor, and developmental disorders in infants and children, brain health status through prevention and diagnosis can be verified.

In addition, according to the present invention, in measuring the brain signal path, the wearable type and child-friendly design product are applied, so that the experience value of the user is increased, sleep and anesthetics are not used, which can help healthy brain development. It works.

In addition, according to the present invention, it is possible to provide an emotional sympathy AI service, which is a human-centered artificial intelligence service using a multi-modal emotional data network construction and N-dimensional emotional mapping space to support non-face-to-face learning and emotional labor fields.

Specifically, according to the present invention, the user's social emotions are recognized, the emotional data network is built and the personal emotional AI service is provided, and this can be used for non-face-to-face learning and industrial field support.

On the other hand, the effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be able

1 is a block diagram of an artificial intelligence-based brain information providing apparatus in relation to the present invention.
2 is a block diagram of a brain signal measuring unit, which is a component of an artificial intelligence-based brain information providing device, in relation to the present invention.
3 is a view for explaining the operation of the brain signal measuring unit in relation to the present invention.
4A and 4B show an example of the form of a brain signal measuring unit in relation to the present invention.
5A and 5B are diagrams for explaining functional near-infrared spectroscopy applied to the present invention.
6A and 6B are flowcharts illustrating a measurement operation of a brain signal measuring unit in relation to the present invention.
7 shows an example of the entire system of the brain signal measuring unit in relation to the present invention.
8 shows an example of the signal measured by the brain signal measuring unit in relation to the present invention.
9 shows an example of visualizing and displaying the signal measured by the brain signal measuring unit as brain-related information in relation to the present invention.
10A to 10C show examples of signals measured by the brain signal measuring unit, visualized information, and processed information in relation to the present invention.
11A to 11C show an example of an authoring tool for data operation and management for artificial intelligence learning in relation to the present invention.
12 is a block diagram of a brain signal stimulation unit, which is a component of an artificial intelligence-based brain information providing device, in relation to the present invention.
13 is a block diagram illustrating a brain activation device of a brain signal stimulation unit in relation to the present invention.
14 shows an example in which the brain signal stimulation unit is constructed in space in relation to the present invention.
15 shows an example of a flowchart for explaining a brain activation method of a brain signal stimulation unit in relation to the present invention.
16A and 16B are block diagrams illustrating a user data collection unit and an image acquisition unit in relation to the present invention.
17A to 17D are diagrams for explaining the relationship between hearing loss and dementia in relation to the present invention.
18 is a diagram for explaining brain volume loss in advanced Alzheimer's disease in relation to the present invention.
19 is a view for explaining the correlation between cognitive impairment due to hearing impairment and dementia in relation to the present invention.
20 is a diagram for explaining the correlation between hearing loss (tinnitus) and Alzheimer's disease (dementia) in relation to the present invention.
21 is a view for explaining the correlation between hearing and cognitive impairment in relation to the present invention.
22 is a view for explaining a process according to a pure tone hearing test in relation to the present invention.
23 is a view for explaining an example in which an activated brain signal structure is applied to a pure tone hearing test in relation to the present invention.
24 shows an example of an active brain medical biometric AI data model/algorithm in relation to the present invention.
25 shows an example of the structure of Inceptions V3 in relation to the present invention.
26 shows an example of a SW conceptual diagram of a data quality management and verification method for artificial intelligence learning in relation to the present invention.
27 shows an example of pre-processing of data quality management and verification method for artificial intelligence learning in relation to the present invention.
28 shows an example of diagnosis of data quality management and verification method for AI learning in relation to the present invention.
29 shows an example of a data quality management and verification method for AI learning in relation to the present invention.
30 shows an example of factor search and structuring for collecting biometric data in relation to the present invention.
31A and 31B show an example of a design example for an example of a connection method of collected data and emotion mapping in relation to the present invention.
32 shows an example of a training program divided into 6 stages according to the measurement result of the child's brain development in relation to the present invention.
33A and 33B show an example of a program result before training and an example of a program result after training in relation to the present invention.
34 is a flowchart illustrating a method of collecting brain-related signals in a state in which the user is active, and determining the user's brain state based on this in relation to the present invention.
35 is a flowchart illustrating a method of displaying a change in brain activation of each of a plurality of divided brain regions over time and determining a user's brain state through this in relation to the present invention.
36 is a flowchart illustrating a method of stimulating a user's brain for a signal collection operation of a brain signal measuring unit in relation to the present invention.
37 is a flowchart illustrating a method of collecting user-related body information and additionally using it to determine a user's brain state in relation to the present invention.
38 is a flowchart for explaining a method of determining a user's hearing impairment degree and additionally using it to determine a user's brain state in relation to the present invention.
39 is a gait information collection unit that collects user's gait information, a stress information collection unit that collects user's stress information, an electrocardiogram information collection unit that collects user's ECG information, and sleep of the user in relation to the present invention; It is a flowchart explaining a method of determining a user's brain state by additionally using a sleep information collecting unit that collects information and a focused information collecting unit that collects user's focused information.
40 is a flowchart illustrating a method of determining a user's brain state by additionally using a non-contact emotion change collecting unit based on eye tracking in relation to the present invention.
41 is a flowchart illustrating a method of determining a user's brain state by additionally using a non-contact emotion change collecting unit based on a voice change in relation to the present invention.
42 is a flowchart illustrating a method of determining a user's brain state by additionally collecting and utilizing medical images related to the user in relation to the present invention.
43 is a flowchart illustrating a method of providing information for improving a user's brain state in response to the user's brain state in relation to the present invention.
44 is a flowchart illustrating a method of stimulating the brain to improve a user's brain state in relation to the present invention.

degenerative brain disease

As of 2019, the proportion of the elderly population (65 years and older) in Korea is 13.2% of the total population, but due to the rapid increase of the elderly population, it is expected that the proportion will reach 20% by 2025.

Over the next 45 years, the proportion of the youth and working age population is expected to decrease by 3.7%p and 23.3%p, respectively, and the proportion of the elderly population is expected to account for 40.1% in 2060.

In the future, the dependency ratio for the elderly will continue to rise, reaching twice that of the United States by 2050, and is expected to become the world's highest country along with Japan in 2060.

There is a rapid increase in the population of dementia predicted in an aging society. The prevalence of dementia increases with age, and from the age of 60, the prevalence doubles every 4-5 years, with more than one third of the population aged 80 years or older. showing symptoms of dementia.

Therefore, dementia is one of the important challenges to be overcome in the present age as we enter a super-aging society in the future, and the number of people with dementia in the world is 4,600, more than the population of Spain, doubling every 20 years, and by 2050 It is expected to surge to 130 million people.

The number of people with dementia in East Asia is expected to increase by 193% by 2050, and the number of dementia patients in Korea is expected to reach 2.71 million by 2050.

Therefore, the national dementia management cost is rising and the economic and social burden is occurring. The rapid increase in the number of dementia patients causes an increase in the national dementia management cost. Management costs are occurring, and are expected to double every 10 years and exceed 130 trillion won (about 2.0% of GDP) by 2050.

The economic cost of inducing dementia is two and three times higher than that of heart disease and cancer, respectively. do.

The social and economic loss in Korea due to dementia is approaching 2 trillion won a year, and it has more than doubled in the past four years (Source: Research Report of Health Insurance Policy Institute).

The incidence of degenerative brain disease is increasing due to the aging of the global population and the resulting social problems. Research for treatment and prevention of degenerative brain diseases such as dementia is active, but it is still at a standstill, and the current dementia diagnosis technology is There is a problem with low psychological, geographical, and economic accessibility as it is conducted through expensive detailed examinations at the specialized hospital level.

Currently, the screening rate of dementia patients is only 45% (more than 90% of the four major cancers), and non-invasive dementia diagnosis methods for dementia patients have been proposed in various fields. does not exist,

In addition, the failure rate of developing a treatment for dementia is about 9.6%, and solanezumab, an amyloid beta-targeting treatment that has gathered much expectation, is in a standing state, such as announcing a final failure in a recent phase 3 clinical trial.

The reason why the success rate of developing a treatment for dementia is less than 1% is that the pathogenesis of dementia has not yet been clearly identified.

Problems of conventional diagnostic techniques

The current diagnosis of dementia is made based on a comprehensive judgment based on various tests such as brain imaging, electroencephalography, blood tests, cerebrospinal fluid tests, and neuropsychological tests.

Regarding cost, brain imaging is widely used because it can directly see the form of brain atrophy, and there are several imaging techniques such as magnetic resonance imaging (MRI), single photon emission imaging (SPECT), and quantum emission tomography (PET). , and dual PET imaging is also helpful in early diagnosis.

Neuropsychological test is a test to examine overall cognitive decline by examining memory, attention, language ability, spatial and spatial sense, and computational ability. do.

However, the biggest problem with this test is that the accuracy of the test is proportional to the dementia progression rate, and there is a problem in that the misdiagnosis rate is high in the very early stage when there are no obvious symptoms.

In addition, the cerebrospinal fluid test has the advantage of being able to directly detect the accumulated biomarkers in the brain, but has the disadvantage that the cerebrospinal fluid collection process is painful and inconvenient.

Problems of the prior art related to the brain development of children

As of 2019, the number of people with intellectual disabilities is 213,000, autistic disorder is 29,000, and the population of ADHD is 2.06 million.

In the case of intellectual disability and autistic disorder, the effect through treatment is slow, whereas the ADHD population requires early diagnosis and treatment because it does not persist into adulthood if appropriate treatment is received during childhood.

In order to measure a child's brain condition with conventional MRI, fMRI, CT, etc., unlike the developmental characteristics of the child, movement must be small, and in the case of infants, the administration of sleeping pills and anesthetics is essential.

At this time, the risk of developing permanent learning disability increases more than twice when infants under 2 years of age are exposed to anesthetics several times.

In addition, in order to make various new drugs such as brain development, it is necessary not only to invest a lot of time, but also to collect bio data in real time for a certain period of time.

In the case of brain development, there is a method to check brain development through multilateral interaction, but the existing product requires a variety of methods for measuring brain development because hyper-scanning for multilateral brain state measurement is difficult. The demand for the development of condition measuring devices is expanding.

Object of the present invention

Therefore, in the present specification, it is intended to contribute to solving social problems of cranial nerve diseases by constructing a dataset for deep learning learning for the development of technologies for early diagnosis and healing of degenerative brain diseases and ear diseases, which are rapidly increasing with aging of the population.

That is, in order to solve the above problems, there is a need for active MRI dementia brain activity evaluation and early diagnosis deep learning technology.

Since the decline of the structure and function of the brain occurs long before clinical symptoms are observed, prediction or early diagnosis through brain function measurement is important for the onset or prognosis of dementia.

In order for AI to recognize and understand itself, it is important to secure large-scale AI learning data processed in a form that AI SW can understand the relationship between objects, and machine learning-based AI performance depends on the utilization of massive data collected in various environments. In the future, synergy between data and AI will be important.

As another object of the present invention, in this specification, it is intended to provide "a multi-party wireless brain signal measuring device for real-time brain measurement" and "brain development test service" of children.

Through the technology of the present invention, in order to check the brain development status of children and to grow healthy, 1) preventing and diagnosing brain diseases such as ADHD, and 2) providing experience value in nurturing concentration in daily life and learning do.

In addition, in the present specification, a device capable of collecting brain state measurement results for a certain period is proposed, predicting the risk of diseases such as brain disease based on the collected data, and brain development such as concentration applicable in daily life and learning We want to provide solutions that provide programs.

As another object of the present invention, in this specification, a multi-modal emotional data network construction and N-dimensional emotional mapping space are used to support non-face-to-face learning and emotional labor fields. ) service can be provided.

As emotional recognition technology, we define the intelligence level of social emotional recognition and emotional sympathy through event stream (user behavior) and non-contact bioreaction* using ICT, and recognize basic emotions and social emotions in the N-dimensional emotional mapping space. technology that can be applied.

In addition, as a non-contact bioreaction, it can be utilized by measuring a biosignal based on the five senses by using a sensor camera, an AI speaker, etc. available in daily life.

In addition, it is possible to gradually expand the user's behavioral events and non-contact bioreactions from a unimodal data network to a multimodal data network that can map social sensibility and propose and build a data network that can grow and evolve through artificial intelligence. there is.

In addition, as an emotional AI service, we intend to provide a new level of human-centered AI service that overcomes the unpredictable behaviors and emotions of learners and emotional workers from a cognitive level to an intuitive level by using a multi-modal emotional data network.

In order to achieve these objects of the present invention, the present invention provides an artificial intelligence-based brain information providing device capable of collecting brain-related signals in a state in which the user is active, and judging the user's brain state based on this and providing feedback. want to

In addition, the present invention irradiates near-infrared rays to the user's brain, a brain signal acquisition unit that detects light that has passed through the cerebral cortex of the brain, and determines the degree of oxygenation of hemoglobin in the user's cerebral bloodstream based on the detected light Continuously with a brain signal measuring unit including a brain signal processing unit and a brain signal analyzing unit for extracting at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation and for a predetermined period of time An object of the present invention is to propose a system including a diagnosis unit that determines the brain state of the user based on the extracted at least one brain activation region.

In addition, the present invention is to provide a system including a brain signal stimulation unit for stimulating the user's brain for the signal collection operation of the brain signal measuring unit.

In addition, the present invention provides an auditory information collection unit that collects the user's auditory information, a gait information collection unit that collects the user's gait information, a stress information collection unit that collects the user's stress information, and the user's electrocardiogram information. Electrocardiogram information collection unit, sleep information collection unit that collects user sleep information, intensive information collection unit that collects user's intensive information, and user's face landmark masking data are collected based on user's eye tracking The user's brain state is further utilized by a non-contact emotion change collecting unit that collects the user's emotional change information in a contact manner, and a non-contact emotional change collecting unit that collects the user's emotional change information based on the user's voice change We want to provide a system that makes precise judgments.

In addition, the present invention additionally uses an image collection unit that collects user-related medical images, and uses an artificial brain activation region, user-related body information, and user-related medical images together to more precisely determine the user's brain state. An object of the present invention is to provide an intelligence-based brain information providing device.

In addition, the present invention responds to the user's brain condition determined by the diagnosis unit and the diagnosis unit that can determine the user's dementia, Parkinson's disease, stroke, epilepsy, brain diseases including brain tumors and developmental disorders, brain condition improvement We would like to propose a system including a management unit that provides information for

In addition, according to the present invention, by applying a preset criterion, based on the user's brain state, it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period of time In this case, in order to improve the user's brain state, it is intended to provide a system that further utilizes a brain signal stimulation unit for stimulating the user's brain.

Hereinafter, an apparatus and method for providing brain information based on artificial intelligence according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

Hereinafter, a preferred embodiment of the present invention will be described with reference to the drawings. In addition, the embodiment described below does not unduly limit the content of the present invention described in the claims, and it cannot be said that the entire configuration described in the present embodiment is essential as a solution for the present invention.

AI-based brain information providing device

1 is a block diagram of an artificial intelligence-based brain information providing apparatus in relation to the present invention.

Referring to FIG. 1 , the apparatus 1 for providing artificial intelligence-based brain information according to the present invention includes a brain signal measurement unit 10 , a brain signal stimulation unit 20 , a user data collection unit 30 , and an image acquisition unit ( 40), a diagnosis unit 50 and a management unit 60 may be included.

First, the brain signal measuring unit 10 provides a function of collecting a signal related to the user's brain.

Specifically, the brain signal measuring unit 10 according to the present invention irradiates near-infrared rays to the user's brain, detects light that has passed through the cerebral cortex of the brain, and based on the detected light, the user's brain blood flow It provides a function of determining the degree of oxygenation of hemoglobin in the brain and extracting at least one activated brain region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation.

As a great feature of the present invention, the signal collection operation of the brain signal measuring unit 10 may be performed while the user is moving.

Next, the brain signal stimulation unit 20 provides a function of stimulating the user's brain for the signal collection operation of the brain signal measurement unit 10 .

Furthermore, the brain signal stimulation unit 20 may provide a function of stimulating the user's brain in order to improve the user's brain condition under the control of the management unit 60 .

In addition, the user data collection unit 30 provides a function of collecting body information related to the user.

The user-related body information may include the user's auditory information, gait information, stress information, electrocardiogram information, sleep information, concentration information, emotion change information, and the like.

Also, the image acquisition unit 40 provides a function of collecting medical images related to the user.

The image acquisition unit 40 according to the present invention includes an MRI image acquisition unit that collects an MRI image related to a user, a CT image acquisition unit that collects a CT image related to the user, and an fMRI image that collects an fMRI image related to the user. An acquisition unit and the like may be used.

In addition, the diagnosis unit 50 provides a function of determining the user's brain state based on the signal collected by the brain signal measuring unit 10 .

The diagnosis unit 50 may determine the user's brain disease, and the target brain disease may include dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder.

Finally, the management unit 60 provides a function of providing information for improving the user's brain condition in response to the user's brain condition determined by the diagnosis unit 50 .

For example, if it is predicted that the user will fall under at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability within a certain period based on the user's brain state by applying a preset criterion , an operation of the brain signal stimulation unit 20 for stimulating the user's brain may be triggered in order to improve the user's brain condition under the control of the manager 60 .

Hereinafter, the specific structure and function of each component of the present invention will be described with reference to the drawings.

brain signal measuring unit

2 is a block diagram of a brain signal measuring unit, which is a component of an artificial intelligence-based brain information providing device, in relation to the present invention.

Referring to FIG. 2 , the brain signal measurement unit 10 includes a brain signal acquisition unit 11 , a brain signal processing unit 12 , a brain signal analysis unit 13 , a communication unit 14 , and a display unit 15 . can do.

First, the brain signal acquisition unit 11 provides a function of irradiating near-infrared rays to the user's brain and detecting light that has passed through the cerebral cortex of the brain.

Here, the signal collection operation of the brain signal measuring unit 11 has a characteristic that is performed while the user is moving.

Next, the brain signal processing unit 12 provides a function of determining the degree of oxygenation of hemoglobin in the user's cerebral blood flow based on the detected light.

The brain signal processing unit 12 determines the degree of hemoglobin oxygenation by extracting the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) in the user's cerebral blood flow based on the detected light.

Also, the brain signal analyzer 13 provides a function of extracting at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation.

Here, in at least one brain activation region, oxygen transported by oxidized hemoglobin in the cerebral bloodstream is consumed, so that the oxidized hemoglobin concentration (Oxy Hb) decreases, and the reduction corresponds to the decrease in the oxidized hemoglobin concentration (Oxy Hb). Hemoglobin concentration (Deoxy Hb) is increased.

The oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) have optical properties that change in a visible light region and a near infrared light region. The signal is collected through functional Near-Infrared Spectroscopy (fNIRS) by irradiation.

On the other hand, the brain signal analyzer 13 divides the user's brain region into a plurality, and whether each of the divided brain regions is changed to the brain activation region in a predetermined time unit or is inactive in the brain activation region It can be determined whether the area has changed.

Based on this, the diagnosis unit 50 may determine the user's brain state using changes in brain activation of each of the plurality of divided brain regions based on a predetermined time unit.

In addition, the communication unit 14 includes components other than the brain signal measurement unit 10 (brain signal stimulation unit 20, user data collection unit 30, image acquisition unit 40, diagnosis unit 50, and management unit). (60), etc.) to establish a network between them to support data communication.

As the wireless communication technology that can be used here, WLAN (Wireless LAN) (Wi-Fi), Wibro (Wireless broadband), Wimax (World Interoperability for Microwave Access), HSDPA (High Speed Downlink Packet Access), etc. may be used. there is.

In addition, as a short range communication technology, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra-Wideband (UWB), ZigBee, etc. may be used.

In addition, the display unit 15 provides a function of displaying changes in brain activation of each of the plurality of divided brain regions over time.

The display unit 15 includes a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), a flexible display ( It may include at least one of a flexible display) and a three-dimensional display (3D display).

3 is a view for explaining the operation of the brain signal measuring unit in relation to the present invention.

Referring to FIG. 3 , the multilateral brain state and development status are measured in real time while the user is moving. That is, a brain signal measuring device capable of simultaneously measuring the brain signals of multiple users is proposed so that the brain signals can be measured and analyzed in real time and wirelessly without any special limitations.

On the other hand, Figures 4a and 4b shows an example of the form of the brain signal measuring unit in relation to the present invention.

According to the present invention, the brain signal measuring unit 10, as shown in (a) of Figure 4a, a full cover (full cover) form covering all of the user's head or as shown in (b) of Figure 4a Likewise, it may be in the form of a front part in close contact with the front part of the user's head.

In addition, as shown in (a) and (b) of Figure 4b in consideration of the book to be used for children, according to the present invention, the brain signal measurement unit 10 applies a child-specific design and configures a user-friendly environment Therefore, UI/UX customized design can be considered to provide experience value.

In the present invention, wearable brain signal measurement technology that can be used for a long time using Bluetooth Low Energy Protocol (BLE) can be applied, and it is possible to make ultra-lightweight brain signal measurement devices designed to fit the shape of a child's head using a flexible substrate.

In addition, it is possible to measure concentration and brain development in physiological aspects through brain activation analysis.

A brain signal analysis algorithm, a data transmission algorithm that can guarantee the reliability of data even if the channel state changes, and an algorithm for recognizing the state of the brain activation region can be utilized.

Meanwhile, FIGS. 5A and 5B are diagrams for explaining the functional near-infrared spectroscopy applied to the present invention.

Referring to FIG. 5A , in the functional near-infrared spectroscopy (fNIRS) technique applied to the present invention, fNIRS is a near-infrared light having a wavelength of about 650 to 1000 nm to the brain to irradiate the brain tissue. It is an imaging technology that can detect the transmitted light and non-invasively measure the change in the concentration of hemoglobin according to the change of blood flow in the brain without risk of surgery.

The fNIRS technology applied to the present invention uses functional near-infrared spectroscopy to perform real-time brain measurements in a live situation in which the patient can move rather than in a stationary situation.

The functional near-infrared spectroscopy method irradiates the brain with light in the near-infrared region with a wavelength of about 650 to 1000 nm to detect the light that has passed through the brain tissue, and non-invasively detects the change in the concentration of hemoglobin according to the change in blood flow in the brain. It is an imaging technology that can be measured without it.

This functional near-infrared spectroscopy method can show results in real time than existing evaluation methods, have lower equipment purchase and management costs, and increase reliability by directly showing sociality of the measured results.

5B is a comparison of the current brain measurement technology and the functional near-infrared spectroscopy applied to the present invention.

Meanwhile, FIGS. 6A and 6B are flowcharts illustrating a measurement operation of a brain signal measuring unit in relation to the present invention, and a data collection procedure for constructing data for artificial intelligence learning is also described.

Referring to FIG. 6A , unidentified MRI brain images are collected (S1), activated MRI brain imaging data and read clinical data are collected (S2), and feedback from a brain image reader is received (S4), By synthesizing these (S3) and pre-processing the image (S5), the clinical information can be structured into a labeled brain imaging image and an Excel file (S6).

In other words, it collects images and clinical information of inactivated MRI brains taken for MRI diagnosis from the medical database operated by the hospital, and uses the image readings and pathological readings to provide accurate hearing loss, tinnitus, and further It is possible to define the criteria for dementia.

In addition, the diagnostic results of MRI images are divided into PSAP result data for hearing loss, tinnitus, and normal PSAP image and MRI brain image images. there is.

As a detailed criterion for hearing loss to be described later, a brain imaging image of a patient confirmed with tinnitus as a result of the PSAP test may be used, and as a detailed criterion for tinnitus, a brain imaging image of a patient confirmed with tinnitus as a result of the PSAP test may be used, and normal details As a reference, a brain imaging image of a patient confirmed to be normal as a result of the PSAP test may be used.

Also, referring to FIG. 6B , an example of a protocol process including data collection and measurement is shown.

Referring to FIG. 6B, the background service is registered (S7), questionnaires are filled out before visiting the institution (S8), the institution is visited on the day of the brain measurement (S9), and an interview such as mood and vitality is conducted before the brain measurement (S9). S10), selecting a task for brain measurement (S11), performing the task (S12), and determining whether sufficient data has been collected (S13) ends the measurement (S14) or performs step S11 again.

In addition, the image data applied to the present invention may be defined as active brain MRI imaging image image data.

In addition, image data de-identification may be processed using a de-identification method and a de-identification tag structure (RAW data format).

In addition, metadata applied to the present invention may be defined as securing reliability by collecting patient's meta information for image information and utilizing it for quality inspection through additional information only for patients with hearing loss or tinnitus.

On the other hand, Figure 7 shows an example of the entire system of the brain signal measuring unit in relation to the present invention.

7 (a) shows an example of data collection equipment, (b) shows an example of moisturizing wearing a functional near-infrared spectroscopy equipment by a user, (c) is a time series of functional near-infrared spectroscopy equipment signals An example of data is shown.

Referring to (a) to (c) of Figure 7, when near-infrared rays are irradiated to the brain, when the light that has passed through the cerebral cortex is detected and processed, the concentration of oxidized hemoglobin and reduced hemoglobin is measured so that the activated region of the brain is safe without surgery. can be checked with

In addition, there is little risk in use, and there is an advantage that it is hardly affected by surrounding noise.

In addition, convenience and portability are very high, and it has the advantage of being inexpensive compared to other brain measurement imaging equipment.

8 shows an example of the signal measured by the brain signal measuring unit in relation to the present invention.

In FIG. 8, the time series block average of the functional near-infrared spectrometer signal for each channel position is shown.

9 shows an example of visualizing and displaying the signal measured by the brain signal measuring unit as brain-related information in relation to the present invention.

In FIG. 9 , an example in which the time series block average of functional near-infrared spectrometer signals for each channel position is mapped to the head is shown.

In the present invention, for the application of machine learning, a data configuration and collection method for preventing algorithm bias can be used as follows.

Regarding sample bias, it can occur when the data collected is not representative or accurately representative of the environment in which the AI system is expected to run.

No algorithm can train on the entire dataset, it usually learns on a dataset from a carefully selected subset of the entire dataset. Therefore, to reduce sample bias, when selecting a subset of datasets, it is important to select a subset that is representative of the entire dataset, not just a sufficiently large dataset.

Also, with respect to measurement bias, there is a problem in that values are systematically distorted by the device used to observe or measure something, and this kind of bias tends to skew the data in a specific direction.

A tool with measurement bias does not replicate the environment in which the model will operate, and the training data distorts the actual data, resulting in biased results. Measurement bias cannot be avoided simply by collecting more data.

Therefore, in order to solve this problem, it is possible to collect data by configuring an environment that does not bring biased results when collecting data.

Meanwhile, FIGS. 10A to 10C show examples of signals measured by the brain signal measuring unit, visualized information, and processed information in relation to the present invention.

10A is an image of an inactivated MRI brain image, and the distribution ratio of water and fat is measured by measuring the distribution of hydrogen atoms through MRI. It is possible to create an image of this other living tissue.

However, the information shown in FIG. 10A has a disadvantage in that it is impossible to grasp a phenomenon when an actual person is acting or in an active state.

In contrast, FIG. 10b shows an example of an activated MRI brain imaging image according to the present invention.

In FIG. 10B, the time series block average of the functional near-infrared spectrometer signal for each channel position is mapped to the head.

Functional Near-Infrared Spectroscopy (fNIRS) is a non-invasive method for measuring brain function that non-invasively measures changes in brain blood flow using near-infrared light. It is based on the measurement of the degree of oxygenation.

By irradiating a living body with near-infrared light to detect and process the fNIRS signal that has passed through the tissue, the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) can be measured.

Active regions of the brain consume oxygen transported by Oxy Hb in the blood, and Oxy Hb turns into Deoxy Hb. These two hemoglobins have optical properties in the visible and near-infrared regions, and their concentrations measured by functional near-infrared spectroscopy can be used as a measure of brain activity.

The method according to the present invention has the advantage of being able to measure while running or moving normally, and that results can be provided within a short time through AI after the examination.

In addition, it has the advantage of being able to measure without the use of drugs and control of ambient sound.

FIG. 10c shows an example of time-series data purification results of a functional near-infrared spectrometer signal in relation to the present invention.

In FIG. 10C , variables having the highest explanatory power are selected from among variables determined to be highly correlated with an anomaly detection target item to create a suitable model.

Also, by measuring adjust R^2 for each subset, the item consisting of the smallest number of variables among the subsets with the largest value is selected.

Meanwhile, FIGS. 11A to 11C show an example of an authoring tool for data operation and management for AI learning in relation to the present invention.

In order to efficiently build data for AI learning, an authoring tool is needed.

In other words, as an image labeling app solution, it is necessary to automatically tag and label the information of the labeling project stored in the server, the labeling target image, the label list, and the reference image.

As an authoring tool method for data operation management for AI learning, image addition, storage, and generation functions may be used.

As shown in FIG. 11A , an image to be examined can be added using the corresponding automated tool authoring tool, and after adding a brain or fracture site using a portable MRI, additional storage and generation functions can be combined by pressing the plus button. there is.

In addition, as an authoring tool utilizing the annotation function, additional information can be entered through tagging and singularity and memo functions for images that are additionally saved and created using the annotation function.

11B and 11C show examples of annotation utilization.

As a utilization method for data operation and management for artificial intelligence learning, cross validation can be performed to verify normal, deafness, and tinnitus reading demonstration models on activated MRI brain images.

In addition, it is possible to evaluate performance using various reading scales, measure sensitivity, specificity, and AUC of ROC curve as evaluation metrics, and build a reading demonstration model of medical AI learning data through github based on this.

As described above, the brain signal measuring unit 10 collects a signal related to the user's brain, and the signal collecting operation of the brain signal measuring unit 10 is performed while the user is moving.

Meanwhile, the diagnosis unit 50 determines the brain state of the user based on the collected signal.

In detail, the diagnosis unit 50 may determine the user's brain state based on the at least one brain activation region continuously extracted for a predetermined period of time.

Here, the diagnosis unit 50 may determine the user's brain disease, and the brain disease may include dementia, Parkinson's disease, stroke, epilepsy, a brain tumor, and a developmental disorder.

Furthermore, in response to the user's brain condition determined by the diagnosis unit 50, the management unit 60 providing information for improving the user's brain condition may be additionally utilized.

For example, if it is predicted that the user will fall under at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability within a certain period based on the user's brain state by applying a preset criterion , the management unit 60 may control the brain signal stimulation unit 20 to be described later so that a signal for stimulating the user's brain is provided to the user in order to improve the user's brain condition.

brain signal stimulator

The brain signal stimulation unit 20 provides a function of stimulating the user's brain for the signal collection operation of the brain signal measurement unit 10 .

Furthermore, the brain signal stimulation unit 20 may provide a function of stimulating the user's brain in order to improve the user's brain condition under the control of the management unit 60 .

12 is a block diagram of a brain signal stimulation unit that is a component of an artificial intelligence-based brain information providing device in relation to the present invention, and FIG. 13 is a block diagram of a brain activation device of a brain signal stimulation unit in relation to the present invention. A configuration diagram is shown.

The brain signal stimulation unit 20 starts by intensively stimulating a synapse, which means a part where brain cells are connected to communicate. In order to activate a brain region with long-term memory, it is necessary to stimulate the brain region through place, emotion, and story.

Therefore, the brain signal stimulation unit 20 intensively strengthens the activity of the targeted brain region through multi-dimensional environmental stimulation based on natural objects arranged in a space where activities such as exhibition, music, and movement are performed, and Optionally activate the function. According to the DSM-V (Diagnostic and Statistical Manual of Mental Disorders) brain region classification criteria defined by the American Psychiatric Association, brain regions are mainly classified into attention, spatiotemporal ability, memory, executive ability, language ability, calculation ability, and sound perception. can do. That is, the present invention

Through multidimensional environmental stimuli such as natural objects, especially plants, sounds, touch, scent, sight, and memory, the activities of various brain areas such as concentration, empathy, creativity, memory, and healing are intensively strengthened and functions are activated.

As shown in FIGS. 12 and 13, the brain activation system 10 according to a preferred embodiment of the present invention activates the function of a brain region by intensively enhancing the activity of a pre-targeted brain region in a space where natural objects are installed. A brain activation device 24 that applies a multi-dimensional environmental stimulus that is systemized to includes

As shown in FIG. 2, the brain activation device 24 implements media art corresponding to the concept to be produced in space in order to intensively enhance the activity of a pre-targeted brain region and activate the function of the brain region. It may include a media unit 30, a sound unit 40 for outputting sound, a lighting unit 50 for irradiating or illuminating light and a fragrance transmitting unit 60 for delivering the fragrance of a natural object to the user.

14 shows an example in which the brain signal stimulation unit is constructed in space in relation to the present invention.

In the embodiment according to FIG. 14 , the space, for example, the exhibition space 11, follows the light illuminating the exhibition space 11, listens to the inner world, essential nature, and finds a new inner world of mine, that is, an imagination to focus on the self. The forest is the main theme, and media art that stimulates the senses and stories through multiple spaces with different concepts is implemented.

Such an exhibition space 11 can be produced through landscape installation art, arrangement of flowers, projection mapping, light art, interactive contents, and sculpture objects.

For example, as shown in FIG. 14 , the exhibition space 11 includes first to fourth spaces 12 to 15 having respective concepts and a fifth space 16 implementing an experience space and a photo tunnel. It can be divided into a total of 5 spaces including

The first to fifth spaces 12 to 16 are designed to stimulate the sympathetic nerve through lighting, objects, and media art that can produce various colors to improve the psychological satisfaction of users who want to explore their own special self. can be implemented.

To this end, the brain activation device 24 can induce immersion into a space where colorful sculptures and media objects are arranged with intense color movements, and can induce sympathy with the user by melting dreamy media into natural objects.

For example, the concept of the first space 12 is 'INTRO - Awaken Your Brain', the concept of the second space 13 is 'STATE - Turning Back Your Nature', and the concept of the third space 14 is 'Creative - Feel Your Nature', the concept of the fourth space 15 may be 'Healing - That's Me!', and the concept of the fifth space 16 may be 'Active - Vivid & Rejuvenation'.

That is, the first space 12 is a space that activates the brain by intensively stimulating synapses while entering the space, and the second space 13 provides unfamiliar stimulation from the current starting point to the state of the brain. ), the third space 14 is a space for stimulating the brain through scent to prune, and the fourth space 15 is the first space 12 to the third space 14 . It is a space where the brain is healed as a space facing oneself from a third person's point of view while going through up to, and the fifth space 16 is a space where the brain's senses are revived through experience.

The media unit 30 includes first to fifth media modules 31 to 35 implementing media art corresponding to the concept of the first to fifth spaces 12 to 16, and the sound unit 40 includes the first to fifth media modules 31 to 35. It may include first to fifth sound modules 41 to 45 for outputting sound corresponding to the concept of the first to fifth spaces 12 to 16 .

Similarly, the lighting unit 50 includes first to fifth lighting modules 51 to 55 for irradiating or illuminating light, and the fragrance delivery unit 60 is a natural object disposed inside the exhibition space 11, that is, a tree and It may include first to fifth delivery modules 61 to 65 for delivering fragrances such as flowers to the user.

Here, each of the media modules 31 to 35 may be provided as a device that displays media art of a photo or a moving image in various ways, such as a projector or a display panel that projects the media art of a photo or video onto each space, wall, or natural object.

Each of the sound modules 41 to 45 may include a speaker and an amplifier that output music or sound, in particular, natural white noise.

Each of the lighting modules 51 to 55 may be configured by combining one or two or more of a laser generator for irradiating a laser, an LED for irradiating light, and neon lighting.

Each of the delivery modules 61 to 65 may include a blower that blows air to effectively deliver a fragrance generated in a natural object disposed in each space or a fragrance that mimics the fragrance of a natural object to a user inside each space.

Here, at least one module may be provided for each partitioned space.

Meanwhile, FIG. 15 shows an example of a flowchart for explaining a brain activation method of a brain signal stimulation unit in relation to the present invention.

In step S10 of FIG. 15 , the brain activation device 24 generates media art, sound, lighting, and fragrance corresponding to the concepts of each partitioned space of the exhibition space 11 , that is, the first to fifth spaces 12 to 16 . transmit

When the user enters the first space 12 of the exhibition space 11, the first media module, the sound module, the lighting module, and the delivery modules 31 to 61 provided in the first space 12 are respectively disposed in the first space ( 11) together with natural objects, that is, plants and objects, output media art, sound, and lighting corresponding to the concept of the first space 12, and deliver the fragrance to the user.

Accordingly, the brain activation device 24 stimulates the synapse through the user's visual, auditory, tactile, and olfactory multi-dimensional senses (S12).

When the user enters the second space 13 in step S14 , the second media module, the sound module, the lighting module, and the delivery module 32 to 62 provided in the second space 13 are respectively located in the second space 13 . Media art, sound, and lighting corresponding to the concept of the second space 13 are output together with the arranged natural objects and objects, and the fragrance is delivered to the user.

Accordingly, the brain activation device 24 implements a neural network that induces a user's sympathy with the exhibition space 11 .

When the user enters the third space 14 in step S16 , the third media module, the sound module, the lighting module, and the delivery modules 33 to 63 provided in the third space 14 are in the third space 14 , respectively. Media art corresponding to the concept of the third space 14 along with the arranged natural objects and objects, sound, lighting, and fragrance are delivered to the user.

Accordingly, the brain activation device 24 guides the movement of the user and prunes the user's brain function.

The module, sound module, lighting module, and delivery module 34 to 64 output media art, sound, and lighting corresponding to the concept of the fourth space 15 together with natural objects and objects arranged in the fourth space 15 , respectively. and deliver the fragrance to the user.

Accordingly, the brain activation device 24 provides healing through relaxation of the brain while the user walks on a forest road that is rendered as if it is alive.

When the user enters the fifth space 16 in step S20 , the fifth media module, sound module, lighting module, and delivery module 35 to 65 provided in the fifth space 16 are respectively located in the fifth space 16 . Media art, sound, and lighting corresponding to the concept of the fifth space 16 are output together with the arranged natural objects and objects, and the fragrance is delivered to the user.

Accordingly, the brain activation device 24 stimulates the five senses, such as sight and touch, through a media art assimilation that the user, especially a child, directly touches and has experiences and stories, along with natural objects made with various finishing materials made of nature. , providing a photo tunnel.

Meanwhile, in this embodiment, in order to explain the configuration of the first to fifth spaces, each space has been sequentially described, but the viewing order of the user is not limited thereto.

That is, a user who enters the second space through the first space may move not only to the third space but also to the fourth space or the fifth space along the arrow direction shown in FIG. 3 .

In addition, the user may move from the third space to the fourth space as well as to the fifth space or move back to the second space.

In addition, the user may move from the fourth space to the fifth space as well as to the second space or the third space.

Also, the user may move from the fifth space back to the second space, the third space, or the fourth space.

Therefore, the user can selectively move, rather than moving through each space in a certain order, and by moving back to the viewed space and viewing it repeatedly, the function can be strengthened by intensively activating the brain regions that are activated in each space.

Through the process as described above, the present invention can activate the brain function of the required part through multi-dimensional stimulation together with natural objects arranged in the exhibition space.

In addition, the present invention can activate various brain functions, such as concentration, creativity, memory, empathy, and healing, through multidimensional stimuli such as natural objects, especially plants and sounds, touch, scent, sight, and memory.

User data collection unit, image acquisition unit, diagnosis unit and management unit

The user data collection unit 30 provides a function of collecting body information related to the user.

The user-related body information may include the user's auditory information, gait information, stress information, electrocardiogram information, sleep information, concentration information, emotion change information, and the like.

16A is a block diagram of a user data collection unit in relation to the present invention.

Referring to FIG. 16A , the user data collection unit 30 may include an auditory information collection unit 31 that collects the user's auditory information.

At this time, the diagnosis unit 50 determines the degree of hearing damage of the user based on the collected auditory information, and uses at least one brain activation region and the degree of hearing damage of the user together to determine the brain state of the user. can judge

Specifically, the user's hearing loss is due to hearing loss, and the diagnosis unit 50 additionally adds user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of hearing loss of the user, and information related to the auditory information collection unit. It is possible to determine the user's brain state using

As another example, the user's hearing impairment is due to tinnitus, and the diagnosis unit 50 additionally adds user-related body information, the average value of the left and right pure tone thresholds of the user, the degree of tinnitus of the user, and the auditory information collection unit-related information. It is possible to determine the user's brain state using

Meanwhile, the signal collection operation of the brain signal measuring unit 10 may be performed while the user moves while the user's hearing damage occurs.

In addition, the user data collection unit 30 includes a gait information collection unit 32 that collects the user's gait information, a stress information collection unit 33 that collects the user's stress information, and the user's electrocardiogram information. It may include an electrocardiogram information collecting unit 34 , a sleep information collecting unit 35 collecting user's sleep information, and a focused information collecting unit 36 collecting user's focused information.

At this time, the diagnosis unit 50, based on the information collected by the user data collection unit 30, determines at least one of the user's gait pattern, stress level, ECG change, sleep state, and concentration change, At least one of the user's gait pattern, stress level, electrocardiogram change, sleep state, and concentration change and the at least one brain activation region are used together to determine the user's brain state.

In addition, the signal collection operation of the brain signal measuring unit 10 is a state in which the gait of the user changes, a state in which the user's stress changes, a state in which the user's electrocardiogram changes, and the sleeping condition of the user In a situation in which at least one of a changing state and a changing state of the user's concentration is applied, it may be performed while the user moves.

In addition, the user data collection unit 30 may include a non-contact emotion change collection unit 37 in a relationship with the user.

The non-contact emotion change collecting unit 37 is a non-contact emotion change collecting unit that collects the user's face landmark masking data based on the user's eye tracking and collects the user's emotion change information in a non-contact manner. can be used

In this case, the diagnosis unit 50 may determine the user's brain state by using the at least one brain activation region and the user's emotion change information together.

In addition, the non-contact emotion change collecting unit 37 may use a non-contact emotion change collecting unit that collects the user's emotion change information based on the user's voice change.

In this case, the diagnosis unit 50 may determine the user's brain state by using the at least one brain activation region and the user's emotion change information together.

In detail, the diagnosis unit 50 may determine the user's brain state based on the at least one brain activation region continuously extracted for a predetermined period of time.

Here, the diagnosis unit 50 may determine the user's brain disease, and the brain disease may include dementia, Parkinson's disease, stroke, epilepsy, a brain tumor, and a developmental disorder.

Furthermore, in response to the user's brain condition determined by the diagnosis unit 50, the management unit 60 providing information for improving the user's brain condition may be additionally utilized.

Hereinafter, functions of the diagnosis unit 50 using the auditory information collection unit 31 and the non-contact emotion change collection unit 37 will be described as representative of the types of the user data collection unit 30 .

First, an embodiment using the auditory information collecting unit 31 will be described in detail.

17A to 17D are diagrams for explaining the relationship between hearing loss and dementia in relation to the present invention, and FIG. 18 is a diagram for explaining the loss of brain volume in advanced Alzheimer's disease in relation to the present invention.

Also, FIG. 19 is a view for explaining the relationship between cognitive impairment due to hearing impairment and dementia in relation to the present invention.

Referring to FIGS. 17A to 17D , the population of dementia according to an aging society is rapidly increasing, and as Korea enters an aging society, dementia, which was only recognized as a problem for the elderly, has recently been extended to young people and is recognized as a social problem.

In particular, the proportion of severe dementia patients among dementia patients changed from 32% in 2016 to 30% in 2018. It is analyzed that early dementia is increasing.

In relation to noise-induced hearing loss, painful and unwanted loud sounds are called noises.

This is a case of damage to the sound-sensing organ, that is, the cochlea. In particular, the outer hair cells are mainly damaged. Sound below 75 dB does not cause hearing loss, but the office or conversation environment is about 60 dB, and the noise in the bus, subway, and restaurant is 80 About dB, when listening to MP3 or portable CD player with earphones, the maximum volume is about 100 dB, for a motorcycle, 120 dB, for an airplane noise, 140 dB, and a gun sound reaches 170 dB. Listening to earphones so that everyone next to you can hear it is about 100 - 115 dB. Continuous exposure to noise above 85 dB can damage your ears.

When exposed for more than 15 minutes at 100 dB without protection, and when exposed regularly at 110 dB for more than 1 minute, there is a risk of hearing loss, and the noise in buses and subways is usually around 80 dB. In order to hear, it is necessary to maintain a loudness of 90 dB or higher, and repeating this repeatedly leads to hearing loss.

The intensity of a sound is determined by the amplitude of the sound wave and is measured in dB (decibels). For every 10 dB increase in the intensity of a sound, the loudness of the sound doubles, and usually below 75 dB there is no hearing damage.

Regarding the symptoms of noise-induced hearing loss, when noise-induced hearing loss occurs, the TV or radio is also turned on loudly.

Even if the surrounding is a little noisy, it is not possible to understand the other person's story accurately and makes a different sound. Hearing loss in the high-pitched range, especially 4 kHz, is often accompanied by tinnitus.

In addition to hearing loss and tinnitus, you may experience difficulties in your normal life, including discomfort, anxiety, insomnia, fatigue, stress, and headaches. In severe cases, it affects the pulse rate and blood pressure, and can cause digestive disorders and autonomic nervous system abnormalities.

In relation to the diagnosis of noise-induced hearing loss, tests that accurately measure hearing by frequency, such as pure-tone audiometry, tinnitus test, otoacoustic radiation test, sound leak test, and auditory brainstem induced response test, are performed, and hearing that progresses slowly after long-term exposure to noise Diagnosis is made when hearing loss is evident, especially in the high-pitched region around 4 kHz.

There is no way to fundamentally restore the hearing once damaged by noise-induced hearing loss, because the already damaged hearing cells cannot be restored. In particular, if there is no sound suddenly within a few days, it may be sudden hearing loss.

In this case, it is necessary to give the ear a rest for a certain period of time so that recovery can be expected.

Sudden hearing loss is a sensorineural hearing loss that occurs suddenly without any apparent cause, called sudden hearing loss, which usually appears in one ear, but rarely bilateral. there is also

It is usually regarded as an emergency disease and requires early hospitalization, and although it is not yet clear, viral infection of the auditory nerve, inner ear blood flow disturbance, rupture of the cochlear membrane, and inner ear immune disease, neurological disease, tumor, ototoxicity The cause is presumed to be drugs.

About 1 to 2% of patients with sudden hearing loss may have an auditory nerve tumor, which is diagnosed through magnetic resonance imaging (MRI). , a combination therapy with metabolic improvers and sedatives. In addition, low-sodium and high-protein diets are recommended. It is known that the cure rate is high in patients who are actively treated at an early stage.

As a result of treatment, complete recovery is usually seen in 1/3, and hearing is improved in 1/3, but it cannot be restored to normal.

18 and 19 , the risk of dementia is doubled in the elderly with mild hearing loss and tripled in the case of moderate hearing loss.

For people with severe hearing loss, the risk is five times higher than those without hearing loss, and according to the Lancet study, hearing loss was the highest at 9% of the nine potential risk factors for developing dementia.

Hearing loss can be an early symptom of Alzheimer's disease, which impairs speech perception, especially in the presence of background noise.

Also, hearing loss increases the demand for cognitive resources, and the brain has to work much harder as hearing diminishes.

Efforts to understand and understand dialogue require the brain to “borrow” cognitive abilities from other parts of the brain, particularly memory, and basically, the more information it receives, the less information it passes into memory.

Hearing loss results in brain reorganization and/or social deprivation, leading to cognitive decline, and a decrease in auditory ability often results in reduced social interaction and engagement.

As a result, the cognitive function required to access and interact with the environment is reduced, and there are research results that can delay the progression of dementia by up to 75% through the prevention of hearing loss.

In addition, early detection and treatment, such as hearing loss, is very important in preventing or reducing cognitive decline.

Among the elderly over 60 years of age, those with mild hearing loss are twice as likely to have normal hearing loss, and those with severe hearing loss are five times more likely to have dementia (Lin, Albert, 2014).

20 is a diagram for explaining the correlation between hearing loss (tinnitus) and Alzheimer's disease (dementia) in relation to the present invention.

Referring to FIG. 20 , 30% of the population with hearing loss experienced tinnitus and required additional examination and treatment for tinnitus in addition to hearing loss treatment (Healthy Hearing, 2019). It was confirmed that it affects cognitive function (especially concentration and thinking) and affects dementia.

Eventually, tinnitus is more likely to lead to dementia (Mahoney, Rohrer, Goll, Fox, Rossor, Warren, 2010).

As a result of analyzing 18 tinnitus related literature from 1996 to 2014, tinnitus lowered cognitive ability and increased anxiety/depression (Tegg-Quinn, Bennett, Eikelboom, & Baguley, 2016).

21 is a view for explaining the correlation between hearing and cognitive impairment in relation to the present invention.

Referring to steps S21 to S24 of FIG. 21 , if there is hearing impairment, the probability of having cognitive impairment is doubled every 30 dB.

Specifically, a 30 dB difference is 2 times higher, and a 60 dB difference is 4 times higher (Uhlmann, Rees, 1989), and the cognitive function level is lowered according to a 25 dB change (As a result of tracking 347 people for 5 years, Lin, Ferrucci, Metter, An , Zonderman, & Resnick, 2011).

In the end, it leads to hearing problems > sentence listening problems > word listening > overall comprehension problems > reaction problems > mild cognitive impairment.

If there is a hearing loss every 30 dB, there is a double chance of having a cognitive impairment. 2014).

22 is a view for explaining a process according to a pure tone hearing test in relation to the present invention, and FIG. 23 is a view for explaining an example in which an activated brain signal structure is applied to a pure tone hearing test in relation to the present invention.

Referring to FIG. 23 , in the present invention, the user data collection unit 30 may include an auditory information collection unit 31 that collects the user's auditory information.

At this time, the diagnosis unit 50 determines the degree of hearing damage of the user based on the collected auditory information, and uses at least one brain activation region and the degree of hearing damage of the user together to determine the brain state of the user. can judge

Specifically, the user's hearing loss is due to hearing loss, and the diagnosis unit 50 additionally adds user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of hearing loss of the user, and information related to the auditory information collection unit. It is possible to determine the user's brain state using

As another example, the user's hearing impairment is due to tinnitus, and the diagnosis unit 50 additionally adds user-related body information, the average value of the left and right pure tone thresholds of the user, the degree of tinnitus of the user, and the auditory information collection unit-related information. It is possible to determine the user's brain state using

Meanwhile, the signal collection operation of the brain signal measuring unit 10 may be performed while the user moves while the user's hearing damage occurs.

Next, the function of the diagnosis unit 50 using the non-contact emotion change collecting unit 37 will be described.

Based on the non-contact emotion change collecting unit 37 , biometric data collection and data augmentation for personal emotion recognition may be enabled.

It is possible to collect text and facial expression data representing six basic emotions with a uniform distribution for various domains (age, gender, presence or absence of accessories, etc.).

For six basic emotions (angry, disgust, fear, happiness, sadness, surprise) and neutral, it is possible to collect large-scale daily voice data in consideration of gender, age, and region and labeling data indicating the corresponding emotion.

Basic emotion label tagging of the collected data is possible, and facial landmark masking data including eye tracking can be collected.

In addition, a voice augmentation technology for collecting pose estimation images and images for identifying the intention of a target and resolving data imbalance between emotions may be provided.

In relation to biometric data-based personal emotion recognition, emotions can be recognized through a trained deep neural network (DNN) model using the collected data.

In order to improve the emotion recognition performance, it is possible to develop and apply pre-processing technologies such as noise removal and voice section extraction for collected data, and a voice feature selection and feature extraction module suitable for emotion recognition can be applied.

In addition, deep learning-based speech unit basic emotion recognition model is applicable, and time series feature extraction of speech is possible for basic emotion inference on target data and personal emotion recognition using speech data.

Among deep learning technologies, an emotion recognition AI model using a recurrent neural network considering time series data and a transformer using an attention-focusing technique can be applied, and the designed emotion recognition AI model learning and bayesian optimization can be applied. The same hyperparameter tuning technique can be used to enable optimal emotion recognition.

In addition, it is possible to recognize eyes, nose, mouth, eyebrows, and facial contours from images and videos, and it is possible to extract concentration from the collected eye tracking data and facial landmark time series data (since continuous data is used, attention ) based artificial neural network technology available).

Social emotion recognition technology through personal emotion and daily situation tracking can be applied, emotion embedding can be advanced, emotions can be expressed in a continuous space rather than classified as one, and reinforced with an embedding space that reflects the continuity of time series data, visual, voice , multi-modal emotion analysis, etc. are applied to this by combining emotion embeddings extracted from natural language.

For the recognition of social emotion, individual emotion extraction and vectorization technology can be applied using text from speech, d-vector based speaker embedding, and previously developed speech emotion recognition technology.

It is possible to extract and apply CMVN (Cepstral Mean and Variance Normalization) to suppress performance degradation in everyday conversational voices whose tone is not flat due to emotions. A language model to which domain adaptation is applied may be applied.

The non-contact emotion change collecting unit 37 is used by a non-contact emotion change collecting unit that collects the user's facial landmark masking data based on the user's eye tracking and collects the user's emotional change information in a non-contact manner. can be

In this case, the diagnosis unit 50 may determine the user's brain state by using the at least one brain activation region and the user's emotion change information together.

Also, the non-contact emotion change collecting unit 37 may use a non-contact emotion change collecting unit that collects the user's emotion change information based on the user's voice change.

In this case, the diagnosis unit 50 may determine the user's brain state by using the at least one brain activation region and the user's emotion change information together.

In detail, the diagnosis unit 50 may determine the user's brain state based on the at least one brain activation region continuously extracted for a predetermined period of time.

Here, the diagnosis unit 50 may determine the user's brain disease, and the brain disease may include dementia, Parkinson's disease, stroke, epilepsy, a brain tumor, and a developmental disorder.

Furthermore, in response to the user's brain condition determined by the diagnosis unit 50, the management unit 60 providing information for improving the user's brain condition may be additionally utilized.

Meanwhile, FIG. 16B is a block diagram of an image acquisition unit in relation to the present invention.

The image acquisition unit 40 according to the present invention provides a function of collecting medical images related to a user.

The image acquisition unit 40 according to the present invention includes an MRI image acquisition unit 41 that collects an MRI image associated with a user, a CT image acquisition unit 42 that collects a CT image associated with the user, and an fMRI associated with the user. An fMRI image acquisition unit 43 that collects images may be used.

In this case, the diagnosis unit 50 may determine the brain state of the user by using the at least one brain activation region, the user-related body information, and the user-related medical image together.

The diagnosis unit 50 may determine the user's brain disease, and the target brain disease may include dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder.

Finally, the management unit 60 provides a function of providing information for improving the user's brain condition in response to the user's brain condition determined by the diagnosis unit 50 .

For example, if it is predicted that the user will fall under at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability within a certain period based on the user's brain state by applying a preset criterion , an operation of the brain signal stimulation unit 20 for stimulating the user's brain may be triggered in order to improve the user's brain condition under the control of the manager 60 .

AI data utilization application service

In relation to active brain medical biometric AI data, it shows high performance in various image analysis and uses inception v3 among the convolutional neural network (CNN)-based models that are most widely used among image analysis deep learning algorithms for artificial intelligence reading demonstration model. This can be applied.

In relation to the need for active brain medical biometric AI data, the present invention aims to provide basic data essential for the development of medical AI reading technology, and to create a virtuous cycle ecosystem that autonomously expands and opens data in accordance with the development of AI technology in the public and private sectors. Therefore, it is necessary to verify the validity of the constructed data.

24 shows an example of an active brain medical biometric AI data model/algorithm in relation to the present invention.

Referring to Figure 24, Inception v3 is a widely used image recognition model, which realizes an accuracy of over 78.1% in the ImageNet dataset, and requires a supervised learning process using a large set of labeled images.

The image also contains a bounding box that specifies the more precise location of the labeled object.

25 shows an example of the structure of Inceptions V3 in relation to the present invention.

Referring to FIG. 25 , the training and evaluation datasets are intentionally kept separate, and only images from the training dataset are used for model training, and only images from the evaluation dataset are used for model accuracy evaluation.

The model itself consists of symmetric/asymmetric components including convolution, average pooling, max pooling, concatenation, dropout, and fully connected layers.

26 shows an example of a SW conceptual diagram of a data quality management and verification method for artificial intelligence learning in relation to the present invention.

Referring to FIG. 26 , through the connection of source code and real-time data, pre-analysis, diagnosis, evaluation, and improvement of data quality, data storage is performed by matching the repository, which is a storage device, to reference information, and the stored data is used with an artificial intelligence engine. to implement quality control.

27 shows an example of pre-processing of data quality management and verification method for artificial intelligence learning in relation to the present invention.

Referring to FIG. 27 , it is possible to perform automation for minimizing human cost by performing quality control prior analysis, and to build an automatic identification system for each data type to automatically create index and method matching.

28 shows an example of diagnosis of data quality management and verification method for AI learning in relation to the present invention.

Referring to FIG. 28 , various types of data diagnosis technology are applied by measuring data quality, and data quality can be measured according to text, image, numerical value, waveform, etc. Data accumulation and feedback are possible.

29 shows an example of a data quality management and verification method for AI learning in relation to the present invention.

Referring to FIG. 29 , improvement and feedback are performed to target the entire quality evaluation lifecycle, evaluation is quality diagnosis level evaluation by various viewpoints, error type classification and improvement guide, and improvement is cluster analysis improvement data Recommendations and feedback on improvement results are performed.

As a method for data collection, it is possible to grasp the learner's daily data and learning life characteristics.

For example, identify data and biometric data that can be collected in daily life, set the target age to 6 years old (preschool) to 15 years old (3rd year middle school), a total of 10 grades, and collect biometric data. It is possible to search for and structure factors for the purpose of data collection, and to search for important factors (behavioral and voice) for collecting related data.

Steps S31 to S39 of FIG. 30 show an example of a factor search and structuring for collecting biometric data in relation to the present invention.

In addition, in relation to the design for emotional mapping, clustering of social emotions, determination of factors to be included for emotional mapping, classification, etc. can be applied through classification according to the characteristics of the learner.

As an example of sentiment mapping accuracy, FIG. 31A is shown.

31B is an example of a method for linking collected data in relation to the present invention. For reliable data transmission and data-to-data network advancement, advanced biometric data collection for accuracy improvement can be applied, social sensibility and learning A collection method for improving the accuracy of biometric data related to activity may be applied.

Meanwhile, FIG. 32 shows an example of a training program divided into 6 steps according to the measurement result of the child's brain development in relation to the present invention.

This model can be applied when the present invention is utilized as a “multi-party wireless brain signal measuring device for real-time brain measurement” and “brain development test service” of children.

Referring to FIG. 32 , the present invention provides a customized brain development program based on real-time brain measurement results based on “multilateral wireless brain signal measuring device” and “brain development test service”.

In other words, it is possible to provide brain development training programs such as providing brain development status measurement results (application) and concentration in daily life and learning. degree of prefrontal activation, occipital lobe activation, temporal lobe activation) and concentration level derivation (Level I-5 min, Level II-10 min, Level III-20 min), and the like.

At this time, the appropriate concentration training program for each level (reducing anxiety, knowing the pleasure of immersion, reducing glance time), brain change program (physical task performance, cognitive task performance, visual training-simple selection, auditory training-simple) selection, continuity, work function forward/reverse) may be provided.

33A and 33B show an example of a program result before training and an example of a program result after training in relation to the present invention.

In FIG. 33A , red indicates a problem related to concentration below average, green indicates high concentration above average, and green in FIG. 33B indicates that there is no problem related to concentration above average.

Accordingly, the diagnosis unit 50 may determine the user's brain condition by using the brain activation region, the user-related body information, and the user-related medical image together based on the AI, and a target brain disease silver, dementia, Parkinson's disease, stroke, epilepsy, brain tumors and developmental disorders.

Also, the management unit 60 may provide a function of providing information for improving the user's brain condition in response to the user's brain condition determined by the diagnosis unit 50 .

How to judge the user's brain state and give feedback

34 is a flowchart illustrating a method of collecting brain-related signals in a state in which the user is active, and determining the user's brain state based on this in relation to the present invention.

Referring to FIG. 34, first, the brain signal acquisition unit 11 of the brain signal measuring unit 10 irradiates near-infrared rays to the user's brain, and detects the light that has passed through the cerebral cortex of the brain (S100) carry out

In particular, in step S100 according to the present invention, the signal collection operation of the brain signal acquisition unit 11 is technically characterized in that it is performed while the user is moving.

Here, as described above, the brain signal measuring unit 11 may be implemented in the form of a full cover covering all of the user's head or in the form of a front part in close contact with the front part of the user's head.

Thereafter, the brain signal processing unit 12 of the brain signal measuring unit 10 determines the degree of oxygenation of hemoglobin in the user's cerebral blood flow based on the detected light ( S200 ).

Specifically, in step S200, the hemoglobin oxygenation degree is determined by extracting the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) in the user's brain bloodstream based on the detected light.

In addition, in at least one brain activation region, oxygen transported by oxidized hemoglobin in the cerebral bloodstream is consumed, so that the oxidized hemoglobin concentration (Oxy Hb) decreases, and the oxidized hemoglobin concentration (Oxy Hb) decreases in response to the decrease in the oxidized hemoglobin concentration (Oxy Hb). It has a characteristic that the reduced hemoglobin concentration (Deoxy Hb) is increased.

After all, the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) have optical properties that change in the visible light region and the near-infrared light region, and the brain signal measuring unit according to the present invention, The signal is collected through functional Near-Infrared Spectroscopy (fNIRS).

After step S200, the brain signal analysis unit 13 of the brain signal measuring unit 10 extracts at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation. to perform the step (S300).

Thereafter, the diagnosis unit 50 determines the brain state of the user based on the at least one brain activation region continuously extracted for a predetermined period ( S400 ).

How to display and judge brain state

Extracting, by the brain signal analysis unit 13 of the brain signal measuring unit 10, an activated at least one brain activation region among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation (S300) will be described in more detail.

35 is a flowchart illustrating a method of displaying a change in brain activation of each of a plurality of divided brain regions over time and determining a user's brain state through this in relation to the present invention.

Referring to FIG. 35 , step S300 includes dividing the user's brain region into a plurality ( S310 ), whether each of the divided brain regions is changed to the brain activation region in a predetermined time unit or in the brain activation region. and determining whether it has changed to an inactive area ( S320 ).

Thereafter, the display unit 15 of the brain signal measuring unit 10 performs a step (S330) of displaying changes in brain activation of each of the plurality of divided brain regions over time.

In addition, based on a predetermined time unit, the step (S400) of determining the user's brain state using changes in brain activation of each of the plurality of divided brain regions is performed.

brain stimulation method

On the other hand, before the step S100, the step of stimulating the brain may be additionally utilized.

36 is a flowchart illustrating a method of stimulating a user's brain for a signal collection operation of a brain signal measuring unit in relation to the present invention.

Referring to FIG. 36 , first, for the signal collection operation of the brain signal measuring unit 10 , the brain signal stimulating unit 20 stimulating the user's brain ( S50 ) is performed.

Thereafter, the brain signal acquisition unit 11 of the brain signal measurement unit 10 irradiates near-infrared rays to the user's brain and detects the light that has passed through the cerebral cortex of the brain ( S100 ).

Meanwhile, in step S50, the brain activation device of the brain signal stimulation unit intensively strengthens the activity of a pre-targeted brain region within a certain space, and applies a multi-dimensional environmental stimulus to activate the function of the brain region (a ) and (b) of the management server of the brain signal stimulation unit controlling the driving of the brain activation device to apply a multi-dimensional environmental stimulus to the targeted brain region in the space.

In addition, the space according to the present invention is partitioned into a plurality of spaces, and in step (a), the media unit implements media art corresponding to a concept to be produced in each of the plurality of spaces, and the sound unit outputs sound. Step, the step of irradiating or illuminating the light by the lighting unit, and the fragrance delivery unit may include the step of delivering the fragrance to the user.

In addition, in step (b), the management server includes a database in which a plurality of media information and sound information corresponding to the concept of each of the plurality of spaces, and a program for controlling the operation of the brain activation device are stored, It is possible to select one or more of a plurality of media information, sound files and programs corresponding to the concept of each space from among various information and programs stored in the database, and control the driving of the brain activation device to change according to the selected information and program .

In addition, the media unit includes a plurality of media modules for implementing media art corresponding to the concept of each space, and each media module includes a projector or a display panel for projecting media art including a photo or a moving image, The sound unit may include a plurality of sound modules for outputting sound corresponding to the concept of each space, and each sound module may include a speaker and an amplifier for outputting sound including music or natural white noise.

In addition, the lighting unit includes a plurality of lighting modules for irradiating or illuminating light corresponding to the concept of each space, and each lighting module includes one or two or more of a laser generator irradiating a laser, an LED irradiating light, and a neon light. Consisting of a combination, the fragrance delivery unit includes a plurality of delivery modules that deliver the fragrance of a natural object disposed in each space to the user, and each delivery module mimics the fragrance or the fragrance of a natural object disposed in each space. It may include a blower for blowing air to deliver a fragrance to a user inside each space.

In addition, after step (b), the manager terminal for controlling the operation of the brain activation device based on the manager's operation, a plurality of information files stored in the database of the management server for each module installed in each space and The method may further include the step (c) of selecting one or more programs and controlling to change media, sound, lighting, etc. periodically or irregularly according to the selected information file and program.

In addition, after step (b), the user terminal capable of communicating with the management server and inputting the user's operation command inputs user information including age, gender, taste, and psychological state from the user, and input (c) of delivering the user information to the management server, and the management server determines the age, gender, taste, and psychological state of the user according to the received user information, and media corresponding to the personalized concept based on the determination result , may further include the step (d) of controlling to provide sound and lighting.

A method of judging a brain state using body information related to a user

According to another embodiment of the present invention, a method of determining a brain state using body information related to a user may be applied.

37 is a flowchart illustrating a method of collecting user-related body information and additionally using it to determine a user's brain state in relation to the present invention.

Referring to FIG. 37 , the brain signal analysis unit 13 of the brain signal measuring unit 10 selects at least one activated brain activation region among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation. The extraction step (S300) is performed.

In this case, before proceeding to step S400, the user data collection unit 20 collects body information related to the user (S360).

In addition, the diagnosis unit 400 determines the user's brain state by using the user-related body information obtained in step S360 together with the at least one brain activation region continuously extracted for a certain period of time (S400) will perform

Hereinafter, specific embodiments of the present invention related to step S360 will be described.

First, FIG. 38 is a flowchart illustrating a method of determining a user's hearing impairment degree and additionally using it to determine a user's brain state in relation to the present invention.

Referring to FIG. 38 , in S400 described above, the diagnosis unit 50 determines the degree of hearing impairment of the user based on the collected auditory information ( S410 ) and the diagnosis unit 50 performs at least one brain activation region. and determining the user's brain state by using the user's degree of hearing impairment together (S420).

For example, the user's hearing loss is due to hearing loss, and the diagnosis unit 50 may collect the user-related body information, the average value of the left and right pure tone thresholds of the user, the degree of hearing loss of the user, and information related to the auditory information collection unit. Additionally, the user's brain state can be determined.

In addition, the user's hearing impairment is due to tinnitus, and the diagnosis unit 50 additionally uses the user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of tinnitus of the user, and the auditory information collection unit-related information. Thus, it is possible to determine the user's brain state.

In addition, in this embodiment, the signal collection operation of the brain signal measuring unit 10 may be performed while the user moves while the user's hearing damage occurs.

In addition, in relation to the present invention, Figure 39 is a gait information collecting unit for collecting the user's gait information, the stress information collecting unit for collecting the user's stress information, the ECG information collecting unit for collecting the user's ECG information, the user It is a flowchart explaining a method of determining a user's brain state by additionally using a sleep information collection unit that collects sleep information of

Referring to FIG. 39 , the user data collection unit 30 collects the user's gait information, the user's stress information, the user's electrocardiogram information, the user's sleep information, and the user's concentration information ( S360 ). .

Thereafter, the diagnosis unit 50 determines, based on the information collected by the user data collection unit, at least one of the user's gait pattern, stress level, ECG change, sleep state, and concentration change (S410). carry out

Thereafter, the diagnosis unit 50 determines the user's brain state by using at least one of the user's gait pattern, stress level, electrocardiogram change, sleep state, and concentration change together with the at least one brain activation region. step (S420) is performed.

In the case of this embodiment, the signal collection operation of the brain signal measuring unit 10 includes a state in which the user's gait changes, a state in which the user's stress changes, a state in which the user's electrocardiogram changes, and the user's In a situation in which at least one of a state in which sleep conditions change and a state in which the user's concentration changes are applied, the operation may be performed while the user moves.

Also, FIG. 40 is a flowchart illustrating a method of determining a user's brain state by additionally using a non-contact emotion change collecting unit based on eye tracking in relation to the present invention.

Referring to FIG. 40 , the user data collection unit 30 collects the user's face landmark masking data based on the user's eye tracking and collects the user's emotional change information in a non-contact manner. (S360) is performed.

Thereafter, the diagnosis unit 50 determines the user's brain state using at least one brain activation region and the user's emotional change information ( S420 ).

41 is a flowchart illustrating a method of determining a user's brain state by additionally using a non-contact emotion change collecting unit based on a voice change in relation to the present invention.

Referring to FIG. 41 , the user data collection unit 30 collects the user's emotion change information based on the user's voice change ( S360 ).

Thereafter, the diagnosis unit 50 determines the user's brain state using at least one brain activation region and the user's emotional change information ( S420 ).

A method of determining a brain state using a medical image related to a user

Meanwhile, according to another embodiment of the present invention, a method for determining a brain state using a medical image related to a user may be provided.

42 is a flowchart illustrating a method of determining a user's brain state by additionally collecting and utilizing medical images related to the user in relation to the present invention.

Referring to FIG. 42 , the brain signal analysis unit 13 of the brain signal measurement unit 10 extracts at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation. to perform the step (S300).

Thereafter, the user data collection unit 30 collects the user-related body information ( S360 ), and before entering the step S400 , the image acquisition unit 40 collects a medical image related to the user. step (S370) is performed.

In step S370, an MRI image acquisition unit for collecting an MRI image related to the user, a CT image acquisition unit for collecting a CT image related to the user, an fMRI image acquisition unit for collecting an fMRI image related to the user, etc. may be utilized. .

Thereafter, the diagnosis unit 50 determines the user's brain state by using at least one brain activation region, the user-related body information, and the user-related medical image together ( S400 ).

How to provide feedback in response to the user's brain state

Also, according to an embodiment of the present invention, a method of providing feedback in response to a user's brain state may be provided.

43 is a flowchart illustrating a method of providing information for improving a user's brain state in response to the user's brain state in relation to the present invention.

Referring to FIG. 43 , the diagnosis unit 50 determines the brain state of the user based on the at least one brain activation region continuously extracted for a predetermined period ( S400 ).

Thereafter, the management unit 60, in response to the user's brain state determined by the diagnosis unit 50, provides information for improving the user's brain state (S500).

In relation to step S500, FIG. 44 is a flowchart illustrating a method of stimulating the brain to improve a user's brain state in relation to the present invention.

Referring to FIG. 44 , by applying a preset criterion, based on the user's brain state, it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability within a certain period of time Step S410 is performed.

In order to improve the user's brain condition under the control of the understanding and management unit 60 , the brain signal stimulation unit 20 may perform the step S510 of stimulating the user's brain.

In addition, if the user's condition corresponds to dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disability, the user's brain health status is checked while recording the current status judgment and tracking changes in the disorder later and manage it.

Effects according to the present invention

As described above, the apparatus for providing brain information based on artificial intelligence according to the present invention may collect brain-related signals while the user is active, determine the user's brain state based on this, and give feedback.

In addition, the present invention irradiates near-infrared rays to the user's brain, a brain signal acquisition unit that detects light that has passed through the cerebral cortex of the brain, and determines the degree of oxygenation of hemoglobin in the user's cerebral bloodstream based on the detected light Continuously with a brain signal measuring unit including a brain signal processing unit and a brain signal analysis unit that extracts at least one activated brain activation region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation and It is possible to provide a system including a diagnosis unit that determines the brain state of the user based on the extracted at least one brain activation region.

In addition, the present invention may provide a system including a brain signal stimulation unit for stimulating a user's brain for a signal collection operation of the brain signal measuring unit.

In addition, the present invention provides an auditory information collection unit that collects the user's auditory information, a gait information collection unit that collects the user's gait information, a stress information collection unit that collects the user's stress information, and the user's electrocardiogram information. Electrocardiogram information collection unit, sleep information collection unit that collects user sleep information, intensive information collection unit that collects user's intensive information, and user's face landmark masking data are collected based on user's eye tracking The user's brain state is further utilized by a non-contact emotion change collecting unit that collects the user's emotional change information in a contact manner, and a non-contact emotional change collecting unit that collects the user's emotional change information based on the user's voice change It is possible to provide a system for precise judgment.

In addition, the present invention additionally uses an image collection unit that collects user-related medical images, and uses an artificial brain activation region, user-related body information, and user-related medical images together to more precisely determine the user's brain state. An intelligence-based brain information providing device may be provided.

In addition, the present invention responds to the user's brain condition determined by the diagnosis unit and the diagnosis unit that can determine the user's dementia, Parkinson's disease, stroke, epilepsy, brain diseases including brain tumors and developmental disorders, brain condition improvement It is possible to provide a system including a management unit that provides information for

In addition, according to the present invention, by applying a preset criterion, based on the user's brain state, it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period of time In this case, in order to improve the user's brain state, a system that further utilizes a brain signal stimulation unit for stimulating the user's brain may be provided.

The present invention can be used as a result of a study on the mechanism of degenerative brain disease. Through brain imaging and cognitive ability measurement experiment design, it is possible to develop and utilize a non-invasive diagnosis and treatment monitoring system, and through brain structure imaging and brain function imaging experiments, the brain It is possible to prepare and share the basis for building an image database, and it becomes possible to derive and propose an optimal treatment technique by identifying the treatment mechanism based on deep learning analysis.

In addition, the present invention can be utilized as an early diagnosis platform for degenerative brain disease, and can be utilized for early diagnosis through the development of a non-invasive dementia diagnosis platform through retina imaging, ophthalmic disease and eye-tracking, etc., and real-time diagnosis without time and space limitations It is possible to improve the quality of life through

In addition, the present invention can be used as an integrated customized medical device for phototherapy and diagnosis of degenerative brain disease, and it is possible to implement a prototype for commercialization of a treatment platform for degenerative brain disease, and based on the research results, it is possible to commercialize medical device products by collaborating with companies In addition, it is possible to maximize the utilization of therapeutic medical devices by linking with local dementia centers under the 'National Dementia Responsibility System'.

Furthermore, according to the present invention, it is possible to prepare standard guidelines for each field of data construction for AI learning, to secure high-quality AI data sets through step-by-step quality verification, to provide basic data essential for AI technology development, and to follow the development of AI technology in the private sector. It is possible to create a virtuous cycle ecosystem that expands and opens data autonomously.

In addition, according to the present invention, it is possible to lead the field of brain science by identifying the relationship between degenerative brain disease and treatment for hearing loss, and to secure competitiveness in the field of advanced medical image analysis through deep learning techniques for analyzing brain image data.

In addition, according to the present invention, it is possible to secure new and original technology for an effective non-invasive method for early diagnosis of degenerative brain disease, and development of a non-invasive brain disease diagnosis and treatment progress monitoring system based on brain imaging and cognitive ability experiments is possible. possible.

In addition, according to the present invention, it is possible to contribute to solving related social problems through the effects of reducing dementia disease-related mortality, increasing the treatment rate, and reducing treatment costs, and contributing to productivity improvement due to the increase in social participation of the elderly.

In addition, according to the present invention, in the age of low fertility, if you have a brain disease or a brain disease symptom is suspected due to an increase in brain diseases such as epilepsy, stroke, brain tumor, and developmental disorders in infants and children, brain health status through prevention and diagnosis can be verified.

In addition, according to the present invention, in measuring the brain signal path, the wearable type and child-friendly design product are applied, so that the experience value of the user is increased, sleep and anesthetics are not used, which can help healthy brain development. It works.

In addition, according to the present invention, it is possible to provide an emotional sympathy AI service, which is a human-centered artificial intelligence service using a multi-modal emotional data network construction and N-dimensional emotional mapping space to support non-face-to-face learning and emotional labor fields.

Specifically, according to the present invention, the user's social emotions are recognized, the emotional data network is built and the personal emotional AI service is provided, and this can be used for non-face-to-face learning and industrial field support.

On the other hand, the effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be able

The above-described embodiments of the present invention may be implemented through various means. For example, embodiments of the present invention may be implemented by hardware, firmware, software, or a combination thereof.

In case of implementation by hardware, the method according to embodiments of the present invention may include one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs). , FPGAs (Field Programmable Gate Arrays), processors, controllers, microcontrollers, microprocessors, and the like.

In the case of implementation by firmware or software, the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above. The software code may be stored in the memory unit and driven by the processor. The memory unit may be located inside or outside the processor, and may transmit and receive data to and from the processor by various known means.

The detailed description of the preferred embodiments of the present invention disclosed as described above is provided to enable any person skilled in the art to make and practice the present invention. Although the above has been described with reference to preferred embodiments of the present invention, it will be understood by those skilled in the art that various modifications and changes can be made to the present invention without departing from the scope of the present invention. For example, those skilled in the art can use each configuration described in the above-described embodiments in a way in combination with each other. Accordingly, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The present invention may be embodied in other specific forms without departing from the spirit and essential characteristics of the present invention. Accordingly, the above detailed description should not be construed as restrictive in all respects but as exemplary. The scope of the present invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present invention are included in the scope of the present invention. The present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. In addition, claims that are not explicitly cited in the claims may be combined to form an embodiment, or may be included as a new claim by amendment after filing.

Claims (30)

a first step of irradiating near-infrared rays to the user's brain by the brain signal acquisition unit of the brain signal measuring unit and detecting light that has passed through the cerebral cortex of the brain;

a second step of determining, by the brain signal processing unit of the brain signal measuring unit, the degree of oxygenation of hemoglobin in the user's cerebral blood flow based on the detected light;

a third step of extracting, by the brain signal analysis unit of the brain signal measurement unit, at least one activated brain region from among a plurality of regions of the user's brain based on the determined degree of hemoglobin oxygenation; and

A fourth step of determining, by a diagnostic unit, the brain state of the user based on the at least one brain activation region continuously extracted for a certain period of time;

The signal collection operation of the brain signal acquisition unit of the first step is performed while the user is moving,

Between the third step and the fourth step,
Steps 3-5 of collecting body information related to the user by the user data collection unit; further comprising,

In steps 3-5,
The user data collection unit,
An auditory information collecting unit for collecting the user's auditory information, a gait information collecting unit for collecting the user's gait information, a stress information collecting unit for collecting the user's stress information, and an electrocardiogram for collecting the user's electrocardiogram information Comprising at least one of an information collection unit, a sleep information collection unit for collecting the user's sleep information, and an intensive information collection unit for collecting the user's focused information,

In the fourth step,
The diagnostic unit,
Based on the collected auditory information, determining at least one of the user's hearing impairment degree, the user's gait pattern, stress level, ECG change, sleep state, and concentration change,
Artificial, characterized in that the user's brain state is determined by using at least one of the user's hearing impairment degree, gait pattern, stress level, electrocardiogram change, sleep state, and concentration change and the at least one brain activation region together. A method of providing intelligence-based brain information.
The method of claim 1,
In the second step,
Based on the detected light, by extracting the oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) in the user's cerebral blood flow, the degree of hemoglobin oxygenation is determined. method. 3. The method of claim 2,
In the at least one brain activation region,
Oxygen transported by oxidized hemoglobin in the cerebral bloodstream is consumed, so that the oxidized hemoglobin concentration (Oxy Hb) decreases, and the reduced hemoglobin concentration (Deoxy Hb) increases in response to the decrease in the oxidized hemoglobin concentration (Oxy Hb) Artificial intelligence-based brain information providing method, characterized in that. 4. The method of claim 3,
The oxidized hemoglobin concentration (Oxy Hb) and the reduced hemoglobin concentration (Deoxy Hb) have optical properties that change in a visible light region and a near-infrared light region,
In the first step, the brain signal measuring unit, through the near-infrared spectroscopy (fNIRS: functional near-infrared spectroscopy) by the near-infrared irradiation, artificial intelligence-based brain information providing method, characterized in that to collect the signal. 5. The method of claim 4,
In the third step, the brain signal analysis unit,
dividing the user's brain region into a plurality,
AI-based brain information providing method, characterized in that it is determined whether each of the plurality of divided brain regions is changed to the brain active region or changed from the brain active region to an inactive region by a predetermined time unit. 6. The method of claim 5,
In the fourth step, the diagnosis unit,
Based on the predetermined time unit,
An artificial intelligence-based brain information providing method, characterized in that the user's brain state is determined by using changes in brain activation of each of the plurality of divided brain regions. 7. The method of claim 6,
Between the third step and the fourth step,
displaying, by the display unit of the brain signal measuring unit, changes in brain activation of each of the plurality of divided brain regions over time; Artificial intelligence-based brain information providing method, characterized in that it further comprises. 8. The method of claim 7,
In the first step, the brain signal measuring unit,
An artificial intelligence-based brain information providing method, characterized in that it is in the form of a full cover covering all of the user's head or in the form of a front part in close contact with the front part of the user's head. The method of claim 1,
Prior to the first step,
a 0.5 step of stimulating the brain of the user by the brain signal stimulation unit for the signal collection operation of the brain signal measuring unit; Artificial intelligence-based brain information providing method, characterized in that it further comprises. 10. The method of claim 9,
The 0.5 step is,
(a) step of applying multi-dimensional environmental stimuli so that the brain activation device of the brain signal stimulator intensively strengthens the activity of a pre-targeted brain region within a predetermined space and activates the function of the brain region; and
(b) controlling, by the management server of the brain signal stimulation unit, the driving of the brain activation device to apply a multi-dimensional environmental stimulus to the targeted brain region in the space; based brain information provision method. 11. The method of claim 10,
The space is divided into a plurality of spaces,
The step (a) is,
implementing media art corresponding to the concept to be produced by the media unit in each of the plurality of spaces;
outputting sound by the sound unit;
illuminating or illuminating the light; and
A fragrance delivery unit delivering a fragrance to the user; Artificial intelligence-based brain information providing method comprising a. 12. The method of claim 11,
In step (b),
management server,
A plurality of media information and sound information corresponding to the concept of each of the plurality of spaces, and a database in which a program for controlling the operation of the brain activation device is stored,
Selecting one or more of a plurality of media information, sound files and programs corresponding to the concept of each space from among various information and programs stored in the database, and controlling the driving of the brain activation device to change according to the selected information and program A method for providing brain information based on artificial intelligence. 12. The method of claim 11,
The media unit includes a plurality of media modules that implement media art corresponding to the concept of each space,
Each of the media modules includes a projector or a display panel for projecting media art including pictures or moving pictures,
The sound unit includes a plurality of sound modules for outputting sound corresponding to the concept of each space,
Each sound module comprises a speaker and an amplifier for outputting sound including music or natural white noise, an artificial intelligence-based brain information providing method. 12. The method of claim 11,
The lighting unit includes a plurality of lighting modules that irradiate or illuminate light corresponding to the concept of each space,
Each lighting module is composed of one or a combination of two or more of a laser generator that irradiates a laser, an LED that irradiates light, and a neon light,
The fragrance delivery unit includes a plurality of delivery modules that deliver the fragrance of natural objects arranged in each space to the user,
Each delivery module is an artificial intelligence-based brain information providing method, characterized in that it includes a blower that blows air to deliver a fragrance generated from a natural object disposed in each space or a fragrance that mimics the fragrance of a natural object to a user inside each space . 12. The method of claim 11,
After step (b),
A manager terminal that controls the operation of the brain activation device based on the manager's operation,
Select one or more of a plurality of information files and programs stored in the database of the management server for each module installed in each space, and periodically or irregularly change media, sound, lighting, etc. according to the selected information file and program (c) controlling the artificial intelligence-based brain information providing method, characterized in that it further comprises. 12. The method of claim 11,
After step (b),
A user terminal capable of communicating with the management server and inputting a manipulation command of the user inputs user information including age, gender, taste, and psychological state from the user, and transmits the input user information to the management server (c) to do; and
(d) controlling the management server to determine the user's age, gender, taste, and psychological state according to the received user information, and to provide media, sound, and lighting corresponding to the personalized concept based on the determination result; Artificial intelligence-based brain information providing method, characterized in that it further comprises. delete The method of claim 1,
In step 3-5, the auditory information collection unit of the user data collection unit collects the user's auditory information,
In the fourth step, the diagnosis unit determines the degree of hearing impairment of the user based on the collected auditory information,
In the fourth step, the diagnosis unit determines the brain state of the user by using the at least one brain activation region and the degree of hearing impairment of the user together.
19. The method of claim 18,
The user's hearing loss is due to hearing loss,
wherein the diagnosis unit additionally uses the user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of hearing loss of the user, and the auditory information collection unit-related information to determine the brain state of the user How to provide brain information. 19. The method of claim 18,
The user's hearing impairment is due to tinnitus,
The diagnosis unit is based on artificial intelligence, characterized in that the user's brain state is determined by additionally using the user-related body information, the average value of the left and right pure tone threshold values of the user, the degree of tinnitus of the user, and the auditory information collection unit-related information. How to provide brain information. 19. The method of claim 18,
In the first step, the signal collection operation of the brain signal measuring unit is performed while the user is moving in a state in which the user's hearing loss has occurred. delete The method of claim 1,
The signal collection operation of the brain signal measuring unit includes a state in which the gait of the user changes, a state in which the user's stress changes, a state in which the user's electrocardiogram changes, a state in which the user's sleep condition changes, and the Artificial intelligence-based brain information providing method, characterized in that it is performed while the user moves in a situation where at least one of the states in which the user's concentration changes is applied. The method of claim 1,
In step 3-5, the user data collection unit collects the user's face landmark masking data based on the user's eye tracking and collects the user's emotional change information in a non-contact manner Emotion change collection unit; further comprising,
In the fourth step, the diagnosis unit determines the brain state of the user by using the at least one brain activation region and the emotion change information of the user together.
The method of claim 1,
In step 3-5, the user data collection unit, a non-contact emotion change collection unit for collecting the emotion change information of the user based on the change in the user's voice; further comprising,
In the fourth step, the diagnosis unit determines the brain state of the user by using the at least one brain activation region and the emotion change information of the user together.
The method of claim 1,
Between step 3.5 and step 4, a step 3.7 of collecting, by an image collecting unit, a medical image related to the user; further comprising,
In the fourth step, the diagnosis unit determines the brain state of the user by using the at least one brain activation region, the user-related body information, and the user-related medical image together. How to provide brain information.
27. The method of claim 26,
The image collection unit of step 3.7,
an MRI image acquisition unit configured to collect MRI images related to the user; a CT image acquisition unit for collecting CT images related to the user; and an fMRI image acquisition unit that collects fMRI images related to the user. The method of claim 1,
In the fourth step, the diagnosis unit determines the user's brain disease,
The brain disease, dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and artificial intelligence-based brain information providing method, characterized in that it includes a developmental disorder. 29. The method of claim 28,
After the fourth step,
A fifth step of providing, by the management unit, information for improving the user's brain condition in response to the user's brain condition determined by the diagnosis unit; 30. The method of claim 29,
When it is predicted that the user will correspond to at least one of dementia, Parkinson's disease, stroke, epilepsy, brain tumor, and developmental disorder within a certain period based on the user's brain state by applying a preset criterion,
The fifth step is
In order to improve the user's brain condition under the control of the management unit, the artificial intelligence-based brain information providing method further comprising the step of stimulating the user's brain by a brain signal stimulation unit.

Images