CN115177209A - Wearable minimally invasive closed-loop regulation and control device for infant epilepsy - Google Patents
Wearable minimally invasive closed-loop regulation and control device for infant epilepsy Download PDFInfo
- Publication number
- CN115177209A CN115177209A CN202210730060.3A CN202210730060A CN115177209A CN 115177209 A CN115177209 A CN 115177209A CN 202210730060 A CN202210730060 A CN 202210730060A CN 115177209 A CN115177209 A CN 115177209A
- Authority
- CN
- China
- Prior art keywords
- infant
- closed
- prediction
- epilepsy
- loop regulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 206010015037 epilepsy Diseases 0.000 title claims abstract description 138
- 230000033228 biological regulation Effects 0.000 title claims abstract description 76
- 206010010904 Convulsion Diseases 0.000 claims abstract description 88
- 208000028329 epileptic seizure Diseases 0.000 claims abstract description 67
- 238000000034 method Methods 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 33
- 210000004556 brain Anatomy 0.000 claims abstract description 22
- 238000004891 communication Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims description 23
- 238000013527 convolutional neural network Methods 0.000 claims description 21
- 238000013135 deep learning Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 13
- 230000000542 thalamic effect Effects 0.000 claims description 11
- 210000001186 vagus nerve Anatomy 0.000 claims description 11
- 210000004761 scalp Anatomy 0.000 claims description 10
- 238000010172 mouse model Methods 0.000 claims description 9
- 230000001037 epileptic effect Effects 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 210000003109 clavicle Anatomy 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000000638 stimulation Effects 0.000 description 37
- 238000010586 diagram Methods 0.000 description 23
- 210000005036 nerve Anatomy 0.000 description 15
- 230000006870 function Effects 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 6
- 210000003625 skull Anatomy 0.000 description 6
- 230000003321 amplification Effects 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 5
- 229940079593 drug Drugs 0.000 description 5
- 239000003814 drug Substances 0.000 description 5
- 238000003199 nucleic acid amplification method Methods 0.000 description 5
- 230000004936 stimulating effect Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 4
- 238000001356 surgical procedure Methods 0.000 description 4
- 238000013526 transfer learning Methods 0.000 description 4
- 238000007599 discharging Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 3
- 238000012805 post-processing Methods 0.000 description 3
- 206010021750 Infantile Spasms Diseases 0.000 description 2
- 238000010171 animal model Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 210000003169 central nervous system Anatomy 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000016507 interphase Effects 0.000 description 2
- 238000002324 minimally invasive surgery Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000031877 prophase Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010001497 Agitation Diseases 0.000 description 1
- 206010003805 Autism Diseases 0.000 description 1
- 208000020706 Autistic disease Diseases 0.000 description 1
- 208000001654 Drug Resistant Epilepsy Diseases 0.000 description 1
- 206010064571 Gene mutation Diseases 0.000 description 1
- 201000006347 Intellectual Disability Diseases 0.000 description 1
- 241000124008 Mammalia Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 208000029726 Neurodevelopmental disease Diseases 0.000 description 1
- 241000288906 Primates Species 0.000 description 1
- 208000006289 Rett Syndrome Diseases 0.000 description 1
- 208000005392 Spasm Diseases 0.000 description 1
- 201000006791 West syndrome Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000004641 brain development Effects 0.000 description 1
- 230000005978 brain dysfunction Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 208000010877 cognitive disease Diseases 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000002920 convulsive effect Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000006735 deficit Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 208000013257 developmental and epileptic encephalopathy Diseases 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 230000007786 learning performance Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009456 molecular mechanism Effects 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000001314 paroxysmal effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000005424 photoluminescence Methods 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 208000005809 status epilepticus Diseases 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4094—Diagnosing or monitoring seizure diseases, e.g. epilepsy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0004—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by the type of physiological signal transmitted
- A61B5/0006—ECG or EEG signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/25—Bioelectric electrodes therefor
- A61B5/251—Means for maintaining electrode contact with the body
- A61B5/256—Wearable electrodes, e.g. having straps or bands
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/30—Input circuits therefor
- A61B5/307—Input circuits therefor specially adapted for particular uses
- A61B5/31—Input circuits therefor specially adapted for particular uses for electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N1/00—Electrotherapy; Circuits therefor
- A61N1/02—Details
- A61N1/04—Electrodes
- A61N1/05—Electrodes for implantation or insertion into the body, e.g. heart electrode
- A61N1/0526—Head electrodes
- A61N1/0529—Electrodes for brain stimulation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N1/00—Electrotherapy; Circuits therefor
- A61N1/18—Applying electric currents by contact electrodes
- A61N1/32—Applying electric currents by contact electrodes alternating or intermittent currents
- A61N1/36—Applying electric currents by contact electrodes alternating or intermittent currents for stimulation
- A61N1/3605—Implantable neurostimulators for stimulating central or peripheral nerve system
- A61N1/3606—Implantable neurostimulators for stimulating central or peripheral nerve system adapted for a particular treatment
- A61N1/36064—Epilepsy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N1/00—Electrotherapy; Circuits therefor
- A61N1/18—Applying electric currents by contact electrodes
- A61N1/32—Applying electric currents by contact electrodes alternating or intermittent currents
- A61N1/36—Applying electric currents by contact electrodes alternating or intermittent currents for stimulation
- A61N1/3605—Implantable neurostimulators for stimulating central or peripheral nerve system
- A61N1/36128—Control systems
- A61N1/36135—Control systems using physiological parameters
- A61N1/36139—Control systems using physiological parameters with automatic adjustment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Neurosurgery (AREA)
- Neurology (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Signal Processing (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- General Engineering & Computer Science (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention relates to a wearable minimally invasive closed-loop regulation and control device for infant epilepsy, which comprises an acquisition processing module, a closed-loop regulation and control module, a discharge module, a cloud server and a mobile phone APP, wherein the acquisition processing module is connected with the closed-loop regulation and control module; the closed-loop regulation and control module predicts whether the infant has the epileptic seizure by adopting a basic epileptic seizure prediction algorithm to obtain a prediction result I, and when the prediction result I is the epileptic seizure, a control signal I is formed; under the condition that the cloud server is communicated with the closed-loop regulation and control module, the closed-loop regulation and control module sends a control signal II, and the discharge module receives the control signal II sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant; under the condition that the cloud server and the closed-loop regulation and control module interrupt communication, the closed-loop regulation and control module sends a control signal I, and the discharge module receives the control signal I sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant. According to the method, the prediction model of the infant epileptic seizure is established, so that the prediction of 30 minutes before the epileptic seizure is realized.
Description
Technical Field
The invention belongs to the field of devices, and relates to a wearable minimally invasive closed-loop regulation and control device for infant epilepsy.
Background
Infant epilepsy is a common nervous system syndrome with complex etiology and repeated attacks in the infant period, is a convulsive attack caused by paroxysmal and temporary brain dysfunction, and seriously affects the daily life, brain development, physical and mental health and the like of infants.
The existing clinical common infant epilepsy treatment methods comprise methods such as drug treatment, surgical treatment, nerve electrical stimulation and the like, and although the methods have certain clinical curative effects, the methods have the defects that the methods cannot overcome: (1) and (3) drug treatment: can be accompanied with the impairment of the central nervous system function of patients, and causes cognitive dysfunction in nearly 90% of patients with symptom disappearance; (2) and (3) surgical treatment: the high risk of the surgery itself has the problems of easy infection of the wound, difficult recovery after the surgery, and the like; (3) nerve electrical stimulation: at present, the open-loop electrical stimulation mode is commonly used for stimulating vagus nerve or anterior thalamic nucleus in clinic to treat epilepsy, and the real reason of the lesion is not considered. On the basis of the above-mentioned therapy, the concept of applying indirect electrical stimulation, i.e. closed-loop electrical stimulation, in accordance with the change in status epilepticus has been proposed. The only closed-loop stimulation systems currently used clinically worldwide are the american NeuroPace Reactive Neurostimulation System (RNS). The Deep Brain Stimulation (DBS) system developed by modern company in the most mature product in the relevant field of nerve electrical Stimulation in China achieves the international advanced level in technology and process, and is certified by CE in 2016, so that more than 1.2 ten thousand cases of products are implanted clinically. However, the deep brain electrical stimulation system still belongs to the open-loop electrical stimulation regulation type, and in China at the present stage, the field of closed-loop stimulation system products is still in a blank state.
Neurochip and Neurochip2 designed in Fetz professor laboratories of the university of Washington, USA can realize 3 stimulating channels and 3 recording channels, and preliminarily realize functional complex nerve regulation and control experiments by combining infrared data transmission and a simple programmable system [1,2] 。
Professor Feng Zhouyan, university of Zhejiang, in 2012, a closed-loop stimulation system was developed to collect and stimulate nerve signals through a data acquisition card, but the flexibility of the system is limited to a certain extent by using a computer as a processing core [3] 。
University of china Yang Xiaofei teaches a wireless neural closed-loop stimulation system developed in 2013. The system mainly comprises a wireless acquisition module, a wireless stimulation module and Labview software, but the whole set of equipment only has 1 recording channel and 1 stimulation channel, and only can complete limited closed-loop experimental detection [4] 。
In 2014, taiwan university of transportation Wei-MingChen et al designed a fully integrated closed-loop stimulation system with 8-channel recording and l-channel stimulation combined. The whole set of equipment is manufactured by adopting a 0.18-micron CMOS process, the area is only 13.47mm, and the power consumption is about 2.8mW. However, the epilepsy prediction algorithm of the system is single and cannot be clinically applied [5] 。
Toronto-university Salam et al proposed a high-throughput closed-loop stimulation system in 2015. The system mainly comprises an integrated front-end chip and a field programmable gate array, although the system has extremely high performance, the non-commercial chip scheme has extremely high cost and long development period, and is not suitable for popularization and application of preclinical animal models and clinical patients [6] 。
In 2016, shoaran et al, the federal institute of technology, switzerland, proposed a closed-loop stimulation system specifically for the treatment of epilepsy. The system has 16 recording channels, 1 constant-current stimulation channel and an integrated epilepsy detection processing module, but the prediction algorithm of the system uses a small data set, and the true clinical effectiveness of the system needs to be evaluated [7] 。
The American Beller College of Medicine (Baylor College of Medicine), dr. The research work of neurodevelopmental diseases such as Rett syndrome, autism and the like is carried out for a long time, and the diseases caused by gene mutation are mostly developed in infancy and are accompanied by typical clinical epilepsy characteristics.
The subject group of Dr.Jeffrey L.Noebes, baylor College of Medicine, USA. The subject group mainly follows how genes regulate the excitability of central nervous system neurons and neural networks of mammals, and establishes various epilepsia mouse models.
The central innovation of brain science and intelligent technology of Chinese academy of sciences, xiong Zhiji advanced researchers topic group. The subject group has long been engaged in the molecular mechanism of epilepsia occurrence and the establishment of a disuse primate epilepsia model, and particularly focuses on the research of children's intellectual development disorder diseases and epilepsia.
The Dr.Jingxing topic group of the Xinxinatii Children hospital establishes a plurality of related algorithms facing electroencephalogram and magnetoencephalogram for long-term pursuing of seizure mechanism and origin focus positioning of intractable epilepsy of children.
(1) Clinically, the existing nerve closed-loop control diagnosis and treatment equipment faces adults, and is fixed on the skull by screws, but the skull of infants is relatively thin and soft and cannot be installed in the same way, so that the existing nerve closed-loop control diagnosis and treatment equipment for infants is in a blank state;
(2) The current nerve electrical stimulation closed-loop regulation and control system clinically adopts a single target point stimulation mode for regulation and control, but the regulation and control effect and the epilepsy inhibition effect are not ideal, because the epileptic seizure starts from a primary focus and is transmitted through a nerve loop, and a plurality of target points can be involved.
Therefore, the multi-target combined electrical stimulation is based on the regulation of the function of the neural circuit, and has larger treatment potential than single-target stimulation;
(3) The prediction of epileptic seizure is an important link for realizing closed-loop regulation and control of electrical stimulation, most of the existing prediction algorithms are based on the traditional machine learning method, the analyzed data volume is small, the data set is small, the used feature extraction methods are all various traditional features oriented to manual design, the features of the epileptic brain waveform in the time domain and the space domain can not be accurately described for a long time and dynamically, and the prediction accuracy is low and generally does not exceed 70%.
In summary, only the adult RNS system is available at present, but because the skull thickness of the infant group is thin, the adult RNS system is required to drill a hole in the skull and fix a related hardware device; the skull of the infant is thin and not grown yet, and cannot be drilled and fixed, so that the popularization of the system in the infant group is limited.
Disclosure of Invention
In order to solve the above problems, the present invention provides the following technical solutions: a wearable minimally invasive closed-loop regulation and control device for infant epilepsy comprises an acquisition processing module, a closed-loop regulation and control module, a discharge module, a cloud server and a mobile phone APP;
the collecting and processing module collects and processes scalp electroencephalogram signals and transmits the processed electroencephalogram signals to the closed-loop regulation and control module;
the closed-loop regulation and control module predicts whether the infant has the epileptic seizure by adopting a basic epileptic seizure prediction algorithm to obtain a prediction result I, and when the prediction result I is the epileptic seizure, a control signal I is formed;
the mobile phone APP receives the processed electroencephalogram signals transmitted by the closed-loop regulation and control module;
the cloud server receives the processed brain electrical signals transmitted by the mobile phone APP, adopts a deep learning prediction method facing infant epilepsy and carries out deep prediction on the infant epileptic seizure to obtain a prediction result II, and when the prediction result II is the infant epileptic seizure, a control signal II is formed and transmitted to the closed-loop regulation and control module through the mobile phone APP;
when the cloud server is communicated with the closed-loop regulation and control module, the closed-loop regulation and control module sends a control signal II, and the discharge module receives the control signal II sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant;
when the cloud server and the closed-loop regulation and control module are in communication interruption, the closed-loop regulation and control module sends a control signal I, and the discharge module receives the control signal I sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant.
Further: the deep learning prediction method for infant epilepsy and the deep prediction of infant epileptic seizure comprise the following steps:
constructing a data set of infant seizure characteristics, wherein the data set is divided into a training set and a testing set;
preprocessing the data set to obtain a preprocessed data set;
constructing an infant epilepsia convolutional neural network model for deeply predicting infant epilepsia attacks;
training the infant epilepsy convolutional neural network model based on training set data to obtain a trained infant epilepsy convolutional neural network model;
and inputting the test set data into the trained infant epilepsy convolutional neural network model to realize the prediction of infant epilepsy attack and obtain a prediction result.
Further, the method comprises the following steps: the process of constructing the infant epilepsy convolutional neural network model for deeply predicting infant epilepsy attack is as follows:
through electroencephalogram data of early epileptic seizure and interphase epileptic seizure of infants collected by a mouse model, learning priori knowledge of the electroencephalogram data of the early epileptic seizure in an off-line training mode, and training to obtain a DNN basic model;
the DNN basic model is trained by using a small amount of electroencephalogram data of infant epilepsy, and is migrated to a task of predicting infant epileptic seizure in a priori knowledge training mode to obtain a trained infant epilepsy convolutional neural network model.
Further: the discharge module comprises a first electrode and a second electrode; the first electrode is disposed at the anterior thalamic nucleus and the second electrode is disposed at the vagus nerve.
Further: the acquisition processing module adopts a wearable 32-channel scalp electroencephalogram acquisition cap.
Further: the prediction result is based on the method for performing smooth correction on the self prediction label by the neighbor label, and the prediction label of the current sample is updated according to the prediction label conditions of the neighbors on the two sides for the current electroencephalogram data sample by setting a cache region for storing the prediction categories of n neighbor samples on the adjacent left side and the adjacent right side.
Further, the method comprises the following steps: the process of predicting whether the infant has the epileptic seizure by the closed-loop regulation and control module by adopting a basic epileptic seizure prediction algorithm is as follows:
and combining the six characteristics of the average value, the extreme value, the standard deviation, the normalized first-order difference, the energy and the power of the time domain signal of the electroencephalogram signal, and then carrying out threshold analysis to predict the epileptic attack.
Further: the closed loop regulation and control module adopts a chip model number XEM6010-LX45, and the XEM6010-LX45 is installed at a clavicle of a wearer.
The wearable minimally invasive closed-loop regulation and control device for infant epilepsy, provided by the invention, has the most core problem of how to effectively integrate the functions of perception, stimulation and closed-loop control and meet the requirements of mobility and expansibility of a system; the device is a wearable, high-accuracy and low-power consumption reactive nerve electrical stimulation closed-loop regulation and control system for infant epilepsy; the device is convenient for carry, dismantle convenient, the wearing formula nerve regulation and control system of flexibility and practicality strong, and this system utilizes implanted electrode to acquire local field potential signal and output electro photoluminescence, predicts in advance to the epileptic seizure of baby's spasm based on degree of depth study to form the regulation and control of closed loop, and wear the relevant embedded chip in patient's waist, chest or near arbitrary health with the elasticity band of adjustable length with the detachable mode.
In order to reduce the running power consumption of the embedded chip, the epileptic seizure prediction algorithm based on deep learning is put in a cloud server, and the wearable embedded chip is responsible for two functions: running a basic epileptic seizure prediction algorithm;
transmitting the acquired electroencephalogram signals to an acquisition APP of the mobile phone through Bluetooth; the mobile phone APP uploads the received electroencephalogram data to the cloud server, the epilepsy prediction algorithm is operated in the cloud server, the prediction result is fed back to the chip, and the chip determines whether to perform electrical stimulation according to the prediction result, so that closed-loop regulation and control are formed.
The stimulation points are set to be the thalamic nucleus and the vagus nerve by the discharging module, and the discharging module is only required to be placed on the shoulders of a user and fixed by a belt without drilling holes on the skull.
This device is towards infant's epilepsy, and diagnosis and treatment equipment that independent research and development response nerve electrostimulation closed loop was regulated and control has multiple efficiency such as miniaturized, low-power consumption, wearable, data acquisition, epilepsy early warning, regulation and control treatment concurrently, and applicable diagnosis and treatment in clinical infant's epilepsy. The chip is an FPGA chip with low power consumption, and a complex prediction algorithm is executed in the cloud, so that only a basic prediction algorithm can be operated on the chip. Overall operating power <2 watts;
on the basis of hardware of a multi-channel closed-loop stimulation system, closed-loop nerve regulation and control are carried out on infant epilepsy by adopting a multi-target point electrical stimulation method combining anterior thalamic nucleus and vagus nerve.
According to the method, the data set of the infant epilepsy is established, and the prediction model of the infant epilepsy is established based on the deep learning method, so that the prediction 30 minutes before the epilepsy is realized, the accuracy rate is more than 90%, and the false detection rate is less than 1 time/hour.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a block schematic diagram of the apparatus of the present application;
FIG. 2 is a pictorial representation of the present application;
FIG. 3 is an overall block diagram of the apparatus of the present application;
FIG. 4 is a block schematic diagram of an acquisition processing module of the present application;
FIG. 5 (a) is a circuit diagram of an analog-to-digital conversion sub-module of the acquisition processing module; (b) A control function sub-module circuit diagram of the acquisition processing module; (c) a TF card module circuit diagram of the acquisition processing module; (d) is a circuit diagram of a wiring submodule of the acquisition processing module; (e) a power supply module circuit diagram of the acquisition processing module; (f) is a voltage-stabilizing sub-module circuit diagram of the acquisition processing module; (g) A front-end amplification sub-module circuit diagram of the acquisition processing module;
FIG. 6 is a circuit schematic of a discharge module;
FIG. 7 is a functional schematic of a closed loop regulatory module;
FIG. 8 is a block diagram of the strategy for transfer learning of the present application;
FIG. 9 is a diagram of a convolutional neural network model for infantile epilepsy;
fig. 10 is a flowchart of the seizure prediction algorithm based on deep learning according to the present application.
Detailed Description
It should be noted that, in the case of conflict, the embodiments and features of the embodiments of the present invention may be combined with each other, and the present invention will be described in detail with reference to the accompanying drawings and embodiments.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise. Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description. Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate. Any specific values in all examples shown and discussed herein are to be construed as exemplary only and not as limiting. Thus, other examples of the exemplary embodiments may have different values. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
In the description of the present invention, it is to be understood that the orientation or positional relationship indicated by the directional terms such as "front, rear, upper, lower, left, right", "lateral, vertical, horizontal" and "top, bottom", etc., are generally based on the orientation or positional relationship shown in the drawings, and are used for convenience of description and simplicity of description only, and in the absence of any contrary indication, these directional terms are not intended to indicate and imply that the device or element so referred to must have a particular orientation or be constructed and operated in a particular orientation, and therefore should not be considered as limiting the scope of the present invention: the terms "inner and outer" refer to the inner and outer relative to the profile of the respective component itself.
For ease of description, spatially relative terms such as "over … …", "over … …", "over … …", "over", etc. may be used herein to describe the spatial positional relationship of one device or feature to another device or feature as shown in the figures. It will be understood that the spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if a device in the figures is turned over, devices described as "above" or "on" other devices or configurations would then be oriented "below" or "under" the other devices or configurations. Thus, the exemplary term "above … …" may include both orientations of "above … …" and "below … …". The device may be otherwise variously oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein interpreted accordingly.
It should be noted that the terms "first", "second", and the like are used to define the components, and are only used for convenience of distinguishing the corresponding components, and the terms have no special meanings unless otherwise stated, and therefore, the scope of the present invention should not be construed as being limited.
FIG. 1 is a block schematic diagram of the apparatus of the present application;
FIG. 2 is a pictorial representation of the present application;
FIG. 3 is an overall block diagram of the apparatus of the present application;
a wearable minimally invasive closed-loop regulation and control device for infant epilepsy comprises an acquisition processing module, a closed-loop regulation and control module, a discharge module, a cloud server, a mobile phone APP,
The collecting and processing module collects and processes scalp electroencephalogram signals and transmits the processed electroencephalogram signals to the closed-loop regulation and control module;
the closed-loop regulation and control module predicts whether the infant has the epileptic seizure by adopting a basic epileptic seizure prediction algorithm to obtain a prediction result I, and when the prediction result I is the epileptic seizure, a control signal I is formed;
the mobile phone APP receives the processed brain electrical signals transmitted by the closed loop regulation and control module;
the cloud server receives the processed electroencephalogram signals transmitted by the mobile phone APP, adopts a depth prediction method facing the infant epilepsy and carries out depth prediction on the infant epilepsy attack to obtain a prediction result II,
when the prediction result II is epileptic seizure, forming a control signal II, and transmitting the control signal II to the closed-loop regulation and control module through a mobile phone APP;
when the cloud server is communicated with the closed-loop regulation and control module, the closed-loop regulation and control module sends a control signal II, and the discharge module receives the control signal II sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant;
when the cloud server and the closed-loop regulation and control module are in communication interruption, the closed-loop regulation and control module sends a control signal I, and the discharge module receives the control signal I sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant.
Furthermore, the acquisition processing module adopts a wearable 32-channel scalp electroencephalogram acquisition cap and a 32-channel scalp electroencephalogram acquisition cap to acquire electroencephalogram signals of the infant;
FIG. 4 is a block schematic diagram of an acquisition processing module of the present application;
the acquisition processing module comprises an analog-to-digital conversion sub-module, a control function sub-module, a TF (TransFlash) clamp module, a wiring sub-module, a power supply sub-module, a voltage stabilizing sub-module and a front-end amplification sub-module;
the output end of the wiring submodule is connected with the input end of the front-end amplification submodule; the output end of the front-end amplification sub-module is connected with the input end of the analog-to-digital conversion sub-module, the output end of the analog-to-digital conversion sub-module is connected with the input end of the control function sub-module, the output end of the control function sub-module is connected with the TF clip module, the output end of the voltage stabilization sub-module is connected with the input end of the power supply sub-module, and the output end of the power supply sub-module is connected with the control function sub-module;
FIG. 5 (a) is a circuit diagram of an analog-to-digital conversion sub-module of the acquisition processing module; (b) A control function sub-module circuit diagram of the acquisition processing module; (c) is a circuit diagram of a TF card clip module of the acquisition processing module; (d) is a circuit diagram of a wiring submodule of the acquisition processing module; (e) a power supply module circuit diagram of the acquisition processing module; (f) is a voltage-stabilizing sub-module circuit diagram of the acquisition processing module; (g) A front-end amplification sub-module circuit diagram of the acquisition processing module;
the wearable scalp electroencephalogram cap developed by using the ADS1299 mainly comprises the following functions: the front-end collecting scalp brain electrical signals, amplifying brain electrical signals, converting analog/digital signals, an ARDUINO main control chip and Bluetooth wireless transmission.
The 32-channel electroencephalogram cap is worn on the head of a user to realize the function of collecting the electroencephalogram of the scalp, and is wirelessly transmitted to the closed-loop regulation and control module through the Bluetooth sub-module;
the closed-loop control module is the core of the whole system and is responsible for analyzing and processing the acquired electroencephalogram signals and predicting whether epileptic seizure occurs in the next 30 minutes, if yes, the epileptic seizure is blocked in a mode of electrically stimulating vagus nerve and anterior thalamic nucleus, and if not, no operation is performed.
FIG. 6 is a circuit schematic of a discharge module; the discharge module comprises a first electrode and a second electrode; the first electrode is disposed at the anterior thalamic nucleus and the second electrode is disposed at the vagus nerve.
According to the results of animal experiments, the application selects the thalamic nucleus and the vagus nerve as electric stimulation targets; then, an electrode is implanted near a target brain region through a minimally invasive surgery, wherein an implanted electrode of the Meidunli company is selected; the electrodes are connected with the embedded chip through leads, and the chip is detachably worn at the waist, the front of the chest or nearby any nearby body of a patient through an elastic belt with adjustable length.
The discharging module uses AD5761 series chips to perform electrical stimulation output, and is mainly considered based on the following points: the load capacity of the chip can reach 1KQ under the output of +/-5V, and the load capacity is improved; the electrical stimulation channels were 2, one for each of the vagus nerve and the anterior thalamic nucleus. The stimulation intensity is 1mA, the stimulation signal is a square wave, the frequency is 50Hz, and the Pulse Width Modulation (PWM) is 0.25.
The discharge module is responsible for electrically stimulating the brain of a user so as to play a role in regulating and controlling a neural circuit and blocking epileptic seizure, electrodes are implanted into the anterior thalamic nucleus and the vagus nerve through a minimally invasive surgery and are connected with the closed-loop regulation module through a lead, and when the discharge is controlled by the closed-loop regulation module;
FIG. 7 is a functional schematic of a closed loop regulatory module; the closed-loop control module is the core of the whole system and is responsible for analyzing and processing the acquired electroencephalogram signals and predicting whether epileptic seizure occurs within 30 minutes in the future, if yes, the epileptic seizure is blocked in a mode of electrically stimulating vagus nerve and anterior thalamic nucleus, and if not, no operation is performed; the epilepsy prediction algorithm is a core part in the core
The process for predicting whether the infant has the epileptic seizure or not by the closed-loop regulation and control module by adopting a basic epileptic seizure prediction algorithm is as follows:
and combining the characteristics of the average value, the extreme value, the standard deviation, the normalized first-order difference, the energy and the power of the time domain signal of the electroencephalogram signal, and then carrying out threshold analysis to predict the epileptic attack.
According to the application, a high-accuracy epileptic seizure prediction model of the light-weight infantile spasm is researched, and then closed-loop control is performed on a reactive nerve regulation and control system according to the prediction result of the model, wherein the main idea is as follows: the method takes a mouse model epileptic seizure prediction model as a basic model, guides the training of the infant epileptic seizure prediction model through the existing epileptic seizure prediction results of the mouse model by a transfer learning method, thereby greatly reducing the model depth and the training time of the infant epileptic seizure prediction model, and simultaneously ensuring the high accuracy of the infant epileptic seizure prediction model. FIG. 8 is a block diagram of the strategy for transfer learning of the present application;
the process of constructing the infant epilepsy convolutional neural network model for deeply predicting infant epilepsy seizure is as follows:
acquiring electroencephalogram data of early epileptic seizure and inter-epileptic seizure of infants through a mouse model, learning priori knowledge of the electroencephalogram data of the early epileptic seizure in an off-line training mode, and training to obtain a DNN basic model (Deep Neural Networks (DNN) is the basis of Deep learning);
the DNN basic model is trained by using a small amount of electroencephalogram data of infant epilepsy, and is migrated to a task of predicting infant epileptic seizure in a priori knowledge training mode to obtain a trained infant epilepsy convolutional neural network model.
Based on a seizure prediction algorithm of an animal model, a transfer learning method is adopted to transfer a deep learning model and experience of a mouse model to an infant seizure prediction task, and the specific method is shown in fig. 7;
the learning system is updated by slightly adjusting the prior knowledge contained in the trained neural network basic model and partial weight and deviation in the original structure, so that the slightly adjusted neural network model can be more suitable for predicting the infant seizure and seizure task, and the overall learning performance is improved.
Meanwhile, the task of predicting the epileptic seizure is designed for closed-loop regulation and control of corresponding nerves of a patient, and the false alarm rate of the task can have great influence on the patient, so that the post-processing technology of epileptic prediction is introduced to further process the prediction result, the false detection rate of the prediction model is reduced, and the prediction performance of the model is improved.
Further: the data set is pre-processed using a sliding time window.
Further, the method comprises the following steps: the process for predicting whether the infant has the epileptic seizure or not by the closed-loop regulation and control module by adopting a basic epileptic seizure prediction algorithm is as follows:
and combining the characteristics of the average value, the extreme value, the standard deviation, the normalized first-order difference, the energy and the power of the time domain signal of the electroencephalogram signal, and then carrying out threshold analysis to predict the epileptic attack.
Further: the closed loop regulation and control module adopts a chip model number XEM6010-LX45, and the XEM6010-LX45 is installed at a clavicle of a wearer.
Further: the depth prediction method for infant epilepsy and depth prediction of infant epileptic seizure comprises the following steps:
s1, constructing a data set of infant seizure characteristics, wherein the data set is divided into a training set and a testing set;
s2, preprocessing the data set to obtain a preprocessed data set;
s3, constructing an infant epilepsia convolutional neural network model for deeply predicting infant epilepsia attacks;
s4, training the infant epilepsy convolutional neural network model based on training set data to obtain a trained infant epilepsy convolutional neural network model;
and S5, inputting the test set data into a trained ISDNN model (infant epilepsy convolutional neural network model) to realize prediction of infant epilepsy attack and obtain a prediction result.
The steps S1, S2, S3, S4 and S5 are executed in sequence;
to better migrate the knowledge of the mouse model seizure prediction model to the clinical infant seizure model, the present application proposes the Infantile Spasms Deep Neural Network (ISDNN) model, as shown in fig. 9.
The electroencephalogram data are segmented, 2 seconds of electroencephalogram data are taken as one segment, the segment is input into a deep learning network, detailed parameter introduction is shown in table 1, and output channels of 13 layers of convolution layers are 64, 64, 128, 128, 256, 256, 256, 512, 512, 512, 512 and 512.
TABLE 1 infantile epilepsy convolutional neural network model parameters
In order to adapt to the feature extraction of brain waveform data, the size of a convolution kernel is set to be a rectangular receptive field of 1 multiplied by 3; the pooling nucleus is set to be a rectangular receptive field of 1 × 2; the stride and fill are sized to correspond to the rectangular configuration.
The epilepsy prediction is a regression problem, only 1 number needs to be output as an output result, so that the output channel of the third fully-connected layer is set to be 1 (the larger the output value is, the larger the probability of the epilepsy attack in 30 minutes is); and after the result processed by the rolling block passes through the self-adaptive average pooling layer, calculating the quantity of parameters required to be received by the first layer of full-connection layer. The output channels of the first and second full-connection layers are set to be the same in number according to the original network, the size of the output channel needs to be between the number of the receiving channels and the final output channel 1, and the output channel is set to be 1024.
The training phase of the infant epilepsy convolutional neural network model (ISDNN model) is mainly divided into two parts:
the first part is that a large amount of electroencephalogram data of epileptic seizure prophase and epileptic seizure interphase collected by an epileptic mouse model, the priori knowledge of the electroencephalogram data of epileptic seizure prophase is learned in an off-line training mode, and a basic model is trained, as shown in the upper square box shown in fig. 10; fig. 10 is a flowchart of the seizure prediction algorithm based on deep learning according to the present application.
In the second part, the underlying model may be trained with a small amount of electroencephalogram data of infant epilepsy, migrated into the task of predicting infant seizures in a way of training based on a priori knowledge, as shown in the middle box of fig. 10; the trained DNN model and a Softmax classifier can be stacked into the ISDNN model, so that the prior knowledge of the mouse seizure model is inherited, and the infant seizure prediction task can be well completed;
after the model is trained, in the prediction phase, the ISDNN model may perform two classification tasks for distinguishing between the early epileptic seizure signals and the inter-epileptic seizure signals of the infant, as shown in the lower box of fig. 10;
meanwhile, in order to ensure the real-time performance of the epileptic seizure prediction algorithm and solve the defects that the computing resources on the chip XEM6010-LX45 are limited and a more effective and more complex deep learning algorithm cannot be written;
the application introduces a means of combining cloud computing with edge computing, namely: the most basic functions of calculating and classifying brain waveform features are completed on the chip XEM6010-LX45, and simple epileptic seizure prediction is carried out, so that closed-loop regulation and control of the whole system can be guaranteed even under the condition of communication failure;
the cloud server runs a highly-configured deep learning server, and can perform a complex artificial intelligence and deep learning algorithm (ISDNN model), so that an epileptic seizure prediction result with higher accuracy can be obtained.
Furthermore, in order to improve accuracy of epileptic seizure prediction and reduce false alarm rate, post-processing technology is introduced to carry out relevant operations.
In the task of predicting infant epileptic seizure, the sliding time window technology is used to divide the original electroencephalogram data of an infant epileptic patient into a plurality of electroencephalogram data segments, and in order to obtain better real-time performance, a shorter sliding time window needs to be set, so that the obtained actual prediction result may fluctuate, and the false alarm rate is improved.
The method comprises the steps of cutting an original electroencephalogram signal according to a section of 10 seconds to obtain n electroencephalogram signal sections, then processing each electroencephalogram signal section by using a prediction algorithm to obtain n prediction results, if the results are all epileptic seizures, and only the result of a certain section in the middle is epileptic seizures, the unique result different from other sections is definitely incorrect, and the post-processing technology is a method for modifying the inconsistent situation into the same with other sections around.
Aiming at the problem, the method for performing smooth correction on the self prediction label based on the neighbor label is adopted, the cache region is arranged for storing the prediction categories of n neighbor samples on the left side and the right side of the neighbor, and for the current electroencephalogram data sample, the prediction label of the current sample is updated according to the prediction label conditions of the neighbors on the two sides.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
1.Jackson A,Moritz CT,Mavoori J.The Neurochip BCI:towards a neural prosthesis for upper limb function,IEEE Transactions on Neural Systems and Rchabilitation Engincering,14(2):187-190,2006.
2.Zanos S,Richardson A G,Shupe L.The Neurochip-2:an autonomous head-fixed computer for recording and stimulating in freely behaving monkeys,IEEE Transactions on Neural Systems and Rehabilitation Engineering,19(4):427-435,2011.
3. Feng Zhouyan, chen Dan, showqiang, a design of a closed-loop neuroelectrical stimulation system, instruments and meters, 33 (2): 279-285,2012.
4. Wang Yang, yang Xiaofei, zhang Yujia, left-ultrasound, wireless closed loop neural signal recording and stimulation system, electronic measurement technique, 36 (1): 83-87,2013.
5.Chen WM,Chiueh H,Chen TJ,Ho CL.A Fully Integrated 8-Channel Closed-Loop Neural-Prosthetic CMOS SoC for Real-Time Epileptic Seizure Control,IEEE JOURNAL OF SOLID-STATE CIRCUITS,49(1):232-247,2014.
6.Salam M T,Velazquez J L P,Genov R.Seizure suppression efficacy of closed-loop versus open-loop deep brain stimulation in a rodent model of epilepsy,IEEE Transactions on Neural Systems and Rehabilitation Engineering,24(6):710-719,2015.
7.Shoaran M.Shahshahani M,Farivar M.A 16-channel 1.1mm2 implantable seizure control SoC with sub-uW/channel consumption and closed-loop stimulation in 0.18 um CMOS,IEEE Symposium on VLSI Circuits(VLSI-Circuits):1-2,2016.
Claims (8)
1. The utility model provides a wearable wicresoft closed loop regulation and control device towards infant's epilepsy which characterized in that: the system comprises an acquisition processing module, a closed-loop regulation and control module, a discharge module, a cloud server and a mobile phone APP;
the collecting and processing module collects and processes scalp electroencephalogram signals and transmits the processed electroencephalogram signals to the closed-loop regulation and control module;
the closed-loop regulation and control module predicts whether the infant has the epileptic seizure by adopting a basic epileptic seizure prediction algorithm to obtain a prediction result I, and when the prediction result I is the epileptic seizure, a control signal I is formed;
the mobile phone APP receives the processed brain electrical signals transmitted by the closed loop regulation and control module;
the cloud server receives the processed brain electrical signals transmitted by the mobile phone APP, carries out deep prediction on the infant epileptic seizure by adopting a deep learning prediction method facing the infant epileptic seizure to obtain a prediction result II, forms a control signal II when the prediction result II is the infant epileptic seizure, and transmits the control signal II to the closed-loop regulation and control module through the mobile phone APP;
when the cloud server is communicated with the closed-loop regulation and control module, the closed-loop regulation and control module sends a control signal II, and the discharge module receives the control signal II sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant;
when the cloud server and the closed-loop regulation and control module are in communication interruption, the closed-loop regulation and control module sends a control signal I, and the discharge module receives the control signal I sent by the closed-loop regulation and control module and electrically stimulates the brain of the infant.
2. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 1, wherein: the infant epilepsy-oriented deep learning prediction method and the infant epilepsy attack deep prediction method comprise the following steps:
constructing a data set of infant epileptic seizure characteristics, wherein the data set is divided into a training set and a testing set;
preprocessing the data set to obtain a preprocessed data set;
constructing an infant epilepsia convolutional neural network model for deeply predicting infant epilepsia attacks;
training the infant epilepsy convolutional neural network model based on training set data to obtain a trained infant epilepsy convolutional neural network model;
and inputting the test set data into the trained infant epilepsy convolutional neural network model to realize the prediction of infant epilepsy attack and obtain a prediction result.
3. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 2, wherein: the process of constructing the infant epilepsy convolutional neural network model for deeply predicting infant epilepsy attack is as follows:
the method comprises the steps that electroencephalogram data of early-stage epileptic seizures and inter-stage epileptic seizures of infants are collected through a mouse model, prior knowledge of the electroencephalogram data of the early-stage epileptic seizures is learned in an off-line training mode, and a DNN basic model is obtained through training;
the DNN basic model is trained by using a small amount of electroencephalogram data of infant epilepsy, and is migrated to a task of predicting infant epileptic seizure in a priori knowledge training mode to obtain a trained infant epilepsy convolutional neural network model.
4. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 1, wherein: the discharge module comprises a first electrode and a second electrode; the first electrode is disposed at the anterior thalamic nucleus and the second electrode is disposed at the vagus nerve.
5. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 1, wherein: the acquisition processing module adopts a wearable 32-channel scalp electroencephalogram acquisition cap.
6. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 2, wherein: the prediction result is based on the method for performing smooth correction on the self prediction label by the neighbor label, and the prediction label of the current sample is updated according to the prediction label conditions of the neighbors on the two sides for the current electroencephalogram data sample by setting a cache region for storing the prediction categories of n neighbor samples on the adjacent left side and the adjacent right side.
7. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 1, wherein: the process for predicting whether the infant has the epileptic seizure or not by the closed-loop regulation and control module by adopting a basic epileptic seizure prediction algorithm is as follows:
and combining the six characteristics of the average value, the extreme value, the standard deviation, the normalized first-order difference, the energy and the power of the time domain signal of the electroencephalogram signal, and then carrying out threshold analysis to predict the epileptic attack.
8. The wearable minimally invasive closed-loop control device for infant epilepsy according to claim 1, wherein: the closed loop regulation and control module adopts a chip model number XEM6010-LX45, and the XEM6010-LX45 is installed at a clavicle of a wearer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210730060.3A CN115177209A (en) | 2022-06-24 | 2022-06-24 | Wearable minimally invasive closed-loop regulation and control device for infant epilepsy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210730060.3A CN115177209A (en) | 2022-06-24 | 2022-06-24 | Wearable minimally invasive closed-loop regulation and control device for infant epilepsy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115177209A true CN115177209A (en) | 2022-10-14 |
Family
ID=83514520
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210730060.3A Pending CN115177209A (en) | 2022-06-24 | 2022-06-24 | Wearable minimally invasive closed-loop regulation and control device for infant epilepsy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115177209A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6658287B1 (en) * | 1998-08-24 | 2003-12-02 | Georgia Tech Research Corporation | Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity |
| CN101261304A (en) * | 2008-04-25 | 2008-09-10 | 山东大学 | Low current grounding system distribution circuit single-phase earth fault automatic position setting method |
| US20080243022A1 (en) * | 2007-03-30 | 2008-10-02 | Donnett James G | Seizure prediction using brain signal telemetry |
| CN104874102A (en) * | 2015-04-02 | 2015-09-02 | 中国科学院苏州生物医学工程技术研究所 | Multi-channel wireless closed loop deep brain neural sensing and control system |
| WO2022010913A1 (en) * | 2020-07-06 | 2022-01-13 | Johnson Controls Tyco IP Holdings LLP | Cloud-based hvac&r control systems and methods |
| CN114082102A (en) * | 2021-11-12 | 2022-02-25 | 江苏科技大学 | Closed-loop vagus nerve stimulator based on scalp electroencephalogram |
| CN114306931A (en) * | 2021-12-28 | 2022-04-12 | 北京品驰医疗设备有限公司 | Pulse generator and control method thereof |
-
2022
- 2022-06-24 CN CN202210730060.3A patent/CN115177209A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6658287B1 (en) * | 1998-08-24 | 2003-12-02 | Georgia Tech Research Corporation | Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity |
| US20080243022A1 (en) * | 2007-03-30 | 2008-10-02 | Donnett James G | Seizure prediction using brain signal telemetry |
| CN101261304A (en) * | 2008-04-25 | 2008-09-10 | 山东大学 | Low current grounding system distribution circuit single-phase earth fault automatic position setting method |
| CN104874102A (en) * | 2015-04-02 | 2015-09-02 | 中国科学院苏州生物医学工程技术研究所 | Multi-channel wireless closed loop deep brain neural sensing and control system |
| WO2022010913A1 (en) * | 2020-07-06 | 2022-01-13 | Johnson Controls Tyco IP Holdings LLP | Cloud-based hvac&r control systems and methods |
| CN114082102A (en) * | 2021-11-12 | 2022-02-25 | 江苏科技大学 | Closed-loop vagus nerve stimulator based on scalp electroencephalogram |
| CN114306931A (en) * | 2021-12-28 | 2022-04-12 | 北京品驰医疗设备有限公司 | Pulse generator and control method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 王耀民: "基于深度神经网络迁移特征学习的癫痫发作检测研究", 《中国优秀硕士学位论文全文数据库;医药卫生科技辑》, 15 January 2021 (2021-01-15), pages 9 - 10 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kassiri et al. | Closed-loop neurostimulators: A survey and a seizure-predicting design example for intractable epilepsy treatment | |
| AU2021202186B2 (en) | Methods of sensing cross-frequency coupling and neuromodulation | |
| Kassiri et al. | Battery-less tri-band-radio neuro-monitor and responsive neurostimulator for diagnostics and treatment of neurological disorders | |
| CN104225790B (en) | Closed loop neural stimulation system | |
| WO2021184439A1 (en) | Multi-target electrical stimulation circuit, electrical stimulator, and signal output method therefor | |
| Hosain et al. | Closed loop deep brain stimulation: an evolving technology | |
| US20070213786A1 (en) | Closed-loop state-dependent seizure prevention systems | |
| Herron et al. | Closed-loop DBS with movement intention | |
| US11324955B2 (en) | Implantable hub and satellite system for neural recording and stimulation | |
| CN109646796A (en) | Channel wireless radio multi closed loop stimulation system for epilepsy therapy | |
| Yoo et al. | Neural interface systems with on-device computing: machine learning and neuromorphic architectures | |
| CN102626538A (en) | Sleep detection-based integrated closed loop deep brain stimulator | |
| CN110694169A (en) | Motor dysfunction nerve bridging system based on motor intention inducing central nervous system micro-electrical stimulation | |
| WO2020095111A1 (en) | Methods and systems for sensing balance between neuronal activity inputs and pathways and neuromodulation | |
| Liao et al. | Neural motor control differs between bimanual common-goal vs. bimanual dual-goal tasks | |
| CN105125186B (en) | A kind of method and system of definite therapeutic intervention mode | |
| WO2008109508A2 (en) | Automatic parameter selection and therapy timing for increasing efficiency in responsive neurodevice therapies | |
| Wang et al. | Effects of tDCS on brain functional network of patients after stroke | |
| CN116549853B (en) | Pulse generator, stimulator, storage medium, and program product | |
| CN115177209A (en) | Wearable minimally invasive closed-loop regulation and control device for infant epilepsy | |
| Shoaran et al. | Intelligent Neural Interfaces: An Emerging Era in Neurotechnology | |
| CN117065217A (en) | Potential signal acquisition device, potential signal acquisition method, medical system, and readable storage medium | |
| Stanslaski et al. | Creating neural “co-processors” to explore treatments for neurological disorders | |
| Volosyak et al. | A novel calibration method for SSVEP based brain-computer interfaces | |
| Basu et al. | Adaptive control of deep brain stimulator for essential tremor: entropy-based tremor prediction using surface-EMG |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |