300 results about "Entire brain" patented technology

The invention discloses a multi-modal feature joint constraint whole-brain dynamics modeling and parameter optimization method which is applied to the field of computational neuroscience and brain science. Aiming at the defects of a current whole brain dynamic model in the aspect of synchronously fitting multi-modal scalp EEG signals and cortical and subcortical BOLD signals across spatial and temporal scales, the capability of the model for comprehensively and accurately representing real dynamic characteristics of brain activities on different scales is limited; in order to overcome the short plate, the core element of multi-modal feature combined constraint is added on the basis of a whole-brain dynamic model and a parameter space rapid optimization method, and a modeling method and a parameter rapid optimization strategy for the multi-modal feature combined constraint whole-brain dynamic model are designed; by applying a high-dimensional parameter rapid optimization algorithm, the capability of the whole-brain dynamic model in synchronous fitting of brain multi-modal spatial-temporal characteristic signals is further improved; the method has important significance and reference value for developing a whole-brain dynamic model modeling method and a parameter optimization system which are more biological rationality and interpretability.

The invention provides an Alzheimer's disease diagnosis method based on multi-modal brain network feature fusion, which is based on a multi-modal brain image, a neural network and an attention mechanism, and has the main innovation points that: 1) effective feature combination of three types of multi-modal data of MRI, DTI and fMRI can reflect all feature information of a patient sample; and 2) aiming at different image data, a targeted feature extractor is used, the three-dimensional convolution is more suitable for feature learning of a three-dimensional MRI whole brain image, the Mama model is more suitable for learning of fMRI time sequence features, and the graph convolution network is more suitable for learning of DTI fiber number structure brain network features. And 3) the innovative mixed attention mechanism is fused with three types of multi-modal features, position coding and a Transform structure are used at the same time, the mixing effect and the classification accuracy of the multi-modal features are improved, and compared with a convolution and full connection layer structure, richer information can be learned.

The invention relates to the field of electroencephalogram signal processing, and particularly discloses a multi-mode electroencephalogram characteristic cognitive ability evaluation system based on a neural network, comprising: an acquisition module used for synchronously acquiring whole brain signals and eye movement data during pilot simulation training; the data preprocessing module is used for primarily processing the received whole brain signals and the eye movement data, monitoring the primarily processed data in real time by utilizing a preset monitoring mechanism, and triggering to generate interaction test starting signals if preset symbolic fluctuation occurs in the whole brain signals or the eye movement data; and the test interaction sub-module is used for generating arithmetic test questions according to a preset rule after receiving the interaction test starting signal, and synchronously recording answer correlation parameters of the pilot. By adopting the technical scheme of the invention, the deep-level reaction mechanism of the brain under the stimulation of the special complex cognitive task of flight training can be fully considered, and the cognitive load change of the pilot in the simulated training can be comprehensively and accurately reflected.

The invention belongs to the technical field of food microorganisms, and relates to lactobacillus plantarum and application thereof in preparation of products for relieving insomnia. The preservation number of the lactobacillus plantarum is GDMCC No: 65061. The lactobacillus plantarum can be applied to efficient fermentation of spina date seed juice, the content of GABA, polypeptide, total phenols and total flavonoids in the spina date seed juice is increased, and the content of spinosin, 6 ''-feruloyl spinosin, jujuboside A, jujuboside B and betulinic acid is increased; the reduction of whole brain coefficient and the increase of liver coefficient caused by long-term insomnia are relieved, the bad condition caused by long-term insomnia in a hippocampus DG region in the brain of the mouse is improved, the sleep time of the mouse is prolonged, and the insomnia symptom of the mouse is relieved; the content of inhibitory neurotransmitters 5-hydroxytryptamine, melatonin and gamma-aminobutyric acid in the hypothalamus of the mouse is increased, the content of excitatory neurotransmitter glutamic acid is reduced, and the reduction of organism immunity and oxidative stress of brain tissues caused by insomnia are relieved.

The invention discloses a Parkinson's dyskinesia individualized SCAN network positioning method based on a multi-modal image and deep learning. The method comprises the steps of obtaining multi-modal medical image data, preprocessing the multi-modal medical image data, obtaining a multi-modal structure image and functional connection data, and calculating a spontaneous neural activity index of a whole-brain voxel level; taking a priori brain region related to the spontaneous neural activity index and dyskinesia as a seed point, constructing a seed point voxel function connection graph representing individual brain function connection, and performing nonlinear feature fusion and extraction through the deep learning network model; the bilinear attention network is adopted to capture the interaction information of the feature data and the individual dyskinesia symptom which is significantly related, an individualized SCAN network positioning result is obtained, the structure-function coupling characteristics of the individual brain are comprehensively described, the cross-modal pathological features related to the dyskinesia can be more sensitively recognized, and the accuracy and accuracy of the diagnosis and treatment of the dyskinesia can be improved. And the accuracy and robustness of abnormal brain region detection are obviously improved.

A juvenile depression cognitive impairment subtype classification and prognosis prediction method relates to the technical field of medical treatment, and mainly comprises the following steps: performing clinical evaluation and therapeutic response evaluation on a subject, performing MRI and magnetoencephalogram data acquisition, constructing a whole brain MSN of the subject, identifying MSN abnormal characteristics, obtaining functional connection change of a frequency band when magnetoencephalogram is abnormal, and determining the cognitive impairment subtype classification and prognosis prediction of the cognitive impairment subtype of the subject. A subtype classification model is established by fusing the MSN and cognitive function evaluation data, and a prognosis prediction model is established by analyzing MSN abnormal features, functional connection changes of frequency bands during abnormality, multi-dimensional treatment reactions and high-risk behaviors. According to the method, different levels of fusion measurement are carried out on the juvenile depression with cognitive function impairment brain mechanism through multi-modal brain images, a subtype classification model with diagnosis and treatment values is established, and a prognosis prediction model with clinical transformation potential is constructed; therefore, a theoretical basis and a technical means are provided for individualized precise diagnosis and treatment of the cognitive impairment of the juvenile depression.

The invention relates to the field of biomedicine, and particularly discloses an epileptic focus artificial intelligence positioning system and method based on magnetic resonance and electroencephalography, and the system comprises the following contents: a data collection module is used for synchronously collecting T1 weighted magnetic resonance images and electroencephalogram signals of scalp; the data preprocessing module is used for performing brain tissue segmentation on the magnetic resonance image, obtaining structural features of each brain region, constructing vectors including whole brain structural features, and obtaining magnetic resonance structural feature vectors; artifacts of the electroencephalogram signals are removed, electroencephalogram features of all brain areas are extracted through time-frequency analysis, a matrix containing whole electroencephalogram physiological features is constructed, and the electrophysiological features are obtained; the cross-modal confidence coefficient dynamic evaluation module is used for establishing a bidirectional constraint rule to perform confidence coefficient calibration on the magnetic resonance structure feature vector and the electrophysiological feature; a positioning model construction and training module; a positioning result output module; according to the technical scheme, noise can be reduced during magnetic resonance and electroencephalogram fusion, and the epileptic focus positioning precision is high.

The invention discloses a tumor space-occupying brain network neural image alignment method based on multi-modal fusion, and belongs to the technical field of medical image processing and artificial intelligence crossing. The method comprises the following core steps of multi-modal image heterogeneous feature decoupling, tumor occupation deformation field modeling, functional network topological structure maintenance, cross-modal feature adversarial alignment, dynamic deformation constraint optimization and clinical interpretability verification, and construction of a three-dimensional non-rigid registration network based on a double attention mechanism. And differential homeomorphic mapping of a tumor focus area and normal brain tissue is realized through the cascaded spatial transformation module. Aiming at the problems of insufficient multi-modal feature alignment and brain network topology distortion in the prior art, the invention provides a function connection constrained cross-modal fusion strategy, a graph convolution network is adopted to encode resting state function connection features, and network node displacement caused by tumor occupation is dynamically corrected in combination with deformable convolution and a bidirectional feature competition mechanism; a space consistency loss function based on white matter fiber bundle tracing is designed, and through diffusion tensor imaging feature guide structure-function bimodal joint optimization, the problems of insufficient registration precision in a focus area and whole brain network connection distortion of a traditional method are solved. Experiments show that the registration precision of the method in glioma cases reaches 0.82 mm and is improved by 37% compared with that of a traditional method, and dissection-function consistency of functional network reconstruction around tumors is remarkably improved.

The invention discloses a sleep real-time staging modeling method, a sleep real-time encoding and decoding method and a sleep real-time staging modeling system. EEG, EOG and EMG signals of a subject are collected, statistics, frequency domain and frequency domain characteristics and other characteristics are extracted after preprocessing, sleep stages marked by experts serve as real labels, a model is trained in a supervised learning mode, and real-time classification of the sleep stages is achieved. Furthermore, high-precision encoding and decoding of sleep content are realized by applying stimulation prompt during sleep, collecting and preprocessing whole-brain EEG signals, distinguishing NREM and REM stages, training encoding and decoding models respectively, and aligning nerve characterization during waking and sleep by utilizing comparative learning.

According to the multi-scale brain age prediction model construction method based on the magnetic resonance image and the application, the constructed brain age prediction model is higher in generalization and robustness, higher prediction precision is kept, the whole brain-sub-network-voxel brain age can be predicted, the predicted brain age has better interpretability in the physiological sense, and the brain age prediction accuracy is improved. The difference of brain ages among different sub-networks and a specific mode of PAD and cognition association are explored, the specific sub-network for regulating cognition is found, the difference mode of aging of different brain regions is seen from the voxel level, and the prediction performance of the model is superior to that of a current mainstream neural network model. The method comprises the following steps: (1) data collection; (2) data preprocessing; (3) constructing a whole-brain and functional sub-network brain age prediction model based on a simple full convolutional neural network SFCN method; (4) constructing a voxel level brain age prediction model based on a ScaledDense U-Net method; and (5) carrying out offset correction on the brain age deviation.

The invention discloses a disturbance of consciousness stimulation regulation and control system and method fused with electroencephalogram connection recognition. The system comprises a simulated electroencephalogram signal data acquisition stage, a connection recognition analysis stage, a stimulation parameter optimization stage and an executable stimulation instruction conversion stage. The method has the following advantages and effects that a whole-electroencephalogram activity distribution diagram is generated by simulating an electroencephalogram signal data acquisition stage, a key connection area is identified, and a brain function connection map is generated by utilizing a function connection analysis network and a phase synchronization algorithm in a connection identification analysis stage; in the stimulation parameter optimization stage, space-time correlation between a whole electroencephalogram activity distribution map and a brain function connection map is established, a multi-objective optimization algorithm is adopted to generate an optimized stimulation parameter set through a fusion network, and finally, in the executable stimulation instruction conversion stage, the optimized stimulation parameter set is converted into an executable stimulation instruction based on a self-adaptive control model. Therefore, the accuracy and the self-adaptive capability of electroencephalogram signal stimulation regulation and control are remarkably improved.

The invention provides an individualized TMS target positioning method and system for spastic cerebral palsy children, and aims to realize TMS precise nerve regulation and control for different types of SCP children. According to the method, the structural features of the healthy child are obtained, and after image processing, accurate division of the SCP child brain region is achieved, the defects of a current brain injury child brain region spectrogram are overcome, and a basis is provided for seed point selection of different SCP symptoms; multi-modal image parameter features are further scanned, whole-brain voxel correlation analysis is performed based on seed points to construct functional connection diagrams, the crossed and overlapped functional connection diagrams are searched to serve as a liability focus network of different symptoms, a voxel set close to the scalp surface serves as a TMS target, and SCP child treatment targets of different clinical symptoms are accurately positioned. Based on a focus network mapping technology, functional connection characteristics related to different symptoms or symptom complexes are better positioned, a specific brain network is identified, and an accurate intervention target is provided for TMS treatment.

The invention provides a voice recognition method and device based on a brain-like model, electronic equipment and a storage medium, and the method comprises the steps: obtaining a whole-brain network topological structure according to a brain function network generated by human brain image data, and carrying out the recognition of a whole-brain network through employing a multi-class neuron model as a node and a synaptic plasticity model as an edge, constructing a multi-brain-region pulse neural network as a brain-like model; constructing a speech recognition framework of the brain-like model; electromagnetic intervention is applied to different brain areas of the brain-like model, optimal electromagnetic intervention parameters are determined by analyzing the voice recognition accuracy of the brain-like model before and after electromagnetic intervention, and brain-like model voice recognition is carried out according to the optimal electromagnetic intervention parameters. According to the invention, the speech recognition performance of the brain-like model can be effectively improved, the biological interpretability and neural information processing capability of the brain-like model are further improved, and the development of brain-like intelligence in the application of a mode recognition task is promoted.

The invention relates to the technical field of medical image processing, and discloses a brain network image analysis method, system and device for a glaucoma patient and a medium, and the method comprises the steps: carrying out the brain region scanning of the glaucoma patient, and obtaining a high-resolution structure image, a resting state function image and a diffusion tensor image; pre-processing the resting state function image and the diffusion tensor image, and respectively registering the resting state function image and the diffusion tensor image with the high-resolution structure image to obtain a registered resting state function image and a registered diffusion tensor image; performing brain region division on a glaucoma patient, wherein each region is used as a node; on the basis of registration resting state functional imaging, Pearson's correlation coefficients between the nodes are calculated to construct a whole-brain functional network; on the basis of registration diffusion tensor imaging, the number of fiber bundle connections between nodes is calculated to construct a whole-brain structure network; and carrying out coupling analysis on the whole brain function network and the whole brain structure network to obtain an F-S coupling coefficient. The method is helpful for deeply researching the influence of glaucoma on the brain network.

The invention provides a TMS target spot positioning method and system based on a function connection weight. The method comprises the following steps: preprocessing structural magnetic resonance image data and functional magnetic resonance image data of a plurality of subjects; in the preprocessed structural magnetic resonance image data, obtaining a region of interest corresponding to a TMS stimulation target of each subject to obtain a plurality of regions of interest; obtaining a functional connection matrix between the plurality of regions of interest and whole brain regions in the preprocessed functional magnetic resonance image data; obtaining a function connection weight matrix between the depression scale score data and the function connection matrix; mapping the function connection weight matrix into a function connection matrix to obtain an estimated curative effect map of the whole brain region; and carrying out clustering on voxels of a left DLPFC area in the estimated curative effect diagram, and determining a clustering center as a TMS treatment target. According to the method, the problem that the TMS used in the related technology is lack of individualized accurate stimulation target spots is solved.

The invention provides a quantitative risk assessment method and system for head and neck artery stenosis, and the method comprises the steps: collecting the medical image of the head and neck artery of a patient, the blood pressure of the upper arm artery, and the heart rate; a narrow local form, a cerebral artery network anatomical structure / geometric size and a microcirculation blood flow automatic regulation mechanism are comprehensively incorporated into hemodynamic analysis through a geometric multi-scale modeling technology, and information such as an arterial stenosis far-end and near-end pressure ratio, cerebral artery blood flow and hemodynamic parameters of a narrow part is output; quantitative information support is provided for risk assessment of head and neck artery stenosis; the method has the advantages that patient data are obtained only through noninvasive measurement, and the risk that complications are increased due to invasive measurement is avoided; according to the method, a geometric multi-scale modeling method is adopted to achieve integrated analysis of the three-dimensional flow field and the whole-brain circulation hemodynamic parameters of the stenosis part, the blood flow regulation effect of brain microcirculation is considered, and the physiological significance and clinical application value of the evaluation result can be remarkably improved.

The invention discloses a brain connection coupling analysis method based on grey matter morphological similarity-white matter fiber bundles, relates to the technical field of neuropsychiatric imaging, and has the technical key points that the method can be used for analyzing grey matter-white matter brain connection group coupling modes of the whole brain level, the level in brain network communities and the level between the brain network communities; the integration characteristics of structural tissues of the brain on different scales can be understood; the reliability of statistical analysis is improved by adopting a spatial autocorrelation correction method; based on a specific machine learning fusion model, the clinical application value of the model is verified.

The invention discloses a control device and computer equipment for a brain health therapeutic apparatus, and relates to the technical field of brain stimulation control, and the method comprises the steps: synchronously collecting an fMRI activity map, sMRI data and whole-brain voxel-level DTI data of a subject, extracting a cortex target region to be stimulated, and reconstructing and generating a triangular mesh model; non-linear registration is carried out on whole-brain voxel-level DTI data and the sMRI space, and a target region vertex direction field is generated; calculating a direction field average direction and a direction field covariance of the direction field at the vertex of the target region; if the covariance of the direction field does not exceed a preset covariance threshold value, controlling the brain health therapeutic apparatus to perform magnetic stimulation operation on the cortex target region according to the average direction of the direction field; and if the direction field covariance exceeds a covariance threshold, screening a first target region vertex sector and a second target region vertex sector, and controlling the brain health therapeutic apparatus to perform time division multiplexing vector scanning between sectors. Therefore, the accuracy of targeted stimulation and the individual suitability are greatly improved.

The invention discloses a progressive migration training decoding method based on GPT, and belongs to the field of optic nerve decoding. According to the method, a GPT-based category and text decoding model is constructed based on functional magnetic resonance signal data recorded when a subject watches a large number of visual perception images, and the method comprises an image encoder used for embedding image feature vectors extracted by adopting a pre-training model into a hidden feature space; prompt containing task information and position codes are embedded into a hidden feature space; the visual information encoder is used for mapping a brain visual region voxel signal to a hidden feature space by using a GRU; the whole-brain information encoder is used for guiding the selection of voxel signals of other regions of the brain by taking a voxel signal of a visual region of the brain as a context, and mapping global information to a hidden feature space; and the language generator fuses all the hidden features and selects the Token with the highest probability to generate the text of the corresponding content. According to the method, the decoding accuracy is improved, and the generalization ability of the model is enhanced.

The invention provides a modeling method of a whole-brain space-time multi-layer functional connection network, which comprises the following steps: preprocessing resting state functional magnetic resonance imaging data, and extracting a blood oxygen level dependent signal of a region of interest; dividing the blood oxygen level dependent signal into a plurality of overlapped time windows by adopting a time window sliding method; based on a Pearson correlation method and an L1 sparse linear regression algorithm, estimating edges and hyperedges between the brain regions; based on a graph theory and a hypergraph theory, modeling the edge and the hyperedge into a graph and a hypergraph respectively, and constructing a spatial multi-layer function connection network; and constructing a spatial multi-layer function connection network in each time window so as to construct a whole-brain space-time multi-layer function connection network. According to the method, comprehensive description of space and time multi-layer dynamic features of the brain FCN can be realized, and the characterization capability of the brain FCN on brain dynamic information is improved.

The invention discloses a white matter microstructure calculation system and method based on multi-shell diffusion-weighted imaging, and the system comprises a data collection module which is used for obtaining diffusion-weighted imaging and structural image magnetic resonance scanning of a testee; the first data processing module is used for obtaining individual fiber orientation distribution and generating a public template of group-specific fiber orientation distribution at the same time; the second data processing module is used for co-registering the individual fiber orientation distribution diagrams to a common template and segmenting the fiber orientation distribution to obtain white matter microstructure indexes of a whole brain level; the third data processing module is used for obtaining a white matter microstructure index of each anatomical fiber bundle level; and the fourth data processing module is used for obtaining a whole brain fiber streamline based on probabilistic fiber bundle tracking and spherical deconvolution information filtering, and generating a white matter microstructure index of a network level in combination with a prior brain map. By adopting the technical scheme of the invention, quantitative calculation and analysis of the white matter microstructure change of the brain are realized.

The invention relates to the field of ultrasonic sleep aiding, and provides an ultrasonic-infrasound coupled sound wave whole-brain sleep regulation and control method and device. The whole-brain sleep regulation and control method based on the ultrasound-infrasound coupling sound waves comprises the steps that electroencephalogram signals of a user are collected through electroencephalogram collection equipment; physiological parameters of the user are obtained through physiological state monitoring equipment; a control device is adopted to control and adjust the fundamental frequency, pulse width, pulse repetition frequency, difference frequency, intensity and other parameters of two columns of ultrasonic waves generated by a sound wave emission device according to the electroencephalogram signals and the physiological parameters; a sound wave emitting device is adopted to generate and emit two columns of ultrasonic waves which are close in frequency and face to face, the two columns of ultrasonic waves are fed into the brain face to face, and the two columns of ultrasonic signals inhibit the cerebral cortex activity when passing through the cerebral cortex; meanwhile, interference occurs in the deep brain target nuclear region, low-frequency infrasound beat frequency waves are formed, electroencephalogram slow wave rhythm resonance is induced, sleep-related neural activities of the deep brain target nuclear region are synchronized, and coordinated regulation and control of the whole brain region are achieved.

The invention provides a lung cancer brain metastasis risk prediction method and system, and the method comprises the steps: obtaining the brain MRI image data of a user, and carrying out the data preprocessing; whole-brain feature extraction and partition feature extraction based on an AAL template are carried out on the preprocessed data, and whole-brain features and partition features are combined to obtain a feature matrix; according to the preprocessed brain MRI image data of the user suffering from lung cancer brain metastasis, training an auto-encoder to obtain a trained auto-encoder; inputting the feature matrix into the trained auto-encoder, and calculating a reconstruction error; according to the reconstruction error, an isolated forest method is adopted to further screen abnormal areas, and an abnormal score is generated for each sample; according to the reconstruction error and the abnormal score, the abnormal score of the whole brain feature and the abnormal score of the partition feature are calculated, the high-risk region of the brain can be accurately marked by combining the abnormal detection of the whole brain feature and the partition feature, and a basis is provided for personalized PCI treatment.

The invention belongs to the technical field of brain dynamics modeling, particularly relates to a regulation and control scheme auxiliary analysis method based on a disease specificity whole brain dynamics model, and aims to realize accurate screening of nerve regulation and control schemes. Comprising the steps that S1, multi-mode magnetic resonance imaging data are obtained and preprocessed, structural connection, a brain region BOLD time sequence, static function connection and dynamic function connection are obtained, and prior constraint values of myelination and cortical atrophy are obtained through calculation based on the preprocessed data; s2, parameterizing the local return coupling strength of the model based on a prior constraint value, coupling the models of different brain regions through the structural connection obtained in the step S1, and constructing a whole brain dynamic model through parameter optimization; s3, training a variational auto-encoder by using different groups of static function connections and dynamic function connections; and S4, applying different regulation and control schemes on the whole brain dynamic model, obtaining a brain state evolution trajectory induced by regulation and control through the trained variational auto-encoder, and screening an effective regulation and control scheme.

The invention relates to a heterogeneous multi-core distributed brain-like bionic architecture based on neural mimicry and a working method thereof. The heterogeneous multi-core distributed brain-like bionic architecture comprises a neural mimicry processor, a multi-node matrix processor, a memristive neural mimicry unit, a neural plastic unit, a neural mimicry routing system, a pulsation coupling unit and a whole-brain space simulation processing unit. The method is inspired by a working mechanism of each brain component and a neuron plasticity mechanism in a brain thinking process, and knowledge such as functional analysis and a graph group theory in mathematics are combined with related theories such as graph reinforcement learning and group decision in artificial intelligence; a bionic chip is constructed by using a hardware technology, and biological components such as thalamus and basal nerve knots are comprehensively simulated; the problems that an artificial intelligence algorithm in the market is low in speed, complex and low in efficiency; the architecture and the method are mainly applied to multi-AI scenes such as image scene analysis, voice recognition, sentiment analysis, intelligent driving, root analysis, risk prediction, decision analysis and safety confrontation.

Methods, systems, and devices are described for reconstructing spatially resolved electrical activity in the brain. In some example embodiments, EEG and MRI data are used to estimate volumetric distribution of electrostatic potential inside the MRI domain throughout the entire brain. Spatially and temporally varying field estimates can be generated using a brain wave model which is based on weakly evanescent transverse cortical wave propagation and constrained using the tissue properties gained from the MRI data. The disclosed techniques enable brain activity imaging with high spatial and temporal resolution, thereby providing a tool for assessing functional brain states and monitoring changes in those states in relation to various normal and pathological conditions.

The invention discloses a dynamic space-time CNN-Transform emotion brain-computer interface decoding method, and relates to the technical field of brain-computer interfaces, a dynamic time feature extraction module is designed according to the sensitivity of multi-scale convolution to an electroencephalogram sequence along a time dimension, and electroencephalogram sequence time feature information is mined by using convolution kernels of different sizes; constructing a local-global spatial feature extraction module by referring to close correlation between asymmetry of left and right brain regions of the brain and an emotional state, and sequentially extracting spatial feature information of the left brain, the right brain and the whole brain of the electroencephalogram sequence; a spatial-temporal feature fusion module is designed, and spatial-temporal feature relations among the left brain, the right brain and the whole brain are mined; the long-time dependency relationship in the electroencephalogram sequence is captured through an attention mechanism by referring to the advantage of Transform on long-time sequence processing, so that the emotion electroencephalogram decoding precision is effectively improved; and finally, performing emotion recognition on the feature sequence subjected to Transform coding by using a multi-layer perceptron to realize end-to-end emotion electroencephalogram decoding.

The invention discloses a PET-based epilepsy postoperative prognosis information generation system and method, and the system comprises a preprocessing module, a target training data generation module, a training module and an application module which are sequentially connected. Through processing the whole brain health PET image, the preoperative whole brain PET image and the postoperative whole brain magnetic resonance imaging, the obtained target training data focuses on the metabolic connection characteristics of the focus area and the whole brain, a new view angle is provided for the postoperative prognosis prediction, so that the target training data is adopted to train the deep learning model, and the prognosis prediction accuracy is improved. A deep learning model focusing on the metabolic connection characteristics of the focus area and the whole brain can be obtained, so that the postoperative whole brain magnetic resonance imaging is accurately processed, and prognosis information is obtained.

The invention discloses an infantile autism diagnosis method based on brain function correlation structure modeling and a default mode network, which comprises the following steps of: firstly, preprocessing acquired infantile autism data, and extracting average time sequence data from the acquired infantile autism data to construct a whole brain function connection matrix and a DMN (default mode network) function connection matrix; a fusion weight parameter is initialized and is used for fusing the two constructed matrixes; constructing a cross-dimension dual adaptive attention module to extract features from a time dimension and a space dimension; a product-based embedded graph generation module is defined to generate a graph structure, the nodes correspond to brain regions, and the weights of the edges correspond to the similarity between the nodes; and high-order features are extracted from the generated graph structure by defining a graph convolutional network predictor, and node features are mapped to classification tags for subsequent diagnosis of autism. The method disclosed by the invention has the beneficial effects that the brain function association relationship of the autism patient is disclosed, the interpretability of clinical application is enhanced, and the autism diagnosis accuracy is improved.

The invention provides a sample similarity knowledge distillation method and device based on multi-channel electroencephalogram, and the method comprises the steps: a first calculation step: inputting the whole brain electroencephalogram data of a sample into a pre-trained teacher network; a second calculation step: sequentially inputting first input data in the multiple groups of current first training data into the student network for training; a verification step: performing comparison verification according to all the student tags and the corresponding soft tags to obtain the accuracy of the student network; and under the condition that the accuracy does not meet the accuracy requirement, optimizing the student network, and sequentially and repeatedly executing the first calculation step, the second calculation step and the verification step at least once until the accuracy meets the accuracy requirement, thereby obtaining the lightweight student network. The problem that in the prior art, traditional electroencephalogram data knowledge distillation neglects the internal similarity relation between input samples, and consequently the expression ability of a student network is insufficient in the learning process is solved.