466 results about "Brain function" patented technology

The invention provides a spatial-temporal feature fusion-based dynamic graph convolution electroencephalogram depression detection method, which comprises the following steps of: firstly, segmenting a sample into fragments with the length of 1 second, and calculating power spectral density (PSD) as an input feature by adopting a Welch method; the time sequence and spatial characteristics of the EEG signals are synchronously extracted through a double-branch architecture, wherein one branch captures the long-term time sequence dependence of the EEG signals by using a GRU; and the other branch adopts an improved TSCN (separable convolution is introduced), multi-scale spatial features from fine to rough are extracted through causal convolution and expansion convolution of residual layered stacking, after double-branch features are adaptively fused based on an attention mechanism, a dynamic graph structure is constructed, functional connection evolution of brain intervals is modeled by using a graph convolution network, and a dynamic graph structure is constructed. The topological structure of the network is optimized through a back propagation process, and finally depression identification is realized through a Softmax classifier. According to the method, the time sequence modeling capability of the GRU and the multi-scale spatial analysis capability of the TSCN are fused, the representation limitation of a single model is broken through, the dynamic change of a brain function network is adaptively captured through dynamic graph convolution, the physiological interpretability is enhanced, deep complementary fusion of EEG spatial and temporal characteristics is realized, the depression recognition accuracy is remarkably improved, and the method is suitable for popularization and application. And an efficient tool is provided for auxiliary diagnosis of mental diseases.

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.

Methods and apparatus for changing a brain state of a person from an initial brain state to a target brain state are described. The method includes receiving information characterizing the initial brain state, the information including a structural composition and a functional architecture of the brain, estimating based, at least in part, on the received information, a potential for the brain to change from the initial brain state to the target brain state, determining based, at least in part, on the received information and the estimated potential for the brain to change from the initial brain state to the target brain state, a non-invasive brain stimulation protocol, and controlling at least one non-invasive brain stimulation device to stimulate the brain according to the non-invasive brain stimulation protocol. The method also includes using brain information to inform the design of computational general artificial intelligence agents.

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 discloses a technology for targeted stimulation of a specific brain region and application of the technology in regulation and control of brain functions, and belongs to the field of animal cognitive behavior research. Techniques for targeted stimulation of specific brain regions include insulated stimulation electrodes that can generate AC voltage stimulation and a stimulation device for adjusting the AC voltage intensity and frequency. By applying the technology (ntTMS) for targeted stimulation of the specific brain region, an obvious magnetic stimulation effect can be achieved in the center of the brain region, cerebral neurons of a target region are activated, social and cognitive behaviors of animals are effectively regulated and controlled, and no obvious side effect is brought to experimental animals.

Disclosed are devices, systems and methods for acquiring, analyzing, and utilizing neurofeedback to promote brain coherence. Neurofeedback is a form of biofeedback that allows an individual to regulate his / her brain activity by providing a visual metaphor of brain function, thereby making it accessible for manipulation. In some embodiments of the present technology, a system includes a brain signal detection device wearable by a subject and a computer device including a display and a brain-computer interface (BCI) configured to monitor brain signals and display visual, auditory, and / or tactile stimuli to the subject according to a neurofeedback threshold-based protocol to deliver brain signal coherence between the left and right hemispheres of a subject's brain.

The invention belongs to the field of electroencephalogram signal classification detection, and particularly relates to a neuropsychiatric disease electroencephalogram diagnosis method based on iteration polar coordinate attention, which comprises the following steps: acquiring a multi-channel EEG (electroencephalogram) signal and preprocessing the multi-channel EEG signal; inputting the EEG preprocessing signal into a pre-trained improved LaBraM model to obtain a plurality of node features; calculating a Pearson's correlation coefficient matrix and a cosine similarity matrix according to the EEG preprocessing signal, and then constructing a brain function fusion connection matrix as an adjacent matrix through an iteration polar coordinate attention mechanism; according to the method, the large-scale pre-trained EEG model and polar coordinate attention are used for brain graph structure construction for the first time, and collaborative optimization of time feature generalization and space structure modeling capacity is achieved.

The present invention describes an artificial intelligence (AI) enabled electroencephalography (EEG) system that integrates Pathway Hierarchical Adaptive Referencing (PHAR) for localized signal detection, large language models (LLMs) for automated EEG reporting, and Internet of Medical Things (IoMT) connectivity for adaptive neuromodulation control. The system can also deliver transcranial electrical stimulation (tES) pulses and function as an electrical impedance tomography (EIT) system. PHAR employs a multi-layered multiplexer hierarchy and adaptive referencing topologies to optimize EEG signal acquisition and spatial resolution. LLM integration enables automated generation of human-readable EEG reports. IoMT connectivity allows closed-loop neuromodulation, where real-time EEG analysis guides the adjustment of stimulation parameters. The system can deliver tES pulses and perform EIT expands its functionality, allowing for targeted neuromodulation and impedance-based brain imaging. This integrated system revolutionizes EEG-based diagnostics, treatment, and research in neurology and neuroscience, offering a comprehensive and versatile tool for understanding and modulating brain function.

The invention discloses an obsessive-compulsive disorder diagnosis system based on brain multi-modal indexes, and belongs to the field of mental diseases. The problem of lack of a cross-modal feature fusion mechanism is solved. The system comprises an electroencephalogram signal acquisition unit used for acquiring an EEG signal of a testee under a preset stimulation normal form and executing preprocessing operation; the brain imaging data acquisition unit is used for synchronously acquiring brain structure imaging data and brain function imaging data of the testee; the multi-modal data fusion unit is used for extracting frequency band power spectrum density characteristics and event-related potential amplitude or incubation period characteristics from the EEG signals; performing standardization processing on the EEG features, the sMRI structural features and the fMRI functional features; integrating modal features by adopting a weighted average fusion algorithm; screening fused feature subsets through a recursive feature elimination method; and the diagnosis model unit is used for inputting the fusion feature vector into a trained SVM classification model and outputting an obsessive-compulsive disorder diagnosis result. Used in the medical field.

The invention relates to a brain fatigue real-time evaluation and prediction method based on a multi-modal signal and deep learning, and solves the problem that the existing brain fatigue monitoring technology mostly depends on static feature analysis of a single physiological signal. Firstly, EEG dynamic brain function network features and EEG time-frequency domain features are fused, and feature fusion driven by time-frequency-space multi-domain EEG features is achieved; furthermore, on the basis of weighted K-means clustering, PPG signals are automatically associated with a fatigue stage, the fatigue level is calibrated with the heart rate center value, subjective labeling is not needed, and objective re-calibration of the brain fatigue degree is achieved; and finally, realizing real-time evaluation and prediction of the brain fatigue degree based on the time sequence deep convolutional network model. The brain fatigue degree can be evaluated and predicted in real time with high precision, an early warning signal is sent to an operator with high brain fatigue degree, technical support is provided for brain fatigue real-time monitoring and warning, and the rate of misoperation accidents caused by brain fatigue is further reduced.

The invention discloses a brain heuristic multi-expert multi-mode emotion recognition method and system, equipment and a medium, and belongs to the technical field of artificial intelligence and biomedical signal processing. The method comprises the following steps: by simulating a brain function partitioning mechanism, dividing an electroencephalogram signal into a plurality of brain regions according to neuroanatomy prior, and designing a special expert network for each region; a global-local double-current encoder is adopted to cooperatively extract spatial-temporal characteristics of each brain region signal, and meanwhile, a multi-scale large-kernel convolution module is utilized to extract peripheral physiological signal characteristics; and finally, dynamically fusing multi-expert features through an adaptive routing network to realize sentiment classification. Expert load balancing and bifurcation regularization joint loss are introduced into the model in training, and effective cooperation and feature diversity of experts are ensured. According to the method, excellent recognition precision is obtained in practice, it is verified through interpretability analysis that the decision-making process conforms to neuroscience cognition, and a high-precision and high-reliability solution is provided for application of brain-computer interfaces, mental health monitoring and the like.

The invention provides a postoperative brain function state evaluation method and system based on multi-modal fusion, and relates to the technical field of postoperative brain function state evaluation.The method comprises the steps that multi-source original data of a postoperative patient is obtained, standardized multi-source data is obtained, and a postoperative immediate brain function baseline is established; a multi-modal feature subset is formed, and then the postoperative-preoperative function offset is obtained; generating a fusion feature vector and a confidentiality suppression index through fusion processing, and obtaining the confidentiality suppression index; a brain function state preliminary classification result and intervened newborn brain response characteristics are output through classification processing and intervention processing, and sound-light-electricity closed-loop intervention is driven; and finally, obtaining a postoperative brain function state evaluation result, and carrying out iterative calculation again until a final evaluation result which is consistent with clinical diagnosis and passes baseline matching is obtained. The method has the advantages that rapid and accurate evaluation and individualized intervention of the postoperative brain function state are achieved, and then the efficiency and quality of postoperative brain function management are improved.

The invention belongs to the technical field of brain function monitoring, and discloses a three-in-one non-invasive brain function monitoring system and method, a medium, equipment and application, and the monitoring method comprises the following steps: monitoring brain oxygen saturation of a critical patient; carrying out transcranial Doppler ultrasonic measurement on intracranial blood vessel blood flow velocity and ultrasonic measurement on the inner diameter of the optic nerve sheath; monitoring a quantitative electroencephalogram of the patient; various brain function monitoring series indexes and continuous changes of the indexes along with disease evolution are obtained, the pathogenic mechanism and pathogenesis of the indexes are found in time, and primary diseases and related complications of the primary diseases are treated. According to the invention, the brain oxygen saturation can avoid too low brain oxygen or low brain oxygen for a long time and too high brain oxygen due to oxygen utilization disorder; cerebral blood flow can be monitored through a transcranial Doppler (TCD) technology to find an optimal craniocerebral perfusion pressure (CPP), and a cerebral blood flow self-adjusting function is enabled to be in an optimal state; the electroencephalogram can avoid early warning of over-sedation or abnormal discharge and the like. Through organic combination of the three, noninvasive, real-time and dynamic monitoring of critical patients is realized.

The invention discloses a task state brain-computer interface training system for closed-loop transcranial magnetic stimulation, and particularly relates to the field of cognitive rehabilitation assistion.The task state brain-computer interface training system comprises the steps that multi-channel electroencephalogram signals and brain function activation records are synchronously collected in the rehabilitation training process, task segments are divided, and frequency domain energy characteristics of all channels are extracted; identifying a state transition candidate segment based on the frequency band energy distribution change; the method comprises the following steps: constructing a cross-channel phase synchronization matrix, and inputting a graph convolutional neural network to generate an activation topology vector reflecting a brain region cooperation mode; the system constructs an activation state recognizer by taking a topological vector and a brain function activation label as supervision data, and realizes real-time judgment of different function states. The recognition result drives a stimulation parameter recommendation module, a preset stimulation strategy mapping table is inquired according to the recognition state, recommended stimulation parameters are dynamically generated and transmitted to a magnetic stimulation execution interface, and personalized closed-loop nerve regulation and control are achieved.

The invention provides an acoustoelectric multi-mode nerve regulation and control method and system for improving the language reading ability, and the method comprises the steps: obtaining brain function activity baseline data based on a positioned abnormal link of the language reading ability of a pupil in combination with resting state and task state electroencephalograms; based on the abnormal links and the brain function activity baseline data, determining a TES target spot and a tFUS target region; starting a TES module based on the TES target spot and starting a tFUS module based on the tFUS target region; a tUS module is started to perform focused ultrasound stimulation intervention on the TES target spot and the tFUS target region and prompt the pupil to start a training task, and real-time electroencephalogram signals of the pupil in the intervention process and behavioral performance data of the pupil in the training task process are obtained; based on the real-time electroencephalogram signals and the behavioral performance data, the stimulation intensity of the tUS module on the TES target spot and the tFUS target region is adjusted. According to the invention, the synergistic interaction capability and the intervention accuracy are improved.

The invention provides a head-mounted diagnosis and treatment device, a brain-computer interface system and a brain activity monitoring intervention method. The head-mounted diagnosis and treatment device comprises a head-mounted structure, at least one active electrode, at least one pair of stimulation electrodes and a control module, the head-mounted structure comprises a head beam supporting part and a frontal lobe wearing part, the two ends of the frontal lobe wearing part are connected with the two opposite sides of the head beam supporting part respectively, and the active electrode is arranged on the inner side of the frontal lobe wearing part. The active electrode is arranged on the head beam supporting part and used for collecting electroencephalogram signals of a frontal lobe area contact part, the stimulation electrode is arranged on the inner side of the head beam supporting part and used for forming electrical stimulation on the frontal lobe area contact part, and the control module is electrically connected with the active electrode and the stimulation electrode and used for determining the brain function state according to the electroencephalogram signals collected by the active electrode and controlling the stimulation electrode according to the determined brain function state. Thus, electrical stimulation therapy is formed. According to the head-mounted diagnosis and treatment device, the brain-computer interface system and the brain activity monitoring intervention method, diagnosis and treatment can be integrated, and the head-mounted diagnosis and treatment device is easy to wear and portable.

The invention discloses an fNIRS adaptive feedback emotion cognition cooperative training device and method, and relates to the technical field of emotion disorder cognition training. The device comprises a data acquisition module used for acquiring an original dual-wavelength light intensity signal from a brain; the data processing module is used for performing multi-stage preprocessing on the original dual-wavelength light intensity signal; the neural feedback module is used for calculating a multi-dimensional evaluation index according to the hemoglobin concentration data and adjusting a difficulty level and a feedback threshold value of a training task by using a dynamic threshold value control algorithm; the training module is used for training participants based on the emotion stimulation task and the cognitive training task; and the evaluation module is used for performing training evaluation according to the comparison result of the comprehensive performance score and the feedback threshold. Brain function equipment is closely combined with cognitive training and emotion stimulation tasks, dynamic evaluation and real-time feedback are added for traditional tasks, and through the brain network analysis technology, accurate evaluation of the user training effect is achieved.

The invention discloses an anesthesia reviving room risk assessment system for postoperative nausea and vomiting prediction, relates to the technical field of postoperative nausea and vomiting prediction, and aims to perform multi-modal fusion on clinical features of a patient and brain function signals to realize basic probability calculation of postoperative nausea and vomiting risks, and introduce real-time dynamic physiological parameters as regulatory factors to realize postoperative nausea and vomiting risk prediction. And performing adaptive adjustment on the basic prediction result to generate a comprehensive risk probability value. The system can collect and process data in real time after a patient enters an anesthesia reviving room, and dynamic and personalized risk assessment is achieved. Compared with a traditional scoring method which only depends on static clinical features, the method has the advantages that tiny fluctuation of the postoperative physiological state of the patient can be captured, prediction precision and timeliness can be improved, clear decision support can be provided for nursing staff, and therefore preventive intervention measures are optimized, and the occurrence rate of complications is reduced.

The invention relates to the technical field of medical information processing, in particular to a multi-modal data integrated analysis system for screening children with autism, which comprises a data acquisition module, a data processing module, a feature extraction module, a feature fusion module, a data screening module, a data analysis module and a data output module, according to the system, behavior data, brain function data and brain structure data of autism children are collected, and after preprocessing and feature extraction are carried out, fusion processing is carried out by a feature fusion module. The feature fusion module comprises a multi-scale topological feature representation sub-module, a heterogeneous feature integration sub-module and a self-adaptive weight adjustment sub-module which are respectively used for realizing multi-scale representation, heterogeneous feature integration and dynamic weight adjustment of features; the feature fusion technology solves the problems of feature scale inconsistency, inter-modal semantic gap, feature importance dynamic change and the like in multi-modal data fusion, improves the autism screening accuracy and early recognition capability, and supports personalized screening and accurate intervention.

The invention discloses an auxiliary decision-making system of a transcranial magnetic stimulation mode, and the system comprises an obtaining module which is used for obtaining the personal feature information of a patient; the determination module is used for calling a preset evaluation model to calculate the brain function retention degree of the patient by adopting the personal characteristic information, and determining the treatment mode of the patient according to the brain function retention degree; wherein the preset evaluation model is a classification model constructed based on demographic characteristic parameters, behavioral evaluation characteristic parameters, neural electrophysiological characteristic parameters and brain image characteristic parameters as independent variables. The brain function retention degree is calculated by using the model, and the transcranial magnetic stimulation mode is determined according to the brain function retention degree without relying on subjective calculation of a doctor, so that the accuracy of determining the treatment mode can be improved; meanwhile, multiple evaluation, detection and calculation are not needed for processing, the processing time can be shortened, the processing efficiency and precision can be further improved, the diagnosis and subsequent treatment of the patient are prevented from being affected, and precise treatment is achieved.

The application discloses a mental health evaluation method and system based on brain area electric stimulation, relates to the technical field of medical and health information science, and creatively integrates brain area electric stimulation into mental health evaluation. The basic cognitive level and neural plasticity potential of a measured person are objectively quantified through comprehensive scores before and after brain area electric stimulation, so that corresponding suitable mental health evaluation test questions are selected based on the basic cognitive level and neural plasticity potential, and the precision, sensitivity of corresponding mental tests and the mental comfort of the measured person are significantly improved. A decay quantization model of a stimulating current from an electrode to the brain is given, the decay quantization model is used to accurately control the electric quantity, and the maximum beneficial effect is realized under the condition of guaranteeing health and safety. Historical data accumulated in the application process can be effectively used to solve medical problems, such as the influence of brain area electric stimulation on child neural development evaluation, early warning of mental illness and brain function reinforcement.

The invention discloses a personalized electroencephalogram emotion recognition method and system based on dynamic brain region division, and belongs to the technical field of electroencephalogram emotion recognition. The method comprises the steps that electroencephalogram signals are preprocessed, and an electroencephalogram signal matrix is constructed; on the basis of the preprocessed electroencephalogram signals, five-frequency-band differential entropy feature values are extracted, multi-head self-attention calculation is carried out in combination with rotation position coding, and channel feature representation with strong emotion intensity is output; on the basis of the constructed electroencephalogram signal matrix, dynamically dividing a brain function region by using a CNM algorithm, optimizing the structure of the brain function region by using a modularity value, and outputting brain region characteristics; and fusing the channel feature representation and the brain region features, predicting an emotion tag through a graph neural network, and outputting an emotion recognition result. According to the method, the individual brain function region can be dynamically divided, the function integration information of the brain region is effectively captured, and deep fusion of brain region function connection and channel emotion information is realized.

The invention provides a denoising method and system suitable for infant brain function magnetic resonance image data, a medium and a terminal, and relates to the field of infant brain function magnetic resonance image data denoising. The invention provides a denoising method suitable for infant brain function magnetic resonance image data. The denoising method comprises the following steps: acquiring an original function magnetic resonance image of an infant brain; performing noise component removal on the lightly preprocessed functional magnetic resonance image, wherein the noise component removal decomposes the lightly preprocessed functional magnetic resonance image into a time independent component and a space independent component to be processed respectively; noise time point removal is carried out on the deep preprocessing functional magnetic resonance image, reconstruction errors are adopted as an anomaly detection standard, time points possibly polluted by noise are recognized and removed, and the time points possibly polluted by noise are recognized and removed. According to the method, noise can be effectively removed, neural signals are accurately reserved, and the data quality and the denoising effect are improved.

The invention relates to a depression risk detection method based on self-adaptive multi-scale neighborhood perception fusion space-time diagram convolution, and aims to solve the problems that traditional depression diagnosis is subjective and existing electroencephalogram detection space-time feature modeling is insufficient. The method comprises the following steps: acquiring electroencephalogram signals of a testee, and extracting frequency spectrum features through a frequency domain feature extraction module; in combination with an adjacent matrix (independent of an electrode physical distance) which is adaptively learned and normalized in training, embedded features are generated through a graph convolutional neural network; and then capturing long-time-history dependence through a dynamic feature extraction module, realizing feature multi-level fusion by means of a multi-cascade multi-scale convolution module, finally integrating local and global features through an adaptive feature fusion module, and inputting a prediction layer to output a detection result. According to the method, electroencephalogram signal time-space domain joint modeling is achieved, brain region function connection dynamic and multi-scale neural features are effectively captured, information redundancy and key feature loss are avoided, classification accuracy and generalization ability are remarkably improved, depression-related key brain region and function network features can be revealed, and the method has high interpretability and potential clinical application value.

The invention relates to a nutritional composition comprising non-digestible oligosaccharides for infants and / or young children. The invention further relates to the use of said nutritional composition for improving the brain function, in particular social recognition.

The invention discloses a personalized MCI electrical stimulation intervention method based on a dynamic brain network, and belongs to the technical field of electrical stimulation target determination. Individual heterogeneity of a brain function network of a patient is considered, specific frequency band selection is carried out on electroencephalogram data sets of a normal group and a cognitive impairment group, an effective data set is constructed, and the effective data set is determined. The method comprises the following steps: respectively constructing average dynamic brain network connection of total test times of a normal group and a cognitive impairment group based on an adaptive method of dynamic time-varying weight optimization of reaction time, and then determining a main abnormal frequency band through the difference of the average dynamic brain network connection of the two groups; and then target points are determined for the core nodes connected with the average dynamic brain network of the main abnormal frequency band, and compared with traditional single-frequency-band analysis, the application proposes that the effective data set is constructed by the specific frequency band to determine the main abnormal frequency band; in addition, stimulation targets are positioned according to the core nodes connected with the average dynamic brain network of the abnormal frequency band, and the reliability of target selection is improved.

The invention relates to the technical field of brain data monitoring, and discloses a brain function intelligent monitoring method and system based on multi-modal data, and the method comprises the steps: collecting brain multi-modal data of a user through a head wearing device; determining a brain dynamic graph object of the user according to the brain multi-modal data, and performing feature fusion operation on the brain dynamic graph object according to the brain dynamic graph object to obtain a brain multi-modal feature fusion result of the user; and determining brain function monitoring data of the user according to a brain multi-modal feature fusion result. It can be seen that the brain function monitoring data of the user can be determined by collecting and intelligently analyzing the brain multi-modal data of the user through the portable head wearable device, so that dependence on large-scale brain monitoring equipment is reduced, and the user experience is improved. And the dynamic change condition of the brain function of the user is accurately reflected through the multi-modal data, so that the monitoring accuracy of the brain function data of the user is improved.

The invention relates to the technical field of medical artificial intelligence, and discloses a brain disease prediction method based on double encoders and a diffusion model, and the method comprises the steps: carrying out the preprocessing of a functional magnetic resonance image, and constructing a brain function network; data enhancement of semantic preservation is achieved through a diffusion model, a dual random matrix and a cosine scheduling strategy are adopted in the noise adding process, and a GraphTransform neural network containing global topological features is utilized in the denoising process; the spatial features of the brain network and the time dynamic features of the BOLD signals are respectively extracted by using double encoders; designing a triple contrast learning mechanism to optimize cross-dimension feature interaction; and finally migrating to a downstream classification task to realize disease prediction. Small sample overfitting is relieved through diffusion enhancement, and the diagnosis reliability is improved; fusing spatial-temporal characteristics to assist multi-dimensional pathological analysis; and the cross-site adaptability of the model is enhanced, and collaborative analysis of multi-center heterogeneous data is supported. The method is suitable for auxiliary diagnosis of cerebral diseases such as infantile autism.

The invention relates to the technical field of brain function network analysis, in particular to a post-stroke aphasia patient fMRI focal recognition method based on deep learning. The method comprises the following steps of: acquiring a region consistency value for geometric structures of matched local regions of two adjacent fMRI images in an fMRI image sequence, similarity of gray distribution and an image matching degree, and dividing a same time window by using the region consistency value; and obtaining a functional connection coefficient according to the difference of the region consistent values of the fMRI images in two adjacent time windows and the difference of the geometric structure features and the gray level distribution features of the local regions, determining the convolution kernel size of deep learning of the fMRI images in the time windows by using the functional connection coefficient, and performing focal recognition on the fMRI images in the time windows. According to the method, the function connection coefficient presenting the cooperation degree of the brain region activity modes in the adjacent time windows is analyzed, the convolution kernel size of the time windows is determined in a self-adaptive mode, and the recognition effect on the local focus of the patient is improved.

The invention provides an EEG function connection prediction method based on fMRI deep cross-modal representation learning, and relates to the crossing field of neuroiconography and artificial intelligence. According to the method, a sample pair is constructed through time alignment of EEG and a blood oxygen level dependent signal, an end-to-end deep neural network architecture composed of a time projection unit, a cross-attention fusion encoder and a connection synthesis head is designed, and direct mapping from a BOLD signal to EEG function connection is achieved. The model adopts a composite loss function, and gives consideration to the consistency of the prediction connection matrix with the real EEG function connection in numerical precision and topological mode, thereby achieving the high-fidelity reconstruction of the EEG function connection map at the frequency domain level. According to the invention, the topological structure stability of the brain function network can be effectively maintained. Under the conditions of incomplete EEG data, serious noise interference or complete loss, the frequency domain function connection characteristics can still be stably recovered, and a reliable analysis substitution path is provided for brain function network research.