155 results about "Functional magnetic resonance imaging" patented technology

The invention discloses a multi-modal feature combined depression auxiliary diagnosis system. The system comprises a sampling unit which is used for constructing a multi-modal depression data set by acquiring a depression screening scale, an electroencephalogram, a magnetoencephalogram and functional magnetic resonance imaging based on acquisition equipment; the feature extraction unit is used for extracting multi-modal brain features based on the depression data set, and the multi-modal brain features comprise power spectral density obtained by electroencephalogram signals, event-related potential, micro-state, prefrontal lobe gamma frequency band power spectral density obtained by magnetoencephalogram and event-related magnetic field; gray matter volume and resting state functional connection density are obtained through functional magnetic resonance imaging; a data preprocessing unit; the diagnosis model unit is used for constructing a multi-modal depression diagnosis model and training the model on the basis of the multi-modal brain features in combination with a fusion strategy; and an analysis and prediction unit. The extracted features are comprehensive and reasonable, the defect of each mode is overcome by the feature fusion method, and the fused features are advanced.

The invention relates to an infantile autism classification method based on a double-branch functional topological graph neural network. The infantile autism classification method can realize the classification of the infantile autism by using functional magnetic resonance imaging data. According to the provided autism classification network, long-distance connection and short-distance connection are divided based on the shortest path between brain intervals, then an exponential decay mask is introduced through a functional topological graph Transform branch to adjust attention weight and accurately extract long-distance dependency features, a graph isomorphic network in the other branch is subjected to multiple neighborhood aggregation operations, short-distance dependency features are captured, and the short-distance dependency features are extracted. According to the method, multi-scale dependence of the brain network is extracted in parallel through a double-branch structure, information redundancy is reduced by means of a topology perception attention mechanism, and the adaptive ability of the model to the heterogeneous brain network is improved by using the adaptive fusion module, so that multi-scale dependence of the heterogeneous brain network is balanced in a self-adaptive manner. The classification accuracy is remarkably superior to that of an existing mainstream method, objective and efficient technical support is provided for autism diagnosis, and high interpretability is achieved.

The invention discloses a brain function magnetic resonance imaging data analysis method based on a contrast graph neural network, and the method comprises the steps: firstly carrying out the data enhancement of a brain function connection graph, and simulating the heterogeneity of brain function magnetic resonance data, so as to improve the diversity of a data set; secondly, the hidden space embedding features of the brain function connection diagram are efficiently learned by fusing a double-Hough-Laplacian diagram convolutional network and a contraction incentive mechanism; the embedded features are mapped to a group of prototype vectors, and prototype allocation codes corresponding to the embedded features are calculated by adopting a Sinkhorn-Knopp algorithm; performing exchange optimization on prototype codes between different enhanced brain connection diagrams of the same subject through a contrast learning strategy, and compelling codes of homologous subjects to be aligned at the minimum cost in combination with a cross entropy loss function; and finally, applying the pre-training model to a functional magnetic resonance imaging data set of the Alzheimer's disease, and carrying out interpretability analysis on learning features to improve the classification efficiency and pathological analysis of the Alzheimer's disease under a limited tag condition.

The invention discloses a brain disease judgment system and method based on fusion of a time sequence affinity graph and a multi-modal network. The method belongs to the technical field of artificial intelligence and medical image crossing. The invention provides a brain disease judgment system based on a time sequence affinity graph fused multi-modal network. The brain disease judgment system comprises a phenotypic feature reconstruction module, a feature extraction module, an affinity graph construction, processing and multi-modal fusion module, a loss construction module and a classification module. The method is used for distinguishing a brain disease patient group from a health control group, and is suitable for a medical image auxiliary diagnosis system, a multi-center brain disease screening platform and an individualized disease risk assessment tool. The core of the method is to solve the problems of insufficient utilization of single-mode information, poor multi-center data robustness and redundant graph structure noise in traditional brain disease classification through time sequence affinity graph construction and multi-mode feature fusion, ROI time sequence data and phenotypic data derived by resting state functional magnetic resonance imaging can be processed, and the accuracy of brain disease classification is improved. The method has application prospects in clinical transformation and multi-center collaborative research.

The invention provides a multi-modal brain network analysis system and method, electronic equipment and a storage medium, and relates to the field of brain networks, the multi-modal brain network analysis system comprises an acquisition module, a preprocessing module, a first processing module, a second processing module, a fusion convolution module and an output module; the acquisition module is used for acquiring functional magnetic resonance imaging data and diffusion tensor imaging data; the preprocessing module is used for processing to obtain an edge time sequence and an initial structure brain network; the first processing module is used for constructing a dynamic effect brain network; the second processing module is used for constructing an edge center structure brain network; and the fusion convolution module is used for carrying out structure fusion and directed space-time diagram convolution. According to the method, structure fusion and directed space-time diagram convolution are carried out on the dynamic effect brain network and the edge center structure brain network based on a self-attention mechanism, target information representing the brain nerve state is obtained, accurate cognitive load evaluation is carried out, and the limitation of a traditional node center method is solved.

The invention provides a multi-modal brain network classification method based on feature decoupling and dynamic graph construction, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining functional magnetic resonance imaging data and structural magnetic resonance imaging data of a to-be-detected person, and obtaining multi-modal brain network data according to the functional magnetic resonance imaging data and the structural magnetic resonance imaging data, and processing the multi-modal brain network data based on a preset multi-modal brain network classification model to obtain the brain network state of the to-be-tested person. According to the method, by combining functional magnetic resonance imaging (fMRI) and structural magnetic resonance imaging (sMRI) data, the advantages of the two modes can be fully utilized, so that the accuracy and comprehensiveness of brain network state classification are improved, a dynamic graph attention module based on a graph attention network (GAT) can capture the dynamic change characteristics of a brain network, dynamic graph representation is constructed, and the classification accuracy of the brain network state is improved. And the sensitivity of the model to the dynamic connection relationship between brain regions is enhanced.

The invention relates to the technical field of brain anomaly detection and artificial intelligence auxiliary diagnosis, in particular to a graph convolutional network brain disease diagnosis method based on sub-graph sampling and multi-feature fusion, and aims to improve the accuracy of brain disease diagnosis. The method comprises the following steps: obtaining resting state functional magnetic resonance imaging data, preprocessing the data, and constructing a brain function connection diagram; and the brain function connection graph represents a brain interval collaborative activation relationship in a graph structure. Subgraph sampling is carried out based on function module division and node degree sorting, and an initial subgraph set is generated; and performing optimization selection on the initial sub-graph set by utilizing reinforcement learning to obtain an optimal sub-graph, introducing a node attention mechanism into the optimal sub-graph, screening key nodes based on attention scores, and generating a discriminant sub-graph. And extracting and fusing position features, neighborhood features and structural features of the discriminant subgraphs, and performing brain disease diagnosis based on the fused features.

The invention discloses an autism detection method based on multi-mode collaborative embedding, and belongs to the technical field of medical image analysis and artificial intelligence. The method comprises the following steps: firstly, obtaining resting state functional magnetic resonance imaging data and non-imaging data of a subject; a Markov transition field is utilized to encode the time sequence into an image so as to retain dynamic features, and feature extraction is carried out through an efficient multi-scale attention module; then realizing effective fusion and semantic alignment of multi-modal information by adopting a three-level fusion architecture and a joint loss function; and then adaptively constructing a graph structure based on the fusion features, dynamically learning a node relationship by using a graph attention network, and completing a classification decision. According to the method, the defects of a traditional method in the aspects of dynamic feature modeling, multi-modal fusion and heterogeneous graph structure processing are effectively overcome, the autism detection accuracy and robustness are remarkably improved, and a reliable tool is provided for clinical intelligent diagnosis.

The invention discloses a coupling tensor decomposition method based on sparse low-rank constraint, which is used for characteristic decomposition of a multi-frequency dynamic function network connection tensor in resting state function magnetic resonance imaging data. According to the algorithm, on the basis of the traditional coupling canonical factorization (CCPD), an optimization model of sparse and low-rank constraint is constructed, and the sparse and low-rank constraint optimization model is constructed by the algorithm. On the spatial connectivity dimension, redundant function connection is reduced through an L1 sparse penalty term, and the spatial specificity of the key brain network is enhanced; and in time and frequency band dimensions, low-rank regularization constraint is adopted to improve discrimination of cross-subject time sequence characteristics. Generally speaking, the method can effectively extract connectivity characteristics with statistical significance and time states of different frequency bands from dynamic function network connection tensors of multiple frequency bands, thereby effectively identifying functional connection heterogeneity characteristics between schizophrenia patients and healthy control groups.

The invention discloses a federated learning-based functional magnetic resonance imaging data classification method. The method comprises the following steps of obtaining multi-site resting state functional magnetic resonance imaging data and performing preprocessing; dividing a brain region and extracting a time sequence; constructing a brain function connection network by adopting a Pearson correlation coefficient; a graph sampling aggregation neural network fusing a residual connection structure and a multi-head attention mechanism is trained at each site, and shallow brain region features are reserved; adopting a linear kernel maximum mean value difference loss function to align the brain region node feature distribution of each site, and minimizing the data distribution difference between the sites; and carrying out classification training by adopting cross validation, aggregating parameters of each station through federal weighting, and evaluating classification performance indexes to obtain a final classification prediction result. According to the method, the graph sampling aggregation neural network and the cross-network layer feature alignment method are combined, the heterogeneity problem of multi-site functional magnetic resonance imaging data can be solved, and therefore generalization and classification performance of a global model are improved.

A multimodal system that measures one or more parameters relating to an individual's response to the presentation of a scent / odor stimuli and providing an evaluation of olfactory function in the individual based on the measure(s). In some instances, the system can be used to determine if the olfactory deficit in the individual is due to underlying olfactory pathology or non-olfactory causes. The system can also further differentiate non-olfactory causes to knowledge or memory-based deficiencies in the individual. The system can include at least two scents from at least two distinct scent families which are presented an individual. The system can also include a questionnaire that evaluates scent classification, description, and discrimination ability of an individual in connection with the presented scents. The system can also include the measurement of neurophysiological activity through an electroencephalogram (EEG) or functional magnetic resonance imaging (fMRI), in response to scent presentation.

The invention provides a schizophrenia core epicenter region identification method based on a multi-mode nerve image, and belongs to the technical field of medical image analysis and neuropsychiatric disease diagnosis. According to the method, a macroscopic-mesoscopic-microscopic covariant network model is established by integrating multi-modal nerve image data (including structural magnetic resonance, functional magnetic resonance imaging and diffusion tensor imaging) and mesoscopic-microscopic scale data (gene expression atlas, neurotransmitter distribution and cell construction characteristics). The method comprises the following steps: firstly, constructing various connection networks, then identifying a core epicenter region with abnormal grey matter thickness based on a graph theory algorithm, carrying out cross-scale matching on the epicenter region and gene expression, neurotransmitter and cell structure characteristics by utilizing spatial correlation analysis, and analyzing a diffusion path of the epicenter region through connection omics. By fusing the multi-scale biomarkers, the limitation of single modal analysis is broken through, and systematic analysis of the schizophrenia pathological network is realized.

The invention discloses an autism spectrum disorder classification method and system based on evidence decision fusion. The method comprises the following steps: creating an autism multi-modal data set; an autism classification model based on evidence decision fusion is constructed and comprises a data preprocessing module, an evidence extraction module, a category credibility and classification result uncertainty estimation module and a decision fusion module. Training a model by using samples in the autism multi-modal data set; and performing autism classification on a newly input subject sample by using the trained model. According to the method, key evidences of three modes of T1 weighted imaging, diffusion tensor imaging and functional magnetic resonance imaging are comprehensively utilized, and a reliable and credible classification decision framework is constructed according to a Dempster combination rule, so that the accuracy and robustness of autism classification can be improved, overall classification uncertainty evaluation of a final decision can be given, and the accuracy and robustness of autism classification are improved. And better model interpretability and decision credibility are provided.

The invention discloses an image processing method, device and equipment based on multi-modal neural image fusion, and relates to the technical field of image recognition. The method comprises the following steps: calculating a functional connection matrix among regions in brain nerves based on functional magnetic resonance imaging data; extracting a first feature of the structural magnetic resonance imaging data and a second feature of the functional link matrix; obtaining a first feature vector and a second feature vector corresponding to the first feature and the second feature, and adding the first feature vector and the second feature vector to obtain a common feature vector; and performing feature enhancement on the first feature vector and the second feature vector through the common feature vector, splicing the enhanced first feature vector and the enhanced second feature vector to obtain a spliced feature vector, and inputting the spliced feature vector into a pre-constructed feature classification network to obtain a classification result. According to the scheme, complementarity among different modal data can be enhanced, so that diagnosis of depression is better assisted.

Disclosed is a method for determining a treatment to be applied to a brain of a patient, such as determining a transcranial magnetic stimulation (TMS) treatment location(s) and treatment strength to be applied to a brain of a patient suffering from depression. A search of the brain, based on coarse-scale parcellations, treatment metric and one or more anatomical metrics, is conducted to select the treatment location(s). In one embodiment, this involves progressively varying a value of a parameter associated with a gyrus threshold and for each value, identifying candidate treatment locations based on a functional connectivity of a subgenual anterior cingulate cortex during resting-state functional magnetic resonance imaging (fMRI) recorded over a plurality of multimodal brain imaging sessions, and selecting as the respective treatment location the candidate location that is closest to all other candidate treatment locations.

Methods, systems, and devices, including computer programs encoded on a computer storage medium are provided for optimizing neurostimulation therapy for treatment of neurological and neurodegenerative diseases. Joint dynamic causal modeling and biophysics modeling are used for optimization of the stimulation targets and parameters. In particular, methods of performing neuromodulation to suppress b-band oscillations in the brain of a subject are provided.

The invention discloses an emotion intervention system and device based on a reminiscence therapy, and belongs to the technical field of mental health intervention and intelligent medical equipment. The device comprises a stimulation presentation subsystem, a physiological signal collection subsystem, a behavior response recognition subsystem, a neural response modeling subsystem, an intervention effect evaluation subsystem and a central control and feedback regulation subsystem, a closed-loop intervention framework is formed through real-time communication bus interconnection, and by presenting an individualized life event image set, multi-modal physiological signals and behavior responses are synchronously collected, so that the intervention effect is evaluated. A neural response model is constructed based on a brain region abnormal activation mode disclosed by functional magnetic resonance imaging, the activation level of a target brain region is predicted, the intervention effect is evaluated by integrating multiple indexes, and a central control system dynamically adjusts a stimulation sequence according to an evaluation result and a neural response predicted value by applying a hierarchical reinforcement learning algorithm. Personalized and precise emotion intervention aiming at the subclinical depression state is realized, and the intervention effect and the nerve regulation pertinence are effectively improved.

The application provides a multi-modal brain network classification method based on feature decoupling and dynamic graph construction, and relates to the technical field of artificial intelligence, and the method comprises the following steps: acquiring functional magnetic resonance imaging (fMRI) data and structural magnetic resonance imaging (sMRI) data of a to-be-tested person, obtaining multi-modal brain network data according to the fMRI data and the sMRI data, processing the multi-modal brain network data based on a preset multi-modal brain network classification model, and obtaining the brain network state of the to-be-tested person. The application can fully utilize the advantages of the two modalities by combining the fMRI data and the sMRI data, thereby improving the accuracy and comprehensiveness of the brain network state classification, and the dynamic graph attention module based on a graph attention network (GAT) can capture the dynamic change characteristics of the brain network, construct a dynamic graph representation, and enhance the sensitivity of the model to the dynamic connection relationship between brain regions.

The invention relates to the technical field of brain disease target positioning, in particular to a brain disease target positioning system and method based on functional magnetic resonance imaging. The method comprises the following steps: calculating a whole brain function connection map of an individual patient based on 4D rs-fMRI data, calculating a Z score of each voxel function connection strength in the whole brain function connection map of the individual patient, and screening abnormal voxels based on a preset threshold value; aggregating the abnormal voxels through a clustering algorithm so as to determine a cluster; the therapeutic effect scores of the multiple clusters can be predicted through the machine learning model, the clusters and the candidate therapeutic targets can be screened based on the therapeutic effect scores, the positions of the candidate targets can be accurately determined, and multiple selections of the candidate targets can be provided; the target spot is converted into an accurate target spot suitable for an individual through a nonlinear registration algorithm; and determining a target positioning state according to the deviation between the accurate target and the actual treatment target, and adjusting corresponding parameters based on the target positioning state. The invention provides a plurality of targets and effectively determines the positions of the targets.

Provided is a bed that enables noninvasive and multipurpose functional MRI imaging.　Provided is a bed system for functional MRI, the bed system having a helmet for immobilizing the head of an animal using a U-shaped appliance.

The invention discloses a dynamic brain network analysis method guided by neural heterogeneity, and is suitable for the technical field of brain image processing and recognition. The method comprises the steps that functional magnetic resonance imaging data are acquired and preprocessed, an overlapped sliding window is used for dividing the functional magnetic resonance imaging data to construct a dynamic functional brain network, and then the dynamic functional brain network is decoupled into a topological consistency network and a time trend network which conform to brain activities; capturing space and time heterogeneity weights of different brain regions in the brain based on a topological consistency network and a time trend network, and identifying key nodes for driving brain network recombination; further weighting the topology consistency network and the time trend network to obtain a heterogeneity dynamic function brain network; propagation of neural information in a time dimension is simulated based on time propagation graph convolution operation, and spatial-temporal features of brain images are extracted from a heterogeneous dynamic function brain network; and finally, inputting the obtained spatial-temporal characteristics of the brain image into a multi-layer perceptron to predict the data category of the brain image to be recognized, and analyzing the influence of the brain disease image characteristics on the spatial-temporal heterogeneity of the brain region to complete the dynamic brain network analysis guided by the neural heterogeneity.

The embodiment of the invention provides an electroencephalogram emotion recognition method based on a graph neural network and federal learning. The method is applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring a resting state functional magnetic resonance imaging time sequence and non-image personalized data; preprocessing the resting state functional magnetic resonance imaging time sequence, and constructing a dynamic graph sequence for the preprocessed resting state functional magnetic resonance imaging time sequence based on a plurality of preset brain maps by adopting a sliding window technology; inputting the dynamic graph sequence into a shared feature layer for feature extraction to obtain a space-time shared feature vector; inputting the non-image personalized data into an independent personalized layer for feature extraction to obtain a personalized feature vector; performing feature fusion processing on the space-time sharing feature vector and the personalized feature vector to obtain a fused feature; and the fused features are mapped into the electroencephalogram emotion category probability through the classifier, an electroencephalogram emotion recognition result is obtained, and the electroencephalogram emotion recognition accuracy and the generalization ability of the model are improved.

This application discloses a spatial segmentation radiotherapy guidance method and system based on target immune region activation, relating to the field of digital electrophysiological data processing. The method achieves three-dimensional visualization of metabolic activity and immune response through multimodal fusion of spectral CT / functional magnetic resonance imaging and ResNet radiomics feature extraction, which significantly improves the recognition accuracy compared to anatomical target volume. Furthermore, the immune thermogram generated by the XGBoost model and SAMed-2 segmentation markers can accurately locate highly immune-responsive voxels, and the Monte Carlo simulation gradient dose distribution ensures that the dose intensity in the immune-activated region is several times that of the conventional region while protecting the inactive region. The XGBoost model training of this application can dynamically adjust the dose hotspot according to the patient's specific immune characteristics, so that the radiation can accurately and sufficiently reach the region in the tumor area that is most capable of activating the immune system.

The invention belongs to the technical field of medical image processing, and provides an improved functional magnetic resonance imaging data phase synchronization dynamic analysis method. According to the improved functional magnetic resonance imaging data phase synchronization dynamic analysis method, a phase synchronization matrix main feature vector extraction module and a spectral clustering module are included, a similar graph between sample points is constructed through absolute values of correlation coefficients, the positive and negative problems in the feature value decomposition step are eliminated, and therefore clustering errors in an original method are eliminated; in addition, a feature vector updating step is added when the class center is defined, so that a more accurate brain phase synchronization state is obtained.

A magnetic resonance-compatible programmable synchronized photostimulation system includes a main control unit and a stimulation execution unit. The main control unit is located outside the magnetic resonance scanning room, while the stimulation execution unit is located inside the scanning room and consists only of an LED light panel shielded by a magnetic field. The main control unit is configured to: receive photostimulation sequence parameters configured by the user in a modular manner through a graphical interface; receive a scan trigger signal from the magnetic resonance equipment, and automatically control the start and stop of the photostimulation sequence based on the trigger signal using hardware triggering; thereby achieving high-precision time synchronization between photostimulation and magnetic resonance scanning while maintaining magnetic resonance compatibility. This invention solves electromagnetic compatibility issues through a split architecture, achieves millisecond-level synchronization accuracy through hardware TTL triggering, and enhances experimental paradigm flexibility through modular programmable sequences, making it suitable for neuroscience research fields such as functional magnetic resonance imaging.

The present invention provides a method for identifying mental illness based on pulse structure-function brain network coupling, which belongs to the field of intelligent auxiliary medical diagnosis technology and effectively solves the technical problem of the neurobiological mechanism between structural connection and functional connection that is often ignored in the traditional diagnosis of mental illness. Its technical solution is: first, functional and structural brain networks are extracted from functional magnetic resonance imaging and diffusion tensor imaging; then, the information feature maps of these two brain networks are extracted through BrainNetCNN; then a pulse coupled neural network is constructed to learn the brain structure-function coupling mechanism, so as to derive the pulse structure-function coupling; finally, the obtained coupling information is input into the classification layer to obtain the disease identification result, and the cross-entropy loss function is used to train and optimize the result. The beneficial effects of the present invention are: the present invention helps to deeply understand the neural mechanism of mental illness and has a wide range of application prospects in clinical applications.

The invention provides an image processing method and an image processing device, and relates to the field of computers. According to the method, feature extraction is carried out on multi-modal image data of functional magnetic resonance image data and structural magnetic resonance image data of a brain, feature fusion is carried out, classification is carried out based on fused features, an image recognition result is obtained, feature extraction is carried out based on a channel attention and space attention decoupling method, and the recognition accuracy is improved. According to the method, the image features can be extracted more accurately, the accuracy of the image recognition result is further improved, and diagnosis of mental diseases such as bidirectional affective disorder and Alzheimer's disease can be better assisted.

The embodiment of the application discloses a kind of Alzheimer's disease classification and key brain area determination method based on random attention and counterfactual contrast learning, it is related to medical image analysis technical field;Alzheimer's disease classification and key brain area are determined by the functional magnetic resonance imaging data of subject;The accuracy, overall performance and stability of the graph convolution network model are significantly better than the existing GNN baseline model in the AD vs.NC task of ADNI real data set;The output key brain area node is highly consistent with clinical medicine priori;Specific brain area and functional connection leading to abnormal classification can be intuitively presented;It conforms to the real pathological mechanism, and whether the verification model in the same framework is really highly dependent on the selected key brain area structure, so that the explanation result can be accepted by clinician in the medical scene with extremely high safety requirement.

A method for identifying brain-state dependent functional areas of unitary pooled activity (FAUPAs) using a statistical model that does not require a priori knowledge of the activity-induced ideal response signal time course is provided. A system for identifying a functional network in a brain of a living object includes a FAUPA identifier configured to identify FAUPAs by analyzing a plurality of images of the brain over a predetermined period. The plurality of images include a plurality of voxels, and the FAUPA identifier analyzes each voxel of the plurality of voxels in relation to one or more surrounding voxels of each voxel until each voxel of the plurality of voxels is evaluated. A brain network identification module configured to construct the functional network based on the identified FAUPAs that are functionally connected. A display module configured to display images of the brain depicting the FAUPAs included in the functional network.

The invention relates to a brain region positioning magnetic resonance data processing method based on nerve stimulation target optimization, and the method comprises the steps: obtaining the functional magnetic resonance imaging data of a subject, extracting the signal response sequence of each brain region before and after nerve stimulation through time domain analysis, and determining the brain region with a characteristic response time window according to the response delay difference; the method comprises the following steps: detecting delayed oscillation characteristics in a cortex signal based on a neural feedback relationship between a cortex region and a deep structure to determine potential regulatory pathways of thalamus and other deep brain regions so as to position indirect stimulation targets which cannot be directly imaged; decoupling analysis of cerebral blood flow signals and nerve activation signals is carried out on the candidate brain area, and when it is detected that blood flow fluctuation exists in the area but synchronous nerve activation is lacked, it is judged that the area is a pseudo activation area and excluded; according to the method, the time delay correlation degree between the stimulation sequence and the brain region signal is calculated through the improved cross-correlation integral function, and the high-frequency noise interference is suppressed in combination with the nonlinear attenuation item, so that the estimation of the brain region response delay is more stable and repeatable.