159 results about "Resting state fMRI" patented technology

The invention discloses a unit, a device and a method for simulating biological neuron and neuronal synapsis on the basis of chalcogenide compounds. The unit comprises a first electrode layer, a function material layer and a second electrode layer. During the neuron simulation, a device receives the stimulation of one or a plurality of electric pulses, the resistance of the function material is changed into the low resistance state from the high resistance state, the simulated neuron is changed into an excitation state from a resting state, and the threshold value excitation and energy accumulation excitation functions are realized. During the neuronal synapsis simulation, the electric conductance of the function material layer of the device can be gradually changed according to input signals, and the synapsis weight regulating function is realized, the ynapsis weight is changed according to time differences of signals input at two ends, and the STDP (spike timing dependent plasticity) function of synapsis is realized. The basic device forming the artificial neural network can be provided.

The invention discloses an early and late mild cognitive impairment classification method and device based on brain function network characteristics and belongs to the medical image processing technology field. According to the method, firstly, sample data is pre-processed, multiple brain region time series are extracted, the brain function network is constructed by utilizing a correlation coefficient between the Pearson's correlation calculation brain area time series, and brain network parameters are calculated; secondly, the stepwise analysis method is utilized to extract characteristics, abinary classifier is trained, corresponding characteristic vectors are extracted from to-be-classified resting state functional magnetic resonance data and are inputted to the trained binary classifier, and the medical image classification result is acquired. Compared with a method in the prior art, the method is advantaged in that classification accuracy, sensitivity and specificity are better.

The invention provides a method and system for extracting abnormal parameters of the brain. The method comprises the following steps: carrying out magnetic resonance diffusion tensor imaging and resting-state functional imaging scanning for the brain, and obtaining diffusion tensor data and resting-state functional data of the brain; building the structural connection network of the brain according to the diffusion tensor data; building the functional connection network of the brain according to the resting-state functional data; coupling the structural connection network with the functional connection network to obtain network coupling indexes; arraying and verifying the network coupling indexes and normal network coupling indexes, and extracting the parameters which are different from the normal network coupling indexes in the network coupling indexes. The method and the system are used for conveniently extracting the parameters sensitively reflecting the abnormity of the structure and the function of the brain.

The invention discloses a blood sample level dependence functional magnetic resonance signal fluctuating frequency clustering analysis method which is characterized by comprising the following steps: step 1, pretreatment analysis is performed on resting state data collected from a magnetic resonance machine; step 2: Hilbert-Huang transform is adopted for decomposing the pretreated signals, so as to obtain intrinsic mode functions with different frequencies; the Hilbert-Huang transform comprises two steps of empirical mode decomposition and Hilbert transform; step 3, quantitative indexes of different frequency components are calculated; step 4, clustering of different voxels in a brain is performed by taking the quantitative indexes in the step 3, as a basis for classification and statistical analysis to the different indexes is performed. The method solves the problem of insufficiency of data drive frequency analysis methods in the fMRI research field, an iconography analytical method is provided for basic science and clinical researches and meanwhile new quantitative indexes and physiological markers are provided for clinical disease diagnosis and pathological mechanisms.

The invention provides a feedback system combining electroencephalogram and functional magnetic resonance signals, which jointly collects and analyzes electroencephalogram and functional magnetic resonance signals, extracts spatial-temporal characteristics of brain specific activity states reflected by the two signals simultaneously, and applies a multi-mode signal to nervous feedback regulation.The system comprises a data detection and pre-processing module, a spatial-temporal characteristic extraction module and a display and feedback module, wherein the data detection and pre-processing module detects the synchronously collected electroencephalogram and functional magnetic resonance signals on line, marks synchronous start points of the two signals, and performs pre-processing respectively; the spatial-temporal characteristic extraction module respectively extracts different electroencephalogram characteristics by aiming at resting state data and task state data, and performs statistic modeling analysis on the electroencephalogram characteristics together with functional magnetic resonance data to extract brain functional regions with the spatial-temporal characteristics; and different characteristics reflecting the same state of the brain are independently or jointly fed back. The system has important application value in clinical rehabilitation, brain-machine interface and other aspects.

The invention relates to an automatic discrimination analysis method of mild cognitive impairment based on a support vector machine. According to the automatic discrimination analysis method of mild cognitive impairment based on the support vector machine, multiple detection index data including aresting state fMRI, structure MRI and neuropsychological examination data are used, and methods of Functional Connectivity (FC) analysis, Voxel-based Morphometry (VBM) analysis and Fiber Tractography are adopted to extract a resting state functional connectivity characteristic, a gray matter structure characteristic and a white matter fiber connectivity characteristic, reduction is conducted on the characteristics based on a rough set approach, and finally classifier construction is conducted on multimodal MRI data based on the SVM method, so that automatic discrimination analysis of mild cognitive impairment is achieved, diagnosis accuracy of mild cognitive impairment is improved, and the accuracy rate of the diagnosis in experimental data reaches more than 90 percent. The method can be applied to practical clinical diagnosis.

The invention discloses a pain evaluation system based on magnetic resonance resting state functional imaging, and the pain evaluation system comprises a pain characteristic specimen library for storing history resting state functional imaging attributes of specimens of patients with different pain levels and pain causes and control normal persons; a functional imaging unit for obtaining resting state functional imaging data from brains of new individual patients suffering from chronic pain and transmitting to an attribute extraction unit; the attribute extraction unit for extracting resting state functional image attributes of new individuals from the resting state functional imaging data; and a pain evaluation unit for comparing the resting state functional image attributes of new individuals and the history resting state functional image attributes in the pain characteristic specimen library, and evaluating pain levels and/or pain causes of new individuals. Hereby, the pain evaluation system disclosed by the invention can objectively and accurately evaluate pain levels and/or pain causes through objective image data of human brains.

The invention belongs to the field of medical image processing and application, relates to a functional image analysis method for predicting the consciousness recovery of a patient with consciousnessdisorder, and particularly relates to a magnetic resonance detection data analysis method based on the function of the patient with consciousness disorder. The data analysis method based on machine learning comprises the steps that based on resting state functional magnetic resonance data (RS-fMRI), a weighted group sparse algorithm is used to construct a human brain functional connection matrix;from the matrix, a sparse representation feature selection method is used to select functional connection features with high contribution to classification for automatic prediction; and a linear support vector machine is used to construct a prediction model to acquire the final prediction result of consciousness recovery. The method is useful as a reference for predicting whether a long-term unconscious patient with brain damage can recover consciousness.

The invention discloses a brain function network classification method based on a variational autoencoder. The method comprises the following steps: The method comprises the following steps of: acquiring T1 weighted MRI and rs-fMRI of a plurality of normal people and patients with brain cognitive impairment; carrying out pretreatment; carrying out double regression analysis by taking the preprocessed rs-fMRI as a regression dependent variable and the brain function network as a regression independent variable to obtain an individual level brain function network; constructing a deep variationalautoencoder (VAE) model, taking the obtained individual level brain function network diagram as the input and output of the VAE, and taking the encoder part as a feature extraction module for obtaining the implicit code of the individual function network; constructing a multi-layer sensor network to classify the codes obtained by the VAE in the step 4; and deducing samples in the test set by using the trained classifiers for different brain function networks, and fusing deduction results of the classifiers to obtain a final classification result.acquiring T1 weighted magnetic resonance imagesT1 Weighted MRI and resting state functional magnetic resonance images rs-of a plurality of normal persons and patients with brain cognitive impairment; fMRI; carrying out pretreatment; pretreated rs- Performing double regression analysis by taking fMRI as a regression dependent variable and taking the brain function network as a regression independent variable to obtain an individual level brainfunction network; constructing a depth variation auto-encoder (VAE) model, taking the obtained individual level brain function network diagram as input and output of the VAE, and taking the encoder part as a feature extraction module for obtaining hidden codes of the individual function network; constructing a multi-layer perceptron network to classify the codes obtained by the VAE in the step 4;inferring samples in the test set by utilizing a plurality of trained classifiers for different brain function networks, and fusing inference results of the plurality of classifiers to obtain a finalclassification result; according According to the invention, the classification accuracy is improved.

The present invention relates to the image processing technology, and concretely provides a resting state function magnetic resonance image data classification method based on a high-order super network. The problem is solved that the traditional magnetic resonance image data classification method is low in classification accuracy. The resting state function magnetic resonance image data classification method based on the high-order super network comprises the following steps: the step S1: performing preprocessing of the resting state function magnetic resonance image; the step S2: performing time window segment of the average time sequence of each brain region; the step S3: calculating the Pearson's correlation coefficients between each two average time sequences of each brain region; the step S4: extracting the values of corresponding elements in the Pearson's correlation matrix; the step S5: employing a sparse linear regression model to construct a high-order super network; the step S6: calculating the local attributes of the high-order super network; the step S7: selecting the classification features and constructing a classifier; and the step S8: performing quantification of the importance degree and the redundancy degree of the selected features. The resting state function magnetic resonance image data classification method based on the high-order super network is suitable for the classification of the magnetic resonance image data.

Methods for quickly estimating apparent diffusion coefficient probability density functions (ADC PDFs) for each image voxel are provided using a “diffusion reproducibility evaluation and measurement” (DREAM) magnetic resonance sequence. Non-diffusion-weighted (reference) images collected simultaneously have blood oxygenation level dependent (BOLD) sensitivity that can be used for resting-state fMRI data to measure functional connectivity, an unbiased parameter reflecting neurological integrity. ADC coefficient of variation (ADC CV) measurements can be used to isolate and label regions of non-enhancing tumor and predict future enhancement independent of FLAIR, T2, or average ADC maps. Functional diffusion mapping (fDMs) using voxel-wise changes in ADC PDFs can be used to spatially visualize and statistically quantify response to treatment. Additionally, the temporal (time-resolved) diffusivity information can be used for real-time MR thermometry, which is useful for cancer treatment monitoring, and for microperfusion quantification, and tumor/tissue characterization.

The invention provides a multi-modal brain image depression identification method and system based on graph node embedding. Deep learning is applied to the depression recognition of a multi-modal brain image, and a bridge is built between a multi-modal brain network and a convolutional neural network (CNN) through graph node embedding, so that the CNN can be used for the depression recognition ofthe multi-modal brain image, and therefore, depression recognition accuracy is improved. The method comprises the following steps of: 1) acquiring the resting state fMRI and DTI image data of a depressive patient and a normal control group; 2) preprocessing the acquired fMRI and DTI image data; 3) respectively constructing a brain function network and a brain structure network according to the preprocessed fMRI and DTI image data to obtain a brain network adjacency matrix, and 4) adopting graph node embedding to express the adjacency matrix as an image, inputting the image into a convolutionalneural network for classifying the image, and establishing a classification model for identifying depressive patients and normal subjects.

The invention discloses a resting-state functional magnetic resonance data processing method based on primary epilepsy, which belongs to the technical field of data processing for disease detection, and comprises the following steps: S1. data collection; S2. data preprocessing; S3 .Data analysis: including ReHo data analysis, ALFF data analysis and FC data analysis, all ReHo, ALFF, FC maps are analyzed separately, and then aggregated to determine the changed brain regions. The data processing method of the present invention is extremely valuable for the exploration of multifocal epileptic lesions, avoids the difference in brain activity caused by complex task stimuli, and basically controls the experimental conditions of different subjects at the same level, and is easy to cooperate and more It is easy to draw reliable conclusions. It understands the pathogenesis of certain clinical symptoms of epileptic patients from the perspective of brain function, can help locate epileptic foci and surrounding functional areas, and guide epilepsy surgery and the resection range of epileptic foci.

The invention discloses an obese patient functional image analysis method based on mutual sample entropy. The method is characterized by comprising the following steps of collecting resting-state functional magnetic resonance data of a brain; selecting interest areas; extracting a time sequence of voxels in each interest area, up-sampling the time sequence of each voxel, and calculating the mutual sample entropy value of any two interest areas; comparing the mutual sample entropy value of any two interested area of the obese patient with the mutual sample entropy value of corresponding two interested areas of a normal testee, and determining the reason for obesity or worsening of the obesity. The method has the beneficial effects that starting with the physiological line of the obese patient through the resting-state imaging way, the changing of physiological activity of the brain of the patient can be accurately reflected; the EEG signal brain network establishing method is applied to fMRI, and the problem that the fMRI time resolution is not high can be overcome by utilizing the up-sampling method; the mutual sample entropy overcomes the matching problem of the mutual approximate entropy.

An artificial immune method for constructing a brain effect connection network from fMRI data. On the basis of a biological immune system, an artificial immune system combined with the fMRI data is disclosed and can be used for construction of the brain effect connection network. The artificial immune method particularly comprises the following steps of: carrying out experimental design, i.e. performing functional magnetic resonance scanning by using a resting-state experiment; carrying out fMRI data acquisition, i.e. under the condition of reducing a head movement and other errors as further as possible, carrying out scanning to obtain fMRI image data; carrying out pre-processing, i.e. performing pre-processing on the data by using a statistical method, and removing errors and noise which are caused by partial outside factors; selecting a region in which the user is interested, and selecting a brain region related to the study; constructing the effect connection network by a method of optimizing Bayesian network structure learning by using the artificial immune system, and searching the effect connection network matched with an fMRI data set by means of the network structure learning; and carrying out analysis, i.e. analyzing the constructed network and mining biological characteristics exposed by a network structure.

The invention relates to an image processing technique and specifically relates to a classifying method for functional magnetic resonance image data based on multi-scale brain network characteristics. The invention solves the problem of low classifying accuracy of the traditional magnetic resonance image data classifying method. The classifying method for functional magnetic resonance image data based on multi-scale brain network characteristics comprises the following steps: S1) pre-processing the resting state functional magnetic resonance image; S2) adopting a dynamic random seed method for performing region segmentation on the image and extracting mean time sequences for the cut brain areas; S3) calculating a relevance degree of each two mean time sequences of the brain areas; S4) performing binarization processing on the incidence matrix; S5) calculating a local property of the resting state functional brain network and an AUC value thereof in a specific threshold space; S6) constructing a classifier; S7) quantizing the importance and redundancy of the selected characteristics in the classifier. The classifying method provided by the invention is fit for the classification of the magnetic resonance image data.

The invention relates to a human brain function network classification method based on a convolutional neural network, and is used for solving the problems that an existing method ignores the modularization characteristics of a brain network and classification accuracy is low. The human brain function network classification method specifically comprises the following steps of: obtaining resting state fMRI data, carrying out preprocessing, utilizing a preprocessed fMRI time sequence signal to calculate the function connection intensity of each brain interval, and constructing a real human brain function network dataset; independently dividing the real dataset and a simulated dataset into a training set, a verification set and a test set; constructing the convolutional neural network CNN-MF based on scale modularization characteristics for classifying the human brain function network; and carrying out model training: carrying out classification by a model which finishes the training so as to realizing discovery and diagnosis aid of brain diseases. The method disclosed by the invention can effectively utilize modularization structure information in human brain function network data so as to more accurately carry out a brain disease diagnosis.

The invention provides a portable electroencephalogram depression detection system in combination with a demographic attention mechanism. On one hand, the accuracy of electroencephalogram signal sequence learning and modeling is improved by using a convolutional neural network, and on the other hand, demographic information of individuals is introduced in combination with the attention mechanism,and more effective depressive disorder detection is realized. The system comprises an electroencephalogram data acquisition module, a data preprocessing module and a depressive disorder detection module, wherein the electroencephalogram data acquisition module is used for acquiring resting state electroencephalogram original data of a subject; the data preprocessing module is used for carrying outdata preprocessing on the collected original data; and the depressive disorder detection module is used for completing depressive disorder detection based on the electroencephalogram data after datapreprocessing, constructing and training a model by adopting an artificial neural network to classify electroencephalogram signals, and fusing the demographic information into a modeling process of the electroencephalogram signals by jointly using convolution operation and the attention mechanism.

The invention discloses an individualized target positioning method based on weight function connection. The method comprises the following steps: S1, performing structural magnetic resonance and resting-state functional magnetic resonance scanning on a patient without magnetic resonance contraindication to obtain structural magnetic resonance and resting-state functional magnetic resonance data;S2, carrying out scale evaluation of mild cognitive impairment on the patient, and determining scores of different dimensions of a scale; S3, preprocessing the resting-state functional magnetic resonance data; S4, determining a deep brain region and a surface contact brain region which need to be intervened; and S5, in combination with the score of the cognitive assessment scale, calculating the maximum weight function connection of the functional connection between the deep brain region related to cognition and the brain region of interest, and determining the point of the brain region of interest corresponding to the maximum weight function connection as an individualized stimulation target of transcranial magnetic stimulation. The method solves the problem that repeated transcranial magnetic stimulation can only be focused on the surface of the brain and cannot stimulate deep brain tissue.

The invention relates to a depression and schizophrenia recognition method. The depression and schizophrenia recognition method comprises the following steps of S1, obtaining the electroencephalogramdata of a patient during opening eyes and closing eyes under resting state; S2, pretreating the electroencephalogram data in the step S1; S3, converting the data obtained during pretreatment in the step S2 into power spectral density PSD; S4, extracting various characteristics of PSD in the step S3, and constructing a characteristic vector; S5, with the characteristic vectors in the step S4 as input of a classifier, classifying the data; S6, performing model assessment on the classification results in the step S5, so as to obtain a classification model being good in generalization ability; andS7, inputting a test set into the classifier model in the step S6, performing classification, and realizing depression and schizophrenia recognition. According to the depression and schizophrenia classification method, a support vector machine classification method is used for treating and classifying the electroencephalogram of the patient, and the method has good divisibility for data, and is higher in accuracy than other classification methods.

The invention discloses an autism electroencephalogram signal classification device based on a resting-state brain network, belongs to the technical field of biomedical information, and particularly relates to a pattern classification method in the field of brain-computer interfaces. According to the device, autism is recognized and diagnosed from the three aspects of information entropy, power spectrum and brain function network, the most common electroencephalogram signal analysis method at the current research stage is included, an SPN filter is innovatively used for diagnosing the autism patient, and the accuracy rate is close to 100%. Particularly, by researching resting-state brain networks of autism children and normal children, the network topology difference of the autism childrenand the normal children is explored, and comparison with other various methods is performed, so that the pathology of autism is explored to a certain extent, and a reliable basis is provided for clinical diagnosis of autism.