160 results about "Biological neural network" patented technology

The invention relates to the field of geological disaster prediction, in particular to a landslide dynamic evolution prediction method and system based on a biological neural network. The method comprises the following steps: acquiring comprehensive monitoring data of a landslide area; establishing a biological neural network architecture for simulating and learning a landslide dynamic evolution rule; obtaining a dynamic prediction result of the chaotic edge state of the landslide system; generating a landslide spatio-temporal evolution path in combination with a topological manifold generation algorithm; and through the landslide spatio-temporal evolution path, realizing visual prediction of landslide dynamic evolution. According to the method, the biological neural network architecture and the critical state neurodynamics model are established, so that deep fusion and high-precision dynamic prediction of multi-source data in the landslide evolution process are realized; a dynamic prediction result is converted into a four-dimensional space-time voxel density field by utilizing a quantum dot neural fluorescence labeling system and a topological manifold generation algorithm, and a high-resolution landslide space-time evolution path is generated, so that multi-dimensional visual expression of a landslide evolution process is realized.

The invention relates to the technical field of medical instruments, and discloses an implantable neural electrode for long-term electrophysiological signal monitoring and a preparation method and application thereof. The neural electrode comprises an upper hydrogel insulation packaging layer, a lower hydrogel insulation packaging layer, an electrode site, a connecting line and a bonding pad, wherein the electrode site, the connecting line and the bonding pad are arranged between the upper hydrogel insulation packaging layer and the lower hydrogel insulation packaging layer, and the electrode site and the bonding pad are connected through the connecting line. The electrode sites, the connecting lines and the bonding pads are made of conductive hydrogel; the upper hydrogel insulation packaging layer and the lower hydrogel insulation packaging layer are made of insulation polymers and hydrogel. The insulating polymer is a plasma surface-modified insulating polymer. The preparation method of the neural electrode is simple and efficient, biotoxic materials are not introduced in the preparation process, good biocompatibility is achieved, continuous monitoring in a living body can be achieved for one year, important application value is achieved for studying signal response and neural circuits of all brain regions, and a beneficial tool is provided for long-term in-vivo monitoring.

The invention discloses a small molecule drug dynamic analysis system based on a spiking neural network, and relates to the field of drug analysis, and the system comprises a management module which is used for controlling the editing and sending of control instructions of all function modules and units; the data collection module is used for collecting known drug compound structure data and corresponding in-vivo performance data; the model construction module is used for constructing a pulse graph neural network model architecture, and taking the drug compound structure data collected by the data collection module and the corresponding in-vivo performance data as training samples for training; the pulse calculation of the biological neural network can better simulate the characteristics of a biological system and the dynamic process of the biological system, so that the prediction precision of complex drug molecule behaviors is improved, training is performed on input data distribution through dynamic adjustment, the pulse graph neural network has higher generalization ability, error sources are captured, and the prediction precision is improved. Model structures and parameters are continuously adjusted from model identification output, and dynamic optimization is achieved.

The invention discloses a time interference electrical stimulation system for suppressing, controlling and adjusting, which comprises an image acquisition module for acquiring functional magnetic resonance images to acquire functional connection data; the stimulation module is used for respectively generating time interference electrical stimulation waveforms in the plurality of stimulation areas so as to stimulate the corresponding stimulation areas and adjusting stimulation parameters; the evaluation module is electrically connected with the image acquisition module and the stimulation module, acquires cognitive emotion evaluation data, and obtains an emotion and cognitive neural circuit regulation result by combining the emotion evaluation data and the function connection data as an inhibition control capability evaluation index; and the control module is electrically connected with the stimulation module and the evaluation module, and controls the stimulation module to adjust the stimulation parameters based on the emotion and cognitive neural circuit adjustment result. According to the system, effective time interference electrical stimulation is carried out on a plurality of inhibition control intervention brain regions, stimulation parameters are improved by implementing stimulation effect evaluation, and a favorable adjustment means is provided for inhibition control of multi-region and complex network modulation.

The invention discloses a closed-loop training device and a closed-loop training method for an in-vitro neural chip to control obstacle avoidance of a trolley, and belongs to the field of neurobiology and in-vitro brain-computer interfaces, and the closed-loop training device shapes a biological neural network structure and regulates and controls a discharge mode in a closed-loop training mode, so that the device makes a correct decision in different obstacle situations to control obstacle avoidance of the trolley. According to the device, by means of sensory electrical stimulation coding, the neuron culture can'sense 'the obstacle condition faced by the trolley in the environment, and then motion control over the external trolley is achieved by collecting and decoding the discharge signals of the neuron. According to the method, training electrical stimulation is adopted to provide feedback for a neural network, a system applies reward or punishment electrical stimulation to the neural network according to whether a trolley successfully avoids an obstacle or not, a synaptic structure and a discharge mode of the neural network are adaptively shaped and optimized, and the safety of the trolley is improved. And a network structure capable of autonomously controlling the trolley to avoid obstacles in different obstacle environments is gradually formed.

The invention relates to the field of artificial intelligence, in particular to an autism spectrum disorder knowledge graph construction method and system. The method comprises the following steps: acquiring multi-source heterogeneous autism spectrum disorder information; constructing an autism spectrum disorder original corpus; performing information extraction by using a hybrid convolutional neural network-long and short-term memory network-hidden Markov model; designing a credibility dynamic evolution mechanism; constructing and optimizing knowledge topology self-organization; and generating an autism spectrum disorder knowledge graph. According to the method, information is efficiently and accurately extracted by constructing the hybrid model, so that the analysis efficiency of the unstructured medical text is remarkably improved; by designing a credibility dynamic evolution mechanism, dynamic iterative optimization of the triad credibility is realized; by introducing an adversarial verification mechanism, the polarity of the triple is efficiently and accurately identified, and the manual intervention requirement is greatly reduced; by constructing a multi-granularity interpretation engine, cross-level correlation analysis of gene pathway-neural circuit-behavior phenotype is realized, and the analysis accuracy is improved.

The invention relates to the technical field of nerve regulation and control, and provides a closed-loop ultrasonic brain-computer interface regulation and control method for improving post-stroke dysfunction, which comprises the following steps: S1, constructing integrated closed-loop ultrasonic brain-computer interface equipment which comprises a behavior control unit, a signal generator unit, a power amplifier unit and an electrophysiological signal monitoring unit; s2, constructing a mouse model, and embedding an electrode provided for the electrophysiological signal monitoring unit for monitoring; s3, dividing the mouse model into a random stimulation group and a brain-computer interface treatment group, and simultaneously performing closed-loop ultrasonic stimulation for preset time; s4, after closed-loop ultrasonic stimulation for a preset time, performing various behavioral test evaluations including representation of a sensory movement function and representation of an emotional emotion function; and S5, carrying out neural circuit remodeling evaluation, and observing the projection change of the high-order thalamus PO nuclei region of the mouse to the primary cortex. In the aspects of targeting, closed-loop control and integrated design, the method is expected to become a novel and effective post-stroke dysfunction treatment method.

The invention discloses a local path planning method and system based on a spatial memory neural mechanism, and aims to improve obstacle avoidance and shortcut discovery capabilities of a robot in a complex environment. The algorithm is inspired by a hippocampus navigation neural circuit, and efficient memory search and path planning are realized by simulating functions of grid cells, position cells and boundary cells. Firstly, a robot explores a feasible path and records memory nodes including grid cell characterization, reward values and cognitive values; secondly, constructing a hierarchical memory module, realizing fine-to-sparse information abstraction from the back side to the ventral side, and realizing information inheritance and memory search through connection at fixed intervals; and finally, performing path planning by utilizing direction decoding and boundary cell neural representation, and dynamically selecting an optimal sub-target point. The algorithm does not need a large amount of training, can be quickly deployed in various robots, and realizes intelligent autonomous navigation.

The present invention discloses a memristor pulse neural network system and method, which belongs to the field of artificial intelligence technology. The memristor pulse neural network system provided by the present invention uses a bionic design concept to accurately simulate the synaptic structure of a biological neural network by connecting pairs of memristors of the same level in series; based on the positive and negative Hebbian plasticity rules, an adaptive learning mechanism is implemented to strengthen the computing function of the memristor's storage and computing performance; a cascaded scalable control architecture combining master control and slave control is adopted to give the system a high degree of scalability. According to the needs of specific tasks, users can flexibly adjust the scale of the neural network to achieve unlimited cascade expansion, thereby easily coping with various complex scenarios. In addition, the built-in ADC acquisition module provides a real-time feedback mechanism, which can improve the accuracy of system monitoring, speed up the response speed, and provide valuable data support for system optimization and adjustment.

A tumor classification method, an apparatus, a terminal device and a storage medium. The method comprises: performing feature processing on specific tumor gene data to be classified, so as to obtain a target feature vector; and, by means of a trained biological neural network model, identifying the target feature vector to obtain a classification result of said specific tumor gene data. As the biological neural network model is constructed by introducing biological knowledge, classification results of specific tumor gene data can be predicted on the basis of specific cancer tumor gene attributes combined with hierarchical information in biology, so that the degree of credibility of the classification results is higher, thus improving the accuracy of tumor classification.

A Biological Neural Network (BNN) core unit comprising a neural cell culture, an input stimulation unit, an output readout unit may be controlled through its various life cycles to provide data processing functionality. An automation system comprising an environmental and chemical controller unit adapted to operate with the BNN stimulation and readout data interfaces facilitates the monitoring and adaptation of the BNN core unit parameters. Pre-processing and post-processing of the BNN interface signals may further facilitate the training and reinforcement learning by the BNN. Multiple BNN core units may also be assembled together as a stack. The proposed system provides a BNN Operating System as a core component for a wetware server to receive, process and transmit data for different client applications without exposing the BNN core unit components to the client user while requiring significantly less energy than conventional silicon-based hardware and software information processing for high-level cognitive computing tasks.

The invention relates to the technical field of biological neural networks, in particular to an ammonia escape concentration real-time prediction and control system based on data analysis, which comprises a shared long short-term memory (LSTM) neural network model. The neural network architecture receives as its input a multi-source timing enhancement feature generated via a particular calculation. The neural network not only learns time sequence dependence to predict future physical quantities, but also quantifies the uncertainty of self prediction through an integrated Dropout mechanism. An output layer of the neural network is designed to be of a double-branch structure, and a predicted value and another key system operation state evaluation value are output respectively. The system uses output information of the neural network model to derive technical parameters. Besides, a neural network performance monitoring and self-adaptive calibration unit is integrated, performance degradation of the neural network model can be identified, re-training and re-deployment of the neural network model can be triggered, and the self-adaptive learning ability of the neural network is shown.

A Biological Neural Network (BNN) core unit comprising a neural cell culture, an input stimulation unit, an output readout unit may be controlled through its various life cycles to provide data processing functionality. An automation system comprising an environmental and chemical controller unit adapted to operate with the BNN stimulation and readout data interfaces facilitates the monitoring and adaptation of the BNN core unit parameters. Pre-processing and post-processing of the BNN interface signals may further facilitate the training and reinforcement learning by the BNN. Multiple BNN core units may also be assembled together as a stack. The proposed system provides a BNN Operating System as a core component for a wetware server to receive, process and transmit data for different client applications without exposing the BNN core unit components to the client user while requiring significantly less energy than conventional silicon-based hardware and software information processing for high-level cognitive computing tasks.

The invention relates to a device and a method for distinguishing adaptive behaviors of a mouse musical scale based on discrete frequency acoustic stimulation. The device and the method are composed of an acoustic stimulation generation module, a behavior response acquisition module and a self-adaptive decision control module, an acoustic stimulation sequence of discrete frequency scales is acquired, and a multi-modal parameter coordinated regulation and control system is constructed in combination with an Arduino development environment; on the basis of an embedded signal control circuit, an autonomously designed closed-loop feedback control unit and a multi-channel acoustic signal generation device are integrated, and a musical scale-behavior associated quantitative model based on operational conditioned reflex is established through a two-stage experimental process of'progressive adaptive training-stepped threshold verification '. The method has the function of synchronously detecting auditory cortex frequency coding precision and prefrontal cortex decision ring synaptic plasticity, and a novel experimental tool is provided for quantitative evaluation of neurodegenerative disease central auditory dysfunction and research on a decision-related neural circuit plasticity mechanism.

The invention relates to a closed-loop transcranial acoustic-magnetic stimulation neural circuit regulation and control system and method, and the system comprises an individualized simulation navigation module which is used for carrying out the simulation navigation of ultrasonic stimulation and magnetic stimulation on a brain region of a regulation and control object, and obtaining a stimulation site; the transcranial ultrasonic stimulation module is used for outputting ultrasonic stimulation by adopting the initialized stimulation parameters according to the stimulation sites and carrying out transcranial ultrasonic stimulation on the regulation and control object; the transcranial magnetic stimulation module is used for outputting magnetic stimulation by adopting the initialized stimulation parameters according to the stimulation sites and carrying out transcranial magnetic stimulation on the regulation and control object; and the electroencephalogram collecting, processing and feedback module is used for synchronously collecting and analyzing electroencephalogram signals in the acoustic and magnetic combined stimulation process and optimizing stimulation parameters. According to the method, high spatial resolution of transcranial ultrasonic stimulation, flexibility of transcranial magnetic stimulation and real-time processing of electroencephalogram data are combined, and accurate individualized neural circuit regulation and control are achieved.

The invention discloses a humanoid robot control method based on a neural circuit, and belongs to the technical field of robot control, and the method comprises the steps: S1, constructing a three-layer pulse neural control architecture comprising spinal cord reflex, brainstem balance and cortex decision, and carrying out the weight adaptive distribution and cooperative operation between layers through the dynamic adjustment of neurotransmitter concentration, organic cooperation of prediction and reflection is realized through a three-stage control system; the cerebellar prediction network executes attitude adjustment in advance through a delay compensation algorithm, and compared with pure feedback, the control energy consumption is reduced; the parallel reflex pathway of the spinal cord layer establishes a nerve bypass mechanism, and the emergency response is realized by bypassing high-level processing in an emergency; the multi-layer safety monitoring improves the triggering accuracy of a protection mechanism through error confidence evaluation and contact stability analysis; a weight transfer mechanism of pulse frequency modulation avoids sudden movement change during control mode switching; by means of the design, the stability retention rate of the robot walking on the inclined plane is remarkably improved compared with a traditional method.

Techniques for using a biological and artificial neural network (BANN) system to perform a task. The BANN system comprises a multi-electrode array (MEA); a biological neural network (BNN) comprising neurons arranged on the MEA, a trained statistical model trained using inputs generated using responses of the BNN to training data inputs; and at least one processor. The method comprises using the BANN system to receive an input signal; encode the input signal to generate a stimulation pattern; stimulate the BNN by using the MEA to generate electrical signals in accordance with the stimulation pattern; measure, using the MEA, a response of the BNN responsive to the stimulating; generate, based on the measured response, an input for the ANN; process the input with the trained statistical model to obtain corresponding output; and use the output from the trained statistical model in furtherance of performing the task.

Method for designing a mechatronic component, characterized by the following features: The component is repeatedly simulated by software (10) with certain design parameters (11) that take on variable values, A fixation point (20) of the designer (30) is captured on a screen, while the software (10) displays the component to the designer (30) on the screen. A neural circuit with LTC neurons is trained to infer the resulting fixation point (20) of the designer (30) from the current values ​​of the design parameters (11), and The values ​​of the design parameters (11) are modified by the neural circuit and fed back to the software (10).

The invention relates to the technical field of industrial container detection, and discloses an assembled enamel tank body state detection system, which comprises a three-dimensional sensor group provided with a three-dimensional heterogeneous sensing network, the three-dimensional sensor group is arranged on the surface of an enamel tank body through pasting to form a biological neural network sensing system, and an energy supply unit comprises thermoelectric power generation and wireless charging embedded magnetic attraction power supply. According to inner wall corrosion diagnosis, codes at the inner wall corrosion position are decoded through a code coating protection film, a decoding result is positioned, full-time and full-domain monitoring of the state of the enamel tank is achieved, and the intelligent operation and maintenance level of industrial equipment is remarkably improved.

The present invention provides a neuron logic gate metal oxide semiconductor νLGMOS circuit, which can simulate the "integration and firing" behavior of neurons in a biological neural network system, and can be fabricated together with multiple digital arithmetic circuits by using industrial CMOS logic process technology. The processing system of the present invention includes an analog νLGMOS circuit, a converter circuit, and a digital processing circuit, which can optimize the power and cost of various applications and can be fabricated into an integrated circuit chip by using the same CMOS logic process technology. At the same time, the above-mentioned analog νLGMOS circuit is inspired by the biological neural network system and can be simulated, designed, and fabricated into an integrated circuit chip for applications in the biomedical field.

The invention relates to a touch fluid neuron specific tracing imaging agent, a preparation method and application, and belongs to the technical field of nuclear medicine imaging, the tracing imaging agent is composed of a nerve tracer agent horseradish peroxidase (HRP), zirconium-89 isotope chelated by deferoxamine and a pharmaceutically acceptable carrier, the effective tracking window of the tracer agent is prolonged to 120 hours, and the effective tracking window of the tracer agent is prolonged to 120 hours. The multi-stage synaptic transfer period of the fluid touching neuron can be completely covered; the spatial resolution is improved by 5-8 times compared with that of traditional FDG-PET, and a fluid touching neuron cluster can be accurately presented; the in-vitro serum stability is excellent, the radiochemical purity is still larger than or equal to 80.5% within 120 hours, PET / MRI bimodal imaging can be adapted, drug administration is conducted through lateral ventricle minimally invasive injection, the probe is suitable for various animal models such as mice, rats and non-human primates, in-vivo dynamic tracing of sympathetic fluid neuron trans-synapses can be achieved on the premise that no tissue damage exists, and the probe has a good application prospect. And a key technical tool is provided for neural circuit analysis, neuroscience basic research and preclinical research of neurodegenerative diseases.

The invention discloses an in-vitro biological neural network accurate regulation and control method and system based on a microelectrode array, and the method comprises the steps: S1, culturing and planting neuronal cells on a microelectrode array chip, and forming an in-vitro neural network with spontaneous activity; s2, preprocessing the in-vitro neural network to enable the in-vitro neural network to enter a stable function state capable of being effectively activated; s3, recording an electrophysiological signal, sampling the structure and function connection of the neural network, and dynamically selecting an optimal stimulation site; s4, applying electrical stimulation to the selected stimulation sites according to a preset regulation and control target, collecting neural network response signals in real time, and constructing a closed-loop regulation and control loop; and S5, performing quantitative analysis on the neural network based on the multi-dimensional evaluation index to form a comprehensive evaluation result, and feeding back the comprehensive evaluation result to the step S4 to optimize a subsequent stimulation strategy. According to the method, organic unification of sensing, intelligent decision making, closed-loop intervention and multi-dimensional quantitative evaluation of a neural network structure and a function state can be realized.

The invention discloses a spacecraft attitude stability control method based on a liquid neural network. The method comprises the following steps: S1, establishing a spacecraft attitude dynamic model under multi-source complex disturbance; s2, designing a conventional spacecraft attitude stability controller, and constructing a data set in combination with the attitude dynamics model; s3, training and obtaining a liquid neural network interference observer by adopting a liquid time constant network and a neural circuit strategy architecture; s4, designing a spacecraft attitude stability controller based on a liquid neural network disturbance observer, and proving the stability of a closed-loop system by using a Lyapunov function; s5, verifying the estimation effect of the liquid neural network interference observer and the stability effect of the spacecraft attitude control system through simulation; results show that the method has the advantages that interference of the spacecraft attitude can be effectively estimated, stable control over the spacecraft attitude can be achieved, high real-time performance, low power consumption and high robustness are achieved, and the spacecraft attitude stable control efficiency can be improved.

The purpose of the present invention is to provide a novel method for targeting a neuron population that is a neural circuit element which has more limited functional localization. The purpose of the present invention is also to provide a method for treating a disease or the like by targeting a specific neuron population involved in a disease or the like by said novel method and controlling the neural activity thereof.　The present invention is a vector set used for targeting a specific neuron population in a neural circuit, said vector set being characterized by comprising first to third vectors, wherein: the first vector is an antegrade trans-synaptic virus vector that contains a base sequence which codes for a split-Cre recombinase consisting of a C-terminal-side sequence of a Cre recombinase; the second vector is a retrograde trans-synaptic virus vector that contains a base sequence which codes for a split-Cre recombinase consisting of an N-terminal-side sequence of a Cre recombinase; the third vector is a non-trans-synaptic virus vector that contains an inverted sequence of a gene sequence which codes for a desired protein and that has a loxP sequence and a loxP mutation sequence which face opposite to each other and which are at both ends of the inverted sequence; and a functional Cre recombinase is formed in a cell that has been infected with the first vector and the second vector.

The invention relates to the field of neural network bionics, in particular to a bionic neural network construction method and system, and the method comprises the steps: obtaining a fluctuation threshold value of each neuron cell, and expanding an input dimension for each neuron cell according to the fluctuation threshold value; establishing a corresponding forward network for each neuron cell, and performing network delay dimension expansion on each neuron cell according to the time delay value of each directed edge in the process of establishing each forward network; performing sample recombination on the input sample according to a dimension expansion result; and the weight of the fluctuation threshold input dimension is fixed to train the bionic neural network to be trained. Compared with the prior art, the method has the advantages that the input dimension is expanded for each neuron cell through the fluctuation threshold, and the network delay dimension expansion is carried out on each neuron cell containing the fluctuation threshold; and the trained bionic neural network can better utilize the fluctuation threshold and the network after delay dimension expansion of each neuron cell network to realize quantification and simulation of the biological neural network.

The invention relates to a method for recording and identifying social behaviors of drosophila melanogaster, and belongs to the technical field of neuroscience. The method comprises the steps of building a shooting platform, training an adaptive model through an existing machine learning model, recognizing postures of fruit flies in a video through the adaptive model, and putting a machine learning coordinate file into a patent program to obtain social behavior statistics. Through collaborative design of an adaptive machine learning model and a structured data processing program, full-automatic identification and index quantification of the social behaviors of the drosophila melanogaster are realized, the technical blank that the social behaviors cannot be identified by a drosophila melanogaster mating identification method is filled up, key data support is provided for research of a neural circuit mechanism, gene-behavior association and the like, and the social behaviors of the drosophila melanogaster mating identification method are optimized. Compared with the prior art, the method is obviously improved.

A method and system for differentiating acute and chronic pain based on the functional state of specific neural circuits is proposed. This method utilizes in vivo two-photon / endoscopic calcium imaging and ex vivo brain slice patch-clamp techniques to longitudinally monitor the activity dynamics of neurons in the DRN and ZI brain regions during the acute and chronic phases of a neuropathic pain model. By extracting and analyzing calcium and electrophysiological signal parameter sets, distinguishing criteria are established: if the activity of 5-HTergic neurons in the DRN is specifically enhanced in the acute phase, the indication is acute pain; if the activity of GABAergic neurons in the ZI is specifically enhanced in the chronic phase, the indication is chronic pain. This invention achieves an objective and precise differentiation of the transition from acute to chronic pain at the level of specific neural circuits, providing a new strategy and tool for pain mechanism research and clinical diagnosis and treatment.

This invention provides a method and device for refractive development assessment based on visual development prediction neural circuits, relating to the field of myopia management technology. The method includes: acquiring color retinal optical imaging signals and refractive ground state quantification values ​​of the subject; performing retinal stabilization preprocessing; extracting multi-level morphological features through a reentrant layered retinal choroid encoder; performing spatial feature aggregation gating on the extracted feature maps to obtain feature vectors; concatenating and integrating the feature vectors with the refractive ground state quantification values ​​to obtain fused feature vectors; constructing a fused feature vector sequence and inputting it into a refractive development dynamics internal model inferrer to generate a future multi-time-point refractive state prediction sequence; converting it into an interpretable refractive state numerical sequence; and providing risk warnings based on preset risk thresholds. This invention explicitly embeds physiological mechanisms prior into the model structure and parameter adaptive strategies, thereby maintaining the stability and interpretability of predictions even in a single medical visit scenario.

The present invention discloses an ultra-low power consumption neural circuit, belonging to the field of neural circuit technology. It includes: a photoelectric receiver, an amplifier, and a comparator of a mirror current source; the photoelectric receiver, the amplifier, and the comparator of the mirror current source are cascaded, and the amplifier and the comparator of the mirror current source have a common part; the amplifier amplifies the input signal of the photoelectric receiver, the photoelectric receiver is used to generate a neural signal, and the amplifier is used to amplify the input neural signal. The circuit of the present invention combines an amplifier composed of two inverters with a comparator composed of five transistors, which effectively reduces the number of branches in the circuit while maintaining the original function, thereby reducing the power consumption of the neural module.

The invention provides an auxiliary analysis method, device and equipment for a neural circuit injury position and a medium, and relates to the field of medical image processing, and the method comprises the steps: obtaining whole brain sample data, including fMRI data and DTI data of a healthy sample and a tested sample; selecting a seed point based on the fMRI data, establishing whole brain function connection based on the seed point, and extracting a time-varying connection coefficient sequence between the seed point and the region of interest; based on the DTI data, determining a connection fiber between the seed point and the region of interest through probabilistic fiber tracking, obtaining a target liposome representing the connection fiber and a surrounding region thereof, and obtaining a white matter bold signal sequence of the target liposome; and performing difference analysis on the connection coefficient sequence and the white matter bold signal sequence of the healthy sample and the tested sample to obtain the abnormal voxel of the tested sample, and determining the neural circuit injury position according to the coordinate of the abnormal voxel, thereby providing important information for the research of neural activity.