Transcranial direct current stimulation intervention system

Abstract

The invention relates to the technical field of transcranial direct current stimulation, and particularly discloses a transcranial direct current stimulation intervention system which analyzes a brain structure network and a brain function network marked as effective before and after treatment by using a data processing and image analysis algorithm based on a machine learning and artificial intelligence technology. According to the method, brain structure-function network features before treatment are captured, and brain network state features of a new individual are reflected by performing feature extraction and splicing on a brain structure network and a brain function network of the new individual when brain positive reaction possibility identification and potential effectiveness detection are actually performed on the new individual. And taking the brain network state features of the new individual as an index to query feature information with similar response from a brain network state database with effective intervention, so as to identify the possibility of positive reaction of the brain of the new individual, thereby predicting whether the classical tDcs intervention normal form is effective to a patient object to be evaluated or not.

Description

Technical Field

[0001] The present application relates to the technical field of transcranial direct current stimulation, and more specifically, to a transcranial direct current stimulation intervention system. Background Art

[0002] Transcranial Direct Current Stimulation (tDCS) is a non-invasive brain stimulation technique that modulates the activity level of neurons in the cerebral cortex by applying a weak direct current through electrodes on the scalp. tDCS has been used for a variety of clinical and research purposes, including improving cognitive function, treating depression, and relieving chronic pain. Although tDCS has potential, its effects vary from person to person and may not be significantly effective for some individuals.

[0003] Chinese patent CN111870808B provides a transcranial direct current stimulation intervention method, which pre-evaluates the treatment response rate of tDCS for an individual by combining non-invasive brain network detection (magnetic resonance scanning and brain network analysis), so as to achieve the purpose of applying tDCS to intervene in emotional problems in a targeted manner. In simple terms, it is to exclude individuals who may not respond to tDCS intervention in a targeted manner, greatly avoiding the waste of medical resources.

[0004] However, in the above-mentioned intervention method of transcranial direct current stimulation, it specifically analyzes multi-person big data of brain images based on imaging data of brain structure network and brain functional network, so as to capture the distribution morphological characteristics of brain networks that have a positive response to the classic tDCS intervention method, and use them to identify the possibility of positive brain response of new individuals. This method captures the distribution morphological characteristics of brain networks that have a positive response and compares them with the distribution morphological characteristics of brain networks of new individuals to identify whether there is a possibility of positive response, but because the brain structure and functional connection patterns of each individual are different, the features captured based on group data may not fully reflect the uniqueness of the individual. In other words, this method of comparing the distribution morphological characteristics of brain networks of groups may make it difficult to fully consider the subtle differences between individuals, thereby affecting the reliability of the assessment of the possibility of response of new individuals.

[0005] Therefore, an optimized transcranial direct current stimulation intervention system is desired. Summary of the invention

[0006] The present application provides a transcranial direct current stimulation intervention system, which can match similar brain network states of personalized patients from a database of brain network states known to be effective for tDCS interventions through a more intelligent query response method, which helps to assist in predicting the potential effectiveness of tDCS interventions for different individual patients, especially for those individuals who may not respond to standard treatment options.

[0007] According to one aspect of the present application, a transcranial direct current stimulation intervention system is provided, comprising: The effective intervention brain network data acquisition module is used to obtain the data sets marked as effective {brain structure network before treatment, brain function network before treatment, brain structure network after treatment, brain function network after treatment}; An effective intervention brain network analysis module is used to extract features and aggregate analysis on each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as effective to obtain a set of pre-treatment brain structure-function network features; A patient subject brain network data acquisition module, used to obtain a brain structure network and a brain function network of the patient subject to be evaluated; A patient subject brain network analysis module, used for extracting and concatenating features of the brain structure network and brain function network of the patient subject to be evaluated to obtain brain structure-function network coding features of the patient subject to be evaluated; A patient subject brain state query response module is used to perform fast query dynamic response processing on the set of the brain structure-function network coding features of the patient subject to be evaluated and the brain structure-function network features before treatment to obtain the brain state query response coding features of the patient subject to be evaluated; The intervention effectiveness detection module is used to determine whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated based on the brain state query response coding characteristics of the patient subject to be evaluated.

[0008] The present application provides a transcranial direct current stimulation intervention system, which uses a data processing and image analysis algorithm based on machine learning and artificial intelligence technology to analyze the brain structure network and brain function network marked as effective before and after treatment, so as to capture the brain structure-function network characteristics before treatment, which is used to reflect the anatomical connectivity and functional interaction between different brain regions where these tDCS interventions are effective, thereby forming an effective brain network state database. When the possibility of positive brain response is actually identified and the potential effectiveness is detected for a new individual, the brain network state characteristics of the new individual are reflected by extracting and splicing the brain structure network and brain function network of the new individual, and the brain network state characteristics of the new individual are used as an index to query the response-similar feature information from the effective brain network state database to identify the possibility of positive brain response of the new individual, thereby predicting whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated. In this way, similar brain network states of personalized patients can be matched from the known effective tDCS intervention brain network state database through a more intelligent query response method, which helps to assist in predicting the potential effectiveness of tDCS interventions for different individual patients, especially for those who may not respond to standard treatment options. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0010] Figure 1 This is a schematic block diagram of a transcranial direct current stimulation intervention system according to an embodiment of the present application.

[0011] Figure 2 This is a data flow diagram of the transcranial direct current stimulation intervention system according to an embodiment of the present application.

[0012] Figure 3 This is a schematic block diagram of an effective brain network analysis module in the transcranial direct current stimulation intervention system according to an embodiment of the present application.

[0013] Figure 4 This is a schematic block diagram of a patient brain state query response module in a transcranial direct current stimulation intervention system according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0015] In response to the above technical problems, in the technical solution of the present application, a transcranial direct current stimulation intervention system is proposed, which can evaluate the potential effectiveness of the classic tDCS intervention paradigm for different patients and the possibility of positive brain response by using machine learning and artificial intelligence technology. Specifically, the transcranial direct current stimulation intervention system uses data processing and image analysis algorithms based on machine learning and artificial intelligence technology to analyze the brain structure network and brain function network marked as effective before and after treatment, so as to capture the brain structure-functional network characteristics before treatment, which are used to reflect the anatomical connectivity and functional interaction between different brain regions where these tDCS interventions are effective, thereby forming an effective brain network state database. When the possibility of positive brain response and potential effectiveness detection are actually performed on new individuals, the brain network state characteristics of the new individual are reflected by extracting and splicing the brain structure network and brain function network of the new individual, and the brain network state characteristics of the new individual are used as indexes to query the response-similar feature information from the effective brain network state database to identify the possibility of positive brain response of the new individual, thereby predicting whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated. In this way, similar brain network states of personalized patients can be matched from a database of brain network states known to be effective for tDCS interventions through a more intelligent query response method, which can help predict the potential effectiveness of tDCS interventions for different individual patients, especially for those who may not respond to standard treatment options.

[0016] like Figure 1 and Figure 2As shown, the transcranial direct current stimulation intervention system includes: an effective intervention brain network data acquisition module 10, used to obtain a data set of {brain structure network before treatment, brain function network before treatment, brain structure network after treatment, brain function network after treatment} marked as effective; an effective intervention brain network analysis module 20, used to extract features and perform aggregation analysis on each {brain structure network before treatment, brain function network before treatment, brain structure network after treatment, brain function network after treatment} in the data set marked as effective to obtain a set of brain structure-function network features before treatment; a patient subject brain network data acquisition module 30, Used to obtain the brain structure network and brain function network of the patient to be evaluated; the patient brain network analysis module 40 is used to extract and splice the brain structure network and brain function network of the patient to be evaluated to obtain the brain structure-function network coding features of the patient to be evaluated; the patient brain state query response module 50 is used to quickly query the set of the brain structure-function network coding features of the patient to be evaluated and the brain structure-function network features before treatment to obtain the brain state query response coding features of the patient to be evaluated; the intervention effectiveness detection module 60 is used to determine whether the classic tDcs intervention paradigm is effective for the patient to be evaluated based on the brain state query response coding features of the patient to be evaluated.

[0017] Exemplarily, in the effective intervention brain network data acquisition module 10, a data set marked as effective {brain structure network before treatment, brain function network before treatment, brain structure network after treatment, brain function network after treatment} is obtained. It should be understood that the efficacy of transcranial direct current stimulation (tDCS) varies from individual to individual. Not all people who receive tDCS intervention will show significant therapeutic effects. By collecting and analyzing the brain structure network before treatment and the brain function network before treatment, relevant feature information about the brain network for which tDCS intervention is effective can be extracted, thereby helping to predict the potential efficacy of new patients. Acquiring these data can help construct an "intervention-effective brain network state data set". This data set contains brain structure and functional network features of individuals who are known to respond well to tDCS. When evaluating new patients, it is possible to predict whether they will respond positively to tDCS by comparing their brain network state with the features in the database. Pre-screening out individuals who may not respond to tDCS can avoid unnecessary treatment attempts, save medical resources and personal time, and improve the success rate of clinical treatment. For patients who may not benefit from standard tDCS protocols, other treatments that may be better suited for them can be explored, allowing for more personalized medical care.

[0018] Exemplarily, in the effective intervention brain network analysis module 20, feature extraction and aggregation analysis are performed on each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as effective {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} to obtain a set of pre-treatment brain structure-function network features. It should be understood that by performing feature extraction and aggregation analysis on the pre-treatment brain structure and function network data set marked as effective, the anatomical connectivity and functional interaction features between those brain regions that have a positive response to tDCS intervention can be captured to form a set containing effective pre-treatment brain structure-function network features, which helps to identify whether new individuals have similar positive response potential before receiving tDCS treatment. .

[0019] In one embodiment, Figure 3 As shown, the effective intervention brain network analysis module 20 includes: a pre-treatment brain feature extraction unit 21, which is used to extract features from each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as effective {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} to obtain a set of pre-treatment brain structure network feature vectors and a set of pre-treatment brain function network feature vectors; a pre-treatment brain feature splicing unit 22, which is used to perform feature splicing on each group of corresponding pre-treatment brain structure network feature vectors and pre-treatment brain function network feature vectors in the set of pre-treatment brain structure network feature vectors and the set of pre-treatment brain function network feature vectors to obtain a set of pre-treatment brain structure-function network feature vectors as the set of pre-treatment brain structure-function network features.

[0020] Exemplarily, in the pre-treatment brain feature extraction unit 21, feature extraction is performed on each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as valid {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} to obtain a set of pre-treatment brain structure network feature vectors and a set of pre-treatment brain function network feature vectors. It should be understood that since each data in the data set marked as valid {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} is image data, the brain structure network (such as a neural fiber network) and the brain function network (such as the functional connection between different brain regions) are from different individuals who have received tDCS intervention and are effective. In order to extract relevant feature information about the brain network in which tDCS intervention is effective from each image data in these data sets, so as to provide a basis for the subsequent evaluation of the effectiveness of tDCS intervention for new individual patients, in the technical solution of the present application, each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as effective {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} is feature extracted to obtain a set of pre-treatment brain structure network feature vectors and a set of pre-treatment brain function network feature vectors. In particular, in a specific example of the present application, feature extraction of different brain network image data can be performed separately by a brain network feature extractor based on a hollow convolutional neural network model, so as to capture the pre-treatment brain structure network features and brain function network features marked as effective, so as to reflect the brain network information of different individuals in which tDCS intervention is effective, such as the degree of connection and activity of specific brain regions. These features help to subsequently build a database for personalized query matching and response of brain network states. For example, higher connection strength between certain regions, or certain regions showing higher activation levels, may be signs of an effective tDCS response.

[0021] In one embodiment, the pre-treatment brain feature extraction unit is used to extract each pre-treatment brain structural network and each pre-treatment brain functional network from the data set, and extract the features of each pre-treatment brain structural network and each pre-treatment brain functional network through a brain network feature extractor based on a dilated convolutional neural network model to obtain a set of pre-treatment brain structural network feature vectors and a set of pre-treatment brain functional network feature vectors. It should be understood that the pre-treatment brain structural network and the pre-treatment brain functional network include high-quality diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) data. These imaging data reflect the strength of nerve fiber connections between different brain regions (structural network) and the change of blood oxygen level dependent signals over time (functional network). The pre-treatment brain structural network and pre-treatment brain functional network of each individual are respectively input into the brain network feature extractor based on the dilated convolutional neural network model. The model uses dilated convolution technology to expand the receptive field without reducing the resolution, so that spatial information in a larger range can be captured. After being processed by the brain network feature extractor based on the dilated convolutional neural network model, the output is a set of brain structural network feature vectors and brain functional network feature vectors for each individual before treatment. These feature vectors not only retain the key information in the original images, but also abstract and condense them through deep learning algorithms, making subsequent aggregate analysis and comparison more efficient and accurate.

[0022] Exemplarily, in the pre-treatment brain feature splicing unit 22, each group of corresponding pre-treatment brain structure network feature vectors and pre-treatment brain function network feature vectors in the set of pre-treatment brain structure network feature vectors and the set of pre-treatment brain function network feature vectors are feature spliced ​​to obtain a set of pre-treatment brain structure-function network feature vectors as the set of pre-treatment brain structure-function network features. It should be understood that the brain structure network (such as white matter fiber connection) and the functional network (such as the temporal correlation of neuronal activity) provide different but complementary information about the brain. The structural network reflects the physical connection mode of the brain, while the functional network shows the dynamic interaction of these connections in actual operation. In order to effectively integrate these two types of network features to obtain a more comprehensive description and representation of the brain state, in the technical solution of the present application, each group of corresponding pre-treatment brain structure network feature vectors and pre-treatment brain function network feature vectors in the set of pre-treatment brain structure network feature vectors and the set of pre-treatment brain function network feature vectors are feature spliced ​​to obtain a set of pre-treatment brain structure-function network feature vectors. Through the feature fusion method of feature splicing, the structural network characteristics and functional network characteristics of the brain before treatment can be combined to capture more complex neural activity patterns, avoiding the problem that considering the structural or functional characteristics of the brain network alone is not enough to fully characterize the individual's response to tDCS intervention, and providing a basis for the subsequent prediction of the effectiveness of tDCS intervention for individual patients.

[0023] In one embodiment, each feature vector in the set of the pre-treatment brain structure network feature vectors and the set of the pre-treatment brain function network feature vectors is first aligned according to the same individual and brain region. Next, for each set of corresponding pre-treatment brain structure network feature vectors and pre-treatment brain function network feature vectors, they are combined into a new, richer pre-treatment brain structure-function network feature vector through a splicing operation. This new pre-treatment brain structure-function network feature vector contains both structural information (such as the strength of nerve fiber connections) and functional information (such as synchronous activity between different brain regions), so that it can more comprehensively describe the state of the brain.

[0024] Exemplarily, in the patient subject brain network data acquisition module 30, the brain structure network and brain function network of the patient subject to be evaluated are obtained. It should be understood that by obtaining the brain structure network and brain function network of the patient to be evaluated, the current state of the brain of the patient subject to be evaluated can be understood in detail. This information provides important basic data for subsequent analysis, enabling doctors to formulate more accurate treatment plans based on individual differences. At the same time, the brain structure network and brain function network of the patient subject to be evaluated can be obtained by comparing with known effective tDCS intervention cases. In this application, a machine learning algorithm is used to match the similarities between the brain network characteristics of new patients and those individuals who respond well to tDCS. If a high degree of similarity is found between the two, it indicates that the new patient may also benefit from the same tDCS intervention; conversely, if the match is low, other treatment methods may need to be considered or tDCS parameters may need to be adjusted to suit individual needs.

[0025] Exemplarily, in the patient object brain network analysis module 40, the brain structure network and brain function network of the patient object to be evaluated are feature extracted and feature spliced ​​to obtain the brain structure-function network coding features of the patient object to be evaluated. It should be understood that when the patient object to be evaluated is evaluated and predicted by the classic tDcs intervention paradigm, the brain structure network and brain function network of the patient object to be evaluated are obtained, and the brain structure network and brain function network of the patient object to be evaluated are feature extracted and feature spliced ​​to obtain the brain structure-function network coding vector of the patient object to be evaluated. It should be understood that since each person's brain structure and functional characteristics are unique, it is difficult to fully characterize the uniqueness of an individual by relying solely on a single type of feature (such as only considering the brain structure network or only considering the brain function network). By analyzing the structure and function network simultaneously, the subtle differences between individuals can be captured more comprehensively, thereby achieving a more personalized tDcs intervention effectiveness prediction. In addition, the same processing method as the above-mentioned data set marked as valid is adopted here, which can provide a more accurate data basis in the subsequent brain state query matching process, facilitating the query response of thickness and the prediction and evaluation of tDcs intervention effectiveness.

[0026] In one embodiment, the patient subject's brain network analysis module is used to: use the brain network feature extractor based on the void convolutional neural network model to respectively extract features of the brain structure network and the brain function network of the patient subject to be evaluated, and then perform feature splicing on the obtained brain structure network feature vector of the patient subject to be evaluated and the brain function network feature vector of the patient subject to be evaluated to obtain a brain structure-function network coding vector of the patient subject to be evaluated as a brain structure-function network coding feature of the patient subject to be evaluated.

[0027] Exemplarily, in the patient subject brain state query response module 50, the set of the brain structure-function network coding features of the patient subject to be evaluated and the set of the brain structure-function network features before treatment are subjected to fast query dynamic response processing to obtain the brain state query response coding features of the patient subject to be evaluated. It should be understood that since the set of the brain structure-function network coding vector of the patient subject to be evaluated and the set of the brain structure-function network feature vector before treatment respectively contain the brain structure-function network state features of the patient subject to be evaluated and the brain network state database marked as effective for tDcs intervention, in predicting and evaluating whether the classic tDcs intervention of the patient subject to be evaluated is effective, it is necessary to query and match the brain network state features of this patient subject with each known effective brain network state in the database to match the similar brain network state of the personalized patient subject for subsequent classification and effectiveness evaluation tasks of the classic tDcs intervention paradigm. However, since the traditional feature matching query method is based on the overall database information for screening, the query response speed and accuracy are low. Therefore, in the technical solution of the present application, the set of the brain structure-function network coding features of the patient to be evaluated and the brain structure-function network features before treatment are further subjected to rapid query dynamic response processing to obtain the brain state query response coding features of the patient to be evaluated. In particular, the process of rapid query dynamic response processing realizes the process from coarse positioning to fine-grained matching and then to deep semantic coding by integrating technologies such as cross entropy calculation, local context perception, linear transformation, and heterogeneous converter structure, aiming to optimize the speed and accuracy of the query response of the brain state of the patient to be evaluated.

[0028] In one embodiment, Figure 4 As shown, the patient object brain state query response module 50 includes: a patient object fast query positioning factor calculation unit 51, which is used to calculate the patient object fast query positioning factor of the brain structure-function network coding vector of the patient object to be evaluated relative to each pre-treatment brain structure-function network feature vector in the set of the pre-treatment brain structure-function network feature vector to obtain a set of patient object fast query positioning factors; a brain state fine-grained matching search window determination unit 52, which is used to determine the patient object brain state fine-grained matching search window based on the set of the patient object fast query positioning factors; a fine-grained query response encoding unit 53, which is used to perform fine-grained query response encoding on the brain structure-function network coding vector of the patient object to be evaluated and the set of the pre-treatment brain structure-function network feature vector based on the patient object brain state fine-grained matching search window to obtain the brain state query response encoding feature of the patient object to be evaluated.

[0029] In one embodiment, the patient object fast query positioning factor calculation unit 51 is used to: calculate the cross entropy of the brain structure-function network encoding vector of the patient object to be evaluated relative to each pre-treatment brain structure-function network feature vector in the set of pre-treatment brain structure-function network feature vectors to obtain the set of patient object fast query positioning factors. Specifically, the process can be expressed as follows:

[0030]

[0031]

[0032] in, is the set of feature vectors of the brain structure-function network before treatment, are the first, second, and third brain structure-function network feature vectors before treatment. and The brain structure-function network feature vector before treatment, is the characteristic value of each position in the brain structure-function network coding vector of the patient to be evaluated, is the number of eigenvalues, It is the The eigenvalues ​​of each position in the eigenvector of the brain structure-function network before treatment, is the brain structure-function network encoding vector of the patient to be evaluated and the The cross entropy between the feature vectors of the brain structure-function network before treatment, i.e., the patient object fast query positioning factor, is a collection of quick query positioning factors for patient objects, The first, second, and and Quickly query positioning factors for each patient object.

[0033] It should be understood that by calculating the cross entropy, the semantic difference between the brain structure-function network encoding vector of the patient to be evaluated and each known effective case in the database can be quantified. Cross entropy, as an indicator to measure the difference between two probability distributions, is used here to evaluate the similarity of two brain network states. A smaller cross entropy value means that the two are closer, and vice versa, it means that the difference is larger. Therefore, this method can quickly narrow the search range and filter out the most likely related features, greatly improving the speed and efficiency of the query. For each patient to be evaluated, finding the most similar effective treatment case to his brain network state is the key to achieving personalized medicine. Cross entropy calculation helps the system identify those historical cases that are most similar to the new patient in structure and function, thereby providing a scientific basis for formulating personalized tDCS intervention plans. In addition, cross entropy as a metric can exclude those obviously mismatched features at an early stage and reduce the risk of misjudgment. This means that only when the patient to be evaluated shows a high degree of consistency with the historical case in multiple dimensions will it be considered as a potential effective treatment target. This not only improves the reliability of the prediction results, but also saves time and resources for subsequent detailed analysis. Based on the set of fast query positioning factors calculated by cross entropy, the system can dynamically adjust its query strategy to prioritize those features with higher similarity for further analysis.

[0034] In one embodiment, the brain state fine-grained matching search window determination unit 52 is used to: use the pre-treatment brain structure-function network feature vector corresponding to the minimum value in the set of the patient object's rapid query positioning factors as the positioning matching pre-treatment brain structure-function network feature vector; based on the positioning matching pre-treatment brain structure-function network feature vector, determine the patient object's brain state fine-grained matching search window, wherein the vector of the center position of the patient object's brain state fine-grained matching search window is the positioning matching pre-treatment brain structure-function network feature vector, and each pre-treatment brain structure-function network feature vector in the patient object's brain state fine-grained matching search window is defined as a fine-grained query pre-treatment brain structure-function network feature vector to obtain a set of fine-grained query pre-treatment brain structure-function network feature vectors. Specifically, the process can be expressed by a formula:

[0035]

[0036]

[0037] in, To return the minimum value , To locate and match the brain structure-functional network feature vector before treatment, It means calculating the square of the vector-norm. It means to calculate the logarithmic function value with base 2. Indicates rounding down. is the matching search window factor, ( ) is the starting point of the fine-grained matching search window for the patient’s brain state, ( ) is the end point of the fine-grained matching search window for the patient's brain state, To query the collection of brain structure-function network feature vectors before treatment in a fine-grained manner, The fine-grained query before treatment brain structure-function network feature vectors in the set of fine-grained query before treatment brain structure-function network feature vectors are respectively queried.

[0038] It should be understood that by selecting the pre-treatment brain structure-function network feature vector corresponding to the minimum value in the fast query positioning factor set as the positioning matching feature vector, the system can accurately locate the historical case that is most similar to the patient to be evaluated. This precise match is the basis for subsequent more detailed analysis, ensuring that the results of the preliminary screening have a high degree of credibility. This step effectively narrows the search range, reduces unnecessary consumption of computing resources, and improves the speed and accuracy of matching. Next, the system constructs a fine-grained matching search window based on the pre-treatment brain structure-function network feature vector of the positioning matching. This window is centered on the positioning matching feature vector, and a set of fine-grained query pre-treatment brain structure-function network feature vectors are defined around it. These feature vectors represent a series of historical cases that are closest to the patient to be evaluated, forming a localized and refined feature space. In this way, the system can perform in-depth analysis in a smaller and more relevant data set, thereby capturing more subtle brain network differences and providing more accurate matching results. The introduction of the fine-grained matching search window enables the system to more carefully compare the similarities and differences between the patient to be evaluated and the historical cases. By further analyzing this local feature space, the system can identify key factors that may affect the effectiveness of tDCS intervention, such as changes in the connection strength or activity patterns of specific brain regions. This refined analysis helps improve the ability of personalized predictions, allowing doctors to develop more accurate treatment plans based on the unique situation of each patient. In addition, each feature vector in the fine-grained matching search window is defined as a collection of fine-grained query brain structure-function network feature vectors before treatment. This process not only enhances the intelligence of the system, but also supports more efficient decision-making assistance.

[0039] In one embodiment, the fine-grained query response encoding unit 53 is used to: perform a linear transformation on the brain structure-function network encoding vector of the patient to be evaluated to obtain the brain structure-function network query vector of the patient to be evaluated and the brain structure-function network value vector of the patient to be evaluated; use each fine-grained query pre-treatment brain structure-function network feature vector in the set of the fine-grained query pre-treatment brain structure-function network feature vector as a key vector, input the brain structure-function network query vector of the patient to be evaluated, the brain structure-function network value vector of the patient to be evaluated and the key vector into a fine-grained query encoding module based on a heterogeneous converter structure to obtain the brain state query response encoding vector of the patient to be evaluated as the brain state query response encoding feature of the patient to be evaluated. Specifically, the process can be expressed as follows:

[0040]

[0041]

[0042] in, is the brain structure-function network coding vector of the patient to be evaluated, and are the query weight matrix and query bias vector respectively, and are the value weight matrix and value bias vector respectively, and are respectively the brain structure-function network query vector of the patient to be evaluated and the brain structure-function network value vector of the patient to be evaluated, The first one in the set of the fine-grained query pre-treatment brain structure-function network feature vectors A fine-grained query of the brain structure-functional network feature vector before treatment, For the said A fine-grained query of the length of the brain structure-function network feature vector before treatment, is matrix multiplication, is the normalized exponential function, A brain state query response encoding vector is provided for the patient subject to be evaluated.

[0043] It should be understood that, first, by linearly transforming the brain structure-function network encoding vector of the patient to be evaluated, the system generates a query vector and a value vector. This transformation not only retains the spatial distribution characteristics of the original features, but also enhances the ability of feature representation through mapping operations. The query vector is used to find the most similar historical cases in the database, while the value vector carries more detailed information, which is helpful for subsequent in-depth analysis. Next, the system inputs these query vectors, value vectors, and each fine-grained query pre-treatment brain structure-function network feature vector extracted from the fine-grained matching search window (as a key vector) into the fine-grained query encoding module based on the heterogeneous transformer structure. The heterogeneous transformer structure allows the system to process different types of data within the same framework, thereby more effectively capturing complex relationships. In this way, the system can efficiently complete the matching task of large-scale data in a short time while ensuring the accuracy of the results. The fine-grained query encoding module uses the attention mechanism to focus on the most relevant historical cases, thereby achieving a deep semantic understanding of the patient's brain state. In this process, the interaction between the query vector and the key vector helps the system identify which historical cases are most similar to the current patient; while the value vector provides additional information, allowing the system to further refine the matching results. This method not only improves the quality of matching, but also reveals the underlying patterns hidden behind the data, providing a more solid foundation for personalized predictions. The brain state query response encoding vector of the patient to be evaluated is finally output as the brain state query response encoding feature, representing a comprehensive description after multiple layers of analysis. This feature integrates information from different dimensions, including structural connectivity and functional interactions, and can more comprehensively reflect the patient's unique brain state. Doctors can evaluate the possibility of tDCS intervention and its expected effects based on these detailed features, so as to develop more precise treatment plans. For example, if the brain state of a new patient is highly consistent with a known effective case, it indicates that the patient may also benefit from similar tDCS interventions; conversely, if significant differences are found, it may be necessary to adjust the intervention strategy or explore other treatment methods.

[0044] In summary, the process of fast query dynamic response processing introduces a fast positioning mechanism, that is, using cross entropy calculation to evaluate the semantic distance between the brain structure-function network encoding vector of the patient to be evaluated and each of the pre-treatment brain structure-function network feature vectors, so that the semantic difference between the two can be quantified, thereby quickly narrowing the search scope to the most likely relevant area and improving the query matching response speed. Then, preliminary screening finds the best matching pre-treatment brain structure-function network feature vector. This method reduces unnecessary computing resource consumption and speeds up the preliminary screening process. Secondly, in order to further improve the matching query accuracy, the system defines a fine-grained matching search window around the best matching pre-treatment brain structure-function network feature vector. This window focuses on the local area around the known intervention-effective pre-treatment brain structure-function network features of the best matching point determined by the preliminary screening. This strategy enhances the model's ability to understand the deep semantic associations between the brain structure-functional network states of the patient to be evaluated and the known effective brain structure-functional network states before treatment, enabling the system to capture more subtle differences in pre-treatment brain states, thereby providing more accurate patient brain state query response results and feature representations, providing a basis for subsequent classification judgments on whether the classic tDcs intervention paradigm is effective for the patient to be evaluated.

[0045] Exemplarily, in the intervention effectiveness detection module 60, based on the brain state query response coding features of the patient subject to be evaluated, it is determined whether the classic tDCS intervention paradigm is effective for the patient subject to be evaluated. In one embodiment, the intervention effectiveness detection module is used to: input the brain state query response coding vector of the patient subject to be evaluated into the classifier-based evaluation module to obtain an evaluation result, and the evaluation result is used to indicate whether the classic tDCS intervention paradigm is effective for the patient subject to be evaluated. That is to say, the brain state query response characterization information of the patient subject to be evaluated is used for classification processing to identify the possibility of positive brain response of the new individual, so as to predict whether the classic tDCS intervention paradigm is effective for the patient subject to be evaluated. In this way, similar brain network states of personalized patients can be matched from the known effective tDCS intervention brain network state database through a more intelligent query response method, which helps to use similar pre-treatment brain network states to screen out brain network features after treatment, thereby assisting in predicting the potential effectiveness of tDCS intervention for different individual patients, especially for those who may not respond to standard treatment plans.

[0046] In a specific embodiment of the present application, the brain state query response encoding vector of the patient to be evaluated is input into the evaluation module based on the classifier to obtain the evaluation result, including: using the fully connected layer of the evaluation module based on the classifier to fully connect the brain state query response encoding vector of the patient to be evaluated to obtain the fully connected encoded brain state query response encoding vector of the patient to be evaluated. The function of the fully connected layer is to convert the input high-dimensional feature vector into a low-dimensional representation while capturing the complex relationship between the features. This process enhances the learning ability of the model so that it can better understand the intrinsic pattern of the input data. Then, the fully connected encoded brain state query response encoding vector of the patient to be evaluated is passed through the Softmax classification function of the evaluation module based on the classifier to obtain the probability distribution belonging to different categories. The Softmax function can convert the original scores on multiple categories into probability values, the sum of which is equal to 1, which is convenient for explaining the relative possibility of each category. In the present application, two probability values ​​are obtained: a first probability belonging to the presence of a positive treatment response and a second probability belonging to the absence of a positive treatment response. These two probability values ​​reflect the probabilities of different outcomes that may occur in the patient to be evaluated after receiving tDCS intervention. Finally, based on the comparison between the first probability and the second probability, the final classification result is determined. For example, if the first probability is higher than the second probability, it is determined that the classic tDcs intervention paradigm is effective for the patient subject to be evaluated; conversely, if the second probability is higher, it means that the classic tDcs intervention paradigm is ineffective for the patient subject to be evaluated, and other treatment options may need to be considered.

[0047] Preferably, passing the brain state query response encoding vector of the patient subject to be evaluated through a classifier-based evaluation module to obtain an evaluation result comprises: Calculate the distance between each pair of eigenvalues ​​of the brain state query response encoding vector of the patient subject to be evaluated, such as the L2 distance, and take the square root of the distance to obtain the brain state query response encoding distance representation matrix of the patient subject to be evaluated, that is,

[0048] in, represents the brain state query response encoding vector of the patient subject to be evaluated, and The first encoding vector representing the brain state query response encoding vector of the patient to be evaluated and Eigenvalues, The first encoding vector representing the brain state query response encoding vector of the patient to be evaluated and The distance between eigenvalues, Represents the brain state query response encoding distance representation matrix of the patient to be evaluated The eigenvalue of the position; The patient subject brain state query response encoding autocorrelation matrix of the patient subject brain state query response encoding vector to be evaluated is obtained, that is, ,in, represents the brain state query response encoding vector of the patient subject to be evaluated, represents the transpose of a vector, represents matrix multiplication, An autocorrelation matrix representing the encoding of the query response of the patient subject's brain state to be evaluated; The brain state query response encoding vector of the patient to be evaluated is matrix-multiplied with the brain state query response encoding distance representation matrix of the patient to be evaluated to obtain the first-level mapping vector of the brain state query response encoding of the patient to be evaluated, that is, ,in, represents the brain state query response encoding vector of the patient subject to be evaluated, represents the query response encoding distance representation matrix of the patient's brain state to be evaluated, represents matrix multiplication, A first-level mapping vector representing a query response encoding a brain state of a patient subject to be evaluated; The brain state query response encoding primary mapping vector of the patient subject to be evaluated is matrix-multiplied with the matrix product of the brain state query response encoding distance representation matrix of the patient subject to be evaluated and the brain state query response encoding autocorrelation matrix of the patient subject to be evaluated to obtain the brain state query response encoding multi-level mapping vector of the patient subject to be evaluated ,in, represents the first-level mapping vector of the query response encoding of the brain state of the patient to be evaluated, represents the query response encoding distance representation matrix of the patient's brain state to be evaluated, represents matrix multiplication, represents the autocorrelation matrix of the query response encoding of the brain state of the patient to be evaluated, A multi-level mapping vector representing a query response encoding a brain state of a patient subject to be evaluated; The multi-level mapping vector of the brain state query response encoding of the patient to be evaluated is interpolated with the associated eigenvector of the brain state query response encoding of the patient to be evaluated (if the eigenvalue is insufficient, interpolation or zero filling is performed) composed of the eigenvalue of the auto-association matrix of the brain state query response encoding of the patient to be evaluated to obtain an optimized brain state query response encoding vector of the patient to be evaluated; The optimized brain state query response encoding vector of the patient subject to be evaluated is passed through a classifier-based evaluation module to obtain an evaluation result.

[0049] Considering that the brain structure-function network coding vector of the patient to be evaluated represents the semantic coding splicing features of the brain structure network and the brain function network of the patient to be evaluated, and the set of the pre-treatment brain structure-function network feature vectors represents the semantic coding splicing features of the pre-treatment brain structure network and the pre-treatment brain function network of multiple patient objects marked as valid, when performing dynamic semantic query encoding based on a fast positioning mechanism, differences in source data formats and additional interference introduced during the data acquisition process will cause positioning accuracy to be imbalanced, resulting in the lack of decision-making of the query coding feature instance of the brain state query response coding vector of the patient to be evaluated obtained by the dynamic semantic query encoding, thereby affecting the accuracy of the evaluation result obtained by the classifier-based evaluation module.

[0050] Therefore, the self-correlation of the complete similarity instantiation of the brain state query response encoding vector of the patient to be evaluated is instantiated by a linear target mapping representation based on the similarity distance representation matrix of the brain state query response encoding vector of the patient to be evaluated, and the negative impact factor of the association mismatch is compensated by the association fusion kernel bias to improve the degree of instance determination of the eigenvalue of the brain state query response encoding vector of the patient to be evaluated under the similarity restriction, that is, the degree of significance of the eigenvalue as an instance for classification regression determination, and improve the accuracy of the evaluation result obtained by the classifier-based evaluation module of the brain state query response encoding vector of the patient to be evaluated. In this way, similar brain network states of personalized patients can be matched from the known effective tDCS intervention brain network state database through a more intelligent query response method, which helps to assist in predicting the potential effectiveness of tDCS intervention for different individual patients, especially for those who may not respond to standard treatment plans.

[0051] In summary, the transcranial direct current stimulation intervention system according to the embodiment of the present application is explained, which uses data processing and image analysis algorithms based on machine learning and artificial intelligence technology to analyze the brain structure network and brain functional network marked as effective before and after treatment, so as to capture the brain structure-functional network characteristics before treatment, which are used to reflect the anatomical connectivity and functional interaction between different brain regions where these tDCS interventions are effective, thereby forming an effective brain network state database. When actually identifying the possibility of positive brain response and detecting the potential effectiveness of new individuals, the brain network state characteristics of the new individual are reflected by extracting and splicing the brain structure network and brain functional network of the new individual, and the brain network state characteristics of the new individual are used as indexes to query the response-similar feature information from the effective brain network state database to identify the possibility of positive brain response of the new individual, thereby predicting whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated. In this way, similar brain network states of personalized patients can be matched from a database of brain network states known to be effective for tDCS interventions through a more intelligent query response method, which can help predict the potential effectiveness of tDCS interventions for different individual patients, especially for those who may not respond to standard treatment options.

[0052] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the method and device in the embodiment of the present invention is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0053] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0054] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose of the present invention and the scope of protection of the present invention, all of which are within the protection of the present invention.

Claims

1. A transcranial direct current stimulation intervention system, characterized in that: include: The effective intervention brain network data acquisition module is used to obtain the data sets marked as effective {brain structure network before treatment, brain function network before treatment, brain structure network after treatment, brain function network after treatment}; An effective intervention brain network analysis module is used to extract features and aggregate analysis on each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as effective to obtain a set of pre-treatment brain structure-function network features; A patient subject brain network data acquisition module, used to obtain a brain structure network and a brain function network of the patient subject to be evaluated; A patient subject brain network analysis module, used for extracting and concatenating features of the brain structure network and brain function network of the patient subject to be evaluated to obtain brain structure-function network coding features of the patient subject to be evaluated; A patient subject brain state query response module is used to perform fast query dynamic response processing on the set of the brain structure-function network coding features of the patient subject to be evaluated and the brain structure-function network features before treatment to obtain the brain state query response coding features of the patient subject to be evaluated; The intervention effectiveness detection module is used to determine whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated based on the brain state query response coding characteristics of the patient subject to be evaluated.

2. The transcranial direct current stimulation intervention system according to claim 1, characterized in that: The effective intervention brain network analysis module includes: A pre-treatment brain feature extraction unit, used for extracting features from each {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} in the data set marked as valid {pre-treatment brain structure network, pre-treatment brain function network, post-treatment brain structure network, post-treatment brain function network} to obtain a set of pre-treatment brain structure network feature vectors and a set of pre-treatment brain function network feature vectors; The pre-treatment brain feature splicing unit is used to perform feature splicing on each group of corresponding pre-treatment brain structure network feature vectors and pre-treatment brain function network feature vectors in the set of pre-treatment brain structure network feature vectors and the set of pre-treatment brain function network feature vectors to obtain a set of pre-treatment brain structure-function network feature vectors as the set of pre-treatment brain structure-function network features.

3. The transcranial direct current stimulation intervention system according to claim 2, characterized in that: The pre-treatment brain feature extraction unit is used to extract each pre-treatment brain structure network and each pre-treatment brain function network from the data set, and perform feature extraction on each pre-treatment brain structure network and each pre-treatment brain function network respectively through a brain network feature extractor based on a void convolutional neural network model to obtain a set of feature vectors of the pre-treatment brain structure network and a set of feature vectors of the pre-treatment brain function network.

4. The transcranial direct current stimulation intervention system according to claim 3, characterized in that: The patient subject brain network analysis module is used to: use the brain network feature extractor based on the void convolutional neural network model to respectively extract features of the brain structure network and the brain function network of the patient subject to be evaluated, and then feature splice the obtained brain structure network feature vector of the patient subject to be evaluated and the brain function network feature vector of the patient subject to be evaluated to obtain a brain structure-function network coding vector of the patient subject to be evaluated as the brain structure-function network coding feature of the patient subject to be evaluated.

5. The transcranial direct current stimulation intervention system according to claim 4, characterized in that: The patient subject brain state query response module comprises: A patient object rapid query positioning factor calculation unit is used to calculate the patient object rapid query positioning factor of the brain structure-function network encoding vector of the patient object to be evaluated relative to each pre-treatment brain structure-function network feature vector in the set of pre-treatment brain structure-function network feature vectors to obtain a set of patient object rapid query positioning factors; A brain state fine-grained matching search window determining unit, configured to determine a brain state fine-grained matching search window of a patient object based on a set of fast query positioning factors of the patient object; A fine-grained query response encoding unit is used to perform fine-grained query response encoding on the set of the brain structure-function network encoding vector of the patient object to be evaluated and the brain structure-function network feature vector before treatment based on the fine-grained matching search window of the patient object's brain state to obtain the query response encoding feature of the patient object's brain state to be evaluated.

6. The transcranial direct current stimulation intervention system according to claim 5, characterized in that: The patient object fast query positioning factor calculation unit is used to calculate the cross entropy of the brain structure-function network encoding vector of the patient object to be evaluated relative to each pre-treatment brain structure-function network feature vector in the set of pre-treatment brain structure-function network feature vectors to obtain a set of patient object fast query positioning factors.

7. The transcranial direct current stimulation intervention system according to claim 6, characterized in that: The brain state fine-grained matching search window determination unit is used to: The brain structure-function network feature vector before treatment corresponding to the minimum value in the set of positioning factors for rapid query of the patient object is used as the brain structure-function network feature vector before treatment for positioning matching; Based on the positioning matching pre-treatment brain structure-function network feature vector, the fine-grained matching search window of the patient's brain state is determined, wherein the vector of the center position of the fine-grained matching search window of the patient's brain state is the positioning matching pre-treatment brain structure-function network feature vector, and each pre-treatment brain structure-function network feature vector in the patient's brain state fine-grained matching search window is defined as a fine-grained query pre-treatment brain structure-function network feature vector to obtain a set of fine-grained query pre-treatment brain structure-function network feature vectors.

8. The transcranial direct current stimulation intervention system according to claim 7, characterized in that: The fine-grained query response encoding unit is used to: Performing a linear transformation on the brain structure-function network encoding vector of the patient subject to be evaluated to obtain a brain structure-function network query vector of the patient subject to be evaluated and a brain structure-function network value vector of the patient subject to be evaluated; Taking each fine-grained query pre-treatment brain structure-function network feature vector in the set of the fine-grained query pre-treatment brain structure-function network feature vectors as a key vector, the brain structure-function network query vector of the patient object to be evaluated, the brain structure-function network value vector of the patient object to be evaluated and the key vector are input into a fine-grained query encoding module based on a heterogeneous converter structure to obtain a brain state query response encoding vector of the patient object to be evaluated as the brain state query response encoding feature of the patient object to be evaluated.

9. The transcranial direct current stimulation intervention system according to claim 8, characterized in that: The intervention effectiveness detection module is used to: input the brain state query response encoding vector of the patient subject to be evaluated into a classifier-based evaluation module to obtain an evaluation result, and the evaluation result is used to indicate whether the classic tDcs intervention paradigm is effective for the patient subject to be evaluated.

Images