Construction method and device of electroencephalogram and symptom-oriented fusion mental disorder evaluation model

Abstract

The application provides a mood disorder evaluation model construction method and equipment based on electroencephalogram and symptom-oriented fusion. The method comprises the following steps: obtaining multiple mood disorder related symptom characteristics of multiple subjects; collecting electroencephalogram characteristics of multiple brain regions of each subject to construct a first training sample set; based on the first training sample set, a multi-level symptom-oriented feature extraction module is constructed, and multiple levels of electroencephalogram characteristic latent variables oriented by symptoms are obtained; based on the electroencephalogram characteristic latent variables of each level, the corresponding mood disorder diagnosis label is labeled, and then the feature fusion module and the classification module are iteratively trained; based on the multi-level symptom-oriented feature extraction module, the feature fusion module and the classification module, a mood disorder evaluation model is obtained. The application solves the problem that in the prior art, a single feature extraction and learning strategy is adopted, and there is no constraint of symptom information, resulting in limited recognition ability of the model for mood disorders of different functional abnormal types.

Description

Technical Field

[0001] This invention belongs to the field of medical informatics technology, and in particular relates to a method and device for constructing a mood disorder assessment model that integrates electroencephalography (EEG) and symptom-guided approaches. Background Technology

[0002] Current diagnostic methods for mood disorders primarily rely on subjective assessments and scale evaluations by clinicians, lacking objective biological markers. In recent years, electroencephalography (EEG), as a non-invasive, real-time, and high-temporal-resolution neurophysiological technique, has shown great potential in the diagnosis of mood disorders. EEG can directly reflect the electrical activity of brain neurons, providing objective physiological indicators for the diagnosis of mood disorders. Existing technologies offer numerous methods and references for applying EEG in disease diagnosis.

[0003] However, mood disorder classification methods based on EEG signals still face several technical bottlenecks. Traditional classifiers can only identify single abnormal patterns in brain regions, performing poorly when faced with the complex pathophysiological mechanisms of mood disorders. Mood disorders may be caused by different mechanisms, such as prefrontal cortex dysfunction, limbic system overactivation, or abnormal connectivity across multiple brain regions. Existing models lack the ability to comprehensively identify these heterogeneous abnormal patterns, limiting their generalization and practicality. Furthermore, existing models employ single feature extraction and learning strategies, failing to fully capture the unique functional patterns of each brain region, resulting in limited ability to identify mood disorders with different functional abnormalities. Additionally, existing methods lack a direct mapping relationship between EEG features and clinical symptom dimensions. Without the constraint of symptom information, the models cannot internally reflect the direct or indirect relationship between electrophysiological indicators and diagnostic results, limiting their clinical applicability. Some studies use both electrophysiological and clinical features for brain disease differentiation. These methods directly use clinical information as feature input for disease differentiation, which weakens the importance of objective indicators, as clinical symptom features are generally strongly correlated with disease definitions. This can cause the model to focus on learning features with discriminative power, while ignoring disease-related information hidden in electrophysiological indicators, affecting the model's ability to identify different diseases with similar distributions of electrophysiological indicators but different symptoms. Summary of the Invention

[0004] Based on the above analysis, the present invention aims to provide a method and device for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches, in order to solve the problem that existing technologies use a single feature extraction and learning strategy without the constraint of symptom information, resulting in limited ability of the model to identify mood disorders of different functional abnormalities.

[0005] The objective of this invention is mainly achieved through the following technical solutions:

[0006] On the one hand, this invention provides a method for constructing a mood disorder assessment model that integrates electroencephalography (EEG) and symptom-guided approaches, comprising:

[0007] Multiple mood disorder-related symptom characteristics were obtained from multiple subjects; and EEG characteristics of multiple brain regions of each subject were collected to construct the first training sample set.

[0008] Based on the first training sample set, a multi-level symptom-oriented feature extraction module is constructed, and multiple levels of symptom-oriented latent EEG feature variables are obtained.

[0009] Based on the latent variables of EEG features at each level corresponding to multiple subjects, corresponding mood disorder diagnostic labels are labeled to construct a second training sample set; and the feature fusion module and classification module are iteratively trained using the second training sample set.

[0010] Based on the multi-level symptom-oriented feature extraction module, the convergent feature fusion module, and the classification module, a mood disorder assessment model is obtained.

[0011] Furthermore, the multi-level symptom-guided feature extraction module includes:

[0012] The primary symptom-oriented latent variable extraction module, constructed using partial least squares based on symptom features and EEG features corresponding to each brain region in the first training sample set, is used to extract multiple primary EEG feature latent variables most relevant to global symptom features for each brain region and the whole brain. ;

[0013] The secondary symptom-oriented latent variable extraction module is used to decouple the EEG characteristics of each brain region and the whole brain from the corresponding primary EEG latent variables, thereby obtaining the secondary EEG latent variables containing symptom specificity for each brain region and the whole brain. ;

[0014] The third-level subtype fusion latent variable extraction module is used for the extraction of the second-level EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region. Based on each EEG subtype, feature extraction and fusion were performed to obtain a three-level EEG feature latent variable of fused symptom subtype features. .

[0015] Furthermore, the primary symptom-oriented latent variable extraction module is constructed using the following method:

[0016] The mean of each symptom feature corresponding to each subject is used as the global symptom feature;

[0017] Based on the global symptom features and the corresponding EEG features of each brain region and the whole brain, a first-level global regression model corresponding to each brain region and the whole brain is obtained by training using partial least squares.

[0018] A first-level latent variable screening unit is constructed to obtain multiple first-level latent EEG feature variables that are most correlated with the global symptom features for each brain region and the whole brain based on the projected weights W of the first-level global regression model. ;

[0019] Based on the first-level global regression model and the first-level latent variable screening unit, the first-level symptom-oriented latent variable extraction module is obtained.

[0020] Furthermore, the first-level global regression model is trained using the following method:

[0021] S1: Standardize the various EEG characteristics and symptom characteristics, and represent them as follows: , ;initialization and ,make , ;in, The mean of the EEG characteristic X is... y represents the mean of the symptom characteristic. The standard deviation of EEG characteristics, The standard deviation of the symptom characteristics This represents the H-th update of EEG feature X, with latent variable dimension H as the target. Let y represent the symptom feature updated for the h-th time with the latent variable dimension H as the target.

[0022] S2: with As the target variable, based on the formula By maximizing and linear combination and The covariance is calculated to obtain the h-th weight vector. , is represented as:

[0023] ;

[0024] S3: Based on and Obtain latent variables , is represented as: m is the number of samples;

[0025] S4: Based on and latent variables The load of the h-th EEG feature was obtained. , is represented as: ;

[0026] S5: Based on and latent variables To obtain the load of symptom characteristics , is represented as: ;

[0027] S6: Receive the next round of input, where the next round of input is represented as: , Iteratively execute S2-S5 to obtain the first-level global regression model corresponding to each brain region and the whole brain. and latent variable matrix EEG feature load matrix and model weight matrix .

[0028] Furthermore, the secondary symptom-guided specific latent variable extraction module is constructed using the following method:

[0029] Based on the EEG feature load matrix corresponding to each brain region and the whole brain Latent variables of primary EEG characteristics of each brain region and the whole brain Perform an inverse transformation to obtain the reconstructed original features. ;

[0030] Based on the difference between the original EEG features corresponding to each brain region and the whole brain and the reconstructed original features, the first-order residual features corresponding to each brain region and the whole brain are obtained. ;

[0031] For each symptom feature, the first-order residual feature is used separately. A corresponding second-order symptom-specific fitting model was established using partial least squares method. ; Utilizing the aforementioned secondary symptom-specific fitting model Inferring the secondary symptom-specific latent variables corresponding to each symptom characteristic ;

[0032] Construct fusion units to merge secondary symptom-specific latent variables corresponding to each symptom. And simultaneously merge the aforementioned primary EEG latent variables. Secondary EEG latent variables were obtained. ;

[0033] Based on the secondary symptom-specific fitting model The fusion unit is used to obtain the secondary symptom-guided specific latent variable extraction module.

[0034] Furthermore, the third-level subtype fusion latent variable extraction module includes:

[0035] The three-level subtype clustering module is used to perform clustering based on the secondary EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region and the whole brain. and the posterior probability of each subtype. ;

[0036] The latent variable extraction module from the perspective of subsymptom subtypes is used to infer subtype latent variables based on the PLS projection model of each pre-trained EEG subtype. ;

[0037] The subtype fusion module from the perspective of subsymptoms is used to combine the posterior probabilities of each EEG subtype. The values ​​are converted into deterministic weights, and the third-level EEG feature latent variables are obtained based on these deterministic weights. .

[0038] Furthermore, the three-level subtype clustering module is constructed using the following method:

[0039] Based on a preset target cluster size range, secondary EEG latent variables of each brain region and the whole brain are utilized. Train Gaussian mixture models corresponding to different target cluster numbers in sequence;

[0040] The optimal number of target clusters is obtained based on the Bayesian Information Criterion (BIC) value of each Gaussian mixture model.

[0041] The Gaussian mixture model is retrained using the target number of clusters as a parameter. The Gaussian mixture model corresponding to the optimal number of target clusters is used as a third-level subtype clustering module for inferring clustering results.

[0042] Furthermore, the construction of the latent variable extraction module from the perspective of subsymptom subtypes includes:

[0043] A secondary symptom-specific fitting model based on each symptom feature in each brain region and the whole brain. For the corresponding secondary symptom-specific latent variables Perform an inverse transformation to obtain the reconstructed specific features. ;

[0044] Utilizing the first-order residual characteristics Reconstruction-specific features The difference is used as a second-order residual characteristic. Based on the aforementioned electrophysiological subtypes Extract the secondary residual feature samples corresponding to each cluster;

[0045] Based on the secondary residual feature samples, a corresponding subtype PLS projection model is established for each cluster using partial least squares method. Complete the construction of the latent variable extraction module from the perspective of the subsymptom subtype.

[0046] Furthermore, the subtype fusion module from the subsymptom perspective obtains the three-level symptom subtype fusion features through the following method:

[0047] The posterior probabilities of each subtype The entropy transformed into a subtype probability distribution is expressed as:

[0048] ;

[0049] Calculate maximum entropy Where N represents the number of target subtypes;

[0050] use This normalizes the entropy into a deterministic fraction.

[0051] Based on the subtype latent variables The deterministic scores yielded the third-level EEG latent variables. .

[0052] On the other hand, a computer device is also disclosed, including at least one processor and at least one memory communicatively connected to said processor;

[0053] The memory stores instructions that can be executed by the processor to implement the aforementioned method for constructing a mood disorder assessment model that integrates EEG and symptom-oriented approaches.

[0054] The beneficial effects of this technical solution are:

[0055] 1. The method provided by this invention employs a multi-level modeling strategy for EEG and symptom-guided fusion, and a brain region segmentation strategy. An independent EEG feature latent variable extraction model is established for each brain region. Each brain region uses features generated by EEG electrodes corresponding to its spatial location. Different levels of models are used for feature extraction to capture the unique functional patterns of each brain region. Global coordination is achieved through a fusion strategy, realizing EEG feature extraction based on symptom information constraints, and improving the model's ability to assess mood disorders of different functional abnormalities.

[0056] 2. This invention utilizes multi-dimensional clinical symptom indicators and employs a multi-layered modeling strategy. It not only establishes a global symptom fitting projection module but also simultaneously constructs linear fitting models for each of the multi-dimensional clinical features. Using the projection matrix as the feature matrix, it fuses the fitting results from multiple clinical symptom dimensions to assess mood disorders, thus improving the accuracy of the assessment. Furthermore, in practical applications, it can simultaneously output fitting results from multiple clinical symptom dimensions, providing doctors with symptom references to aid in diagnosis.

[0057] 3. This invention establishes a mapping relationship between EEG features and the symptom space through a fitting projection mechanism. Clinical indicators are only used to train the projection matrix, realizing the conversion from EEG to the symptom semantic space. This ensures that the final classification is mainly based on objective electrophysiological indicators, maintaining the objectivity of the diagnosis. Attached Figure Description

[0058] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0059] Figure 1 This is a flowchart of the method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to an embodiment of the present invention. Detailed Implementation

[0060] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0061] One embodiment of the present invention provides a method for constructing a mood disorder assessment model that integrates electroencephalography (EEG) and symptom-guided approaches, such as... Figure 1 As shown, it includes:

[0062] Step S1: Obtain multiple mood disorder-related symptom characteristics from multiple subjects; and collect EEG characteristics from multiple brain regions and the whole brain of each subject to construct the first training sample set;

[0063] Specifically, for symptom characteristics, multiple mood disorder-related assessment scales were obtained from the participants. These scales include, but are not limited to, HAMD17 and HCL33. The symptom assessment results for each symptom item were obtained based on the assessment scales, and these results were used as the symptom characteristics for each symptom item. The symptom items used in this embodiment are shown in Table 1.

[0064] Table 1: Examples of Symptom Items

[0065]

[0066] For acquiring EEG features, a clinically standard EEG cap was first used to collect EEG data from multiple brain regions of the subject, followed by preprocessing. Preprocessing included: resampling, rereference, notch filtering to remove power frequency, removal of base drift, removal of invalid high-frequency components, ICA to remove electrooculography (EOG), electromyography (EMG), and electrocardiogram (ECG) components, removal of abnormal amplitude segments, and segmentation of the processed segments into fixed-length sample segments. The removal of abnormal amplitude segments included: time-slicing the bandpass-filtered EEG data, calculating the amplitude of the slice samples, and deleting time slices with amplitudes outside a preset range. In this embodiment, 120 seconds was used as the target slice length.

[0067] Multi-angle EEG features were further extracted from the preprocessed EEG signals, including frequency band energy features, connectivity features, amplitude-phase coupling features, time-spectrum features, microstate features, and pre-trained model encoding features. Based on the brain region extents of the frontal, parietal, occipital, left temporal, right temporal, and central regions, the multi-angle EEG features corresponding to different brain regions were grouped to obtain EEG feature sets for each brain region. Furthermore, a global EEG feature set based on spatial location was obtained across the entire brain. Finally, the EEG feature sets of each brain region were obtained. and global EEG feature set The overall original feature set G is represented as: { }, used for subsequent model building and training, where, For frontal region EEG feature set, For the central region EEG feature set, For the parietal region EEG feature set, For the occipital region EEG feature set, The left temporal EEG feature set, The right temporal EEG feature set, This is a global EEG feature set.

[0068] Step S2: Based on the first training sample set, construct a multi-level symptom-oriented feature extraction module and obtain multiple levels of symptom-oriented latent EEG feature variables;

[0069] Specifically, the multi-level symptom-guided feature extraction module includes:

[0070] The primary symptom-oriented latent variable extraction module, constructed using partial least squares based on symptom features and EEG features corresponding to each brain region and the whole brain in the first training sample set, is used to extract multiple primary EEG feature latent variables most relevant to the global symptom features for each brain region and the whole brain. ;

[0071] The secondary symptom-oriented latent variable extraction module is used to decouple the EEG characteristics of each brain region and the whole brain from the corresponding primary EEG latent variables, thereby obtaining the secondary EEG latent variables containing symptom specificity for each brain region and the whole brain. ;

[0072] The third-level subtype fusion latent variable extraction module is used to extract latent variables based on the second-level EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region and the whole brain. Based on each EEG subtype, feature extraction and fusion were performed to obtain a three-level EEG feature latent variable of fused symptom subtype features. .

[0073] Preferably, the primary symptom-oriented latent variable extraction module is constructed using the following method:

[0074] The mean of each symptom feature corresponding to each subject is used as the global symptom feature;

[0075] Based on the global symptom features and the corresponding EEG features of each brain region and the whole brain, a first-level global regression model corresponding to each brain region and the whole brain is obtained by training using partial least squares.

[0076] A first-level latent variable screening unit is constructed to obtain multiple first-level latent EEG feature variables that are most correlated with the global symptom features for each brain region and the whole brain based on the projected weights W of the first-level global regression model. ;

[0077] Based on the first-level global regression model and the first-level latent variable screening unit, the first-level symptom-oriented latent variable extraction module is obtained.

[0078] Specifically, this embodiment uses the mean of the symptom set as the global symptom feature, with a dimension of (m, 1), where m represents the sample size. As global symptom features, for each brain region and the whole brain's EEG feature set, as well as the global EEG feature set, a corresponding first-level global regression model was trained using partial least squares (PLS), including:

[0079] S21: Standardize the various EEG characteristics and symptom characteristics, and represent them as follows: , ;initialization and ,make , ;in, The mean of the EEG characteristic X is... y represents the mean of the symptom characteristic. The standard deviation of EEG characteristics, The standard deviation of the symptom characteristics This represents the EEG input features updated for the hth time with latent variable dimension H as the target. This represents the h-th update of the symptom input features, targeting the latent variable dimension H.

[0080] S22: with As the target variable, based on the formula By maximizing and linear combination and The covariance is calculated to obtain the h-th weight vector. ;in ;

[0081] Represented as: ;

[0082] S23: Based on and Obtain latent variables , is represented as: m is the sample size;

[0083] S24: Based on and latent variables The load of the h-th EEG feature was obtained. , is represented as: ;where the load This represents the association between the latent variable and X;

[0084] S25: Based on and latent variables To obtain the load of symptom characteristics , is represented as: ; among which the load This represents the association between the latent variable and y;

[0085] S26: Receive the next round of input, whereby the next round of input is represented as: , Iteratively execute S2-S5 to obtain the first-level global regression model corresponding to each brain region and the whole brain. and latent variable matrix EEG feature load matrix and model weight matrix Where W represents the brain region Model parameters.

[0086] For each brain region and global EEG features, the corresponding PLS model is obtained sequentially. And obtain the latent variables of global symptom features corresponding to each brain region and the whole brain EEG feature set. The latent variable is obtained from the projection weights W (dimension of original EEG feature dimension * preset latent variable dimension H) trained by PLS. Each column of W is the optimal linear combination direction of the original EEG features. This direction maximizes the correlation between the projection of the EEG feature in that direction and the global symptoms. Therefore, based on the first-level latent variable screening unit, multiple first-level EEG feature latent variables that are most relevant to the global symptom features can be selected. The latent variables obtained by linearly changing the projection weights W can retain the original feature set and the patterns related to global symptoms, while irrelevant patterns are compressed. The preset latent variable dimension H needs to be obtained through multiple comparisons using the cross-validation method.

[0087] Furthermore, the secondary symptom-guided specific latent variable extraction module is constructed using the following method:

[0088] Based on the EEG feature load matrix corresponding to each brain region and the whole brain Latent variables of primary EEG characteristics of each brain region and the whole brain Perform an inverse transformation to obtain the reconstructed original features. ;

[0089] Based on the difference between the original EEG features corresponding to each brain region and the whole brain and the reconstructed original features, the first-order residual features corresponding to each brain region and the whole brain are obtained. ;

[0090] For each symptom feature, the first-order residual feature is used separately. A corresponding second-order symptom-specific fitting model was established using partial least squares method. ; Utilizing the aforementioned secondary symptom-specific fitting model Inferring the secondary symptom-specific latent variables corresponding to each symptom characteristic ;

[0091] Construct fusion units to merge secondary symptom-specific latent variables corresponding to each symptom. And simultaneously merge the aforementioned primary EEG latent variables. Secondary EEG latent variables were obtained. ;

[0092] Based on the secondary symptom-specific fitting model The fusion unit is used to obtain the secondary symptom-guided specific latent variable extraction module.

[0093] It should be noted that this embodiment establishes a secondary symptom-oriented specific latent variable extraction module, which decouples the information of the original features and the global symptom features to obtain symptom-specific features.

[0094] For each brain region and the whole brain, the corresponding EEG feature latent variables are used, along with the corresponding EEG feature loading matrix from the first-level global regression model. Perform an inverse transform to reconstruct the original features. This reconstruction of the original features reconstructs the EEG manifestations of global symptom features. In its inverse transformation process, it is based on the corresponding first-level EEG feature latent variables. and EEG feature load matrix Using the reconstruction formula The reconstructed original features are obtained. The difference between these features and the original input EEG features is that the original EEG features include components related to symptom features, as well as individual differences or components related to symptom subtypes. The reconstructed features, however, only include components related to symptom features. Using G-... The first-order residual features are obtained, which contain symptom-specific patterns and individual differences.

[0095] For each symptom item, a second-level symptom-specific fitting model is established using the first-level residual features corresponding to each brain region and the whole brain respectively, through partial least squares method. The potential dimension N of each fitted model also needs to be determined through cross-validation; a pre-trained secondary symptom-specific fitting model is used. Inferring latent variables as symptom-specific latent variables For each brain region and whole-brain feature set, the symptom-specific latent variable and the first-order EEG feature latent variable are combined for each symptom sub-item. Then the input features of the third-level model are obtained. This feature includes both features related to global symptoms and specific features related to corresponding sub-symptoms.

[0096] Furthermore, the third-level subtype fusion latent variable extraction module includes:

[0097] The three-level subtype clustering module is used to perform clustering based on the secondary EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region and the whole brain. and the posterior probability of each subtype. ;

[0098] The latent variable extraction module from the perspective of subsymptom subtypes is used to infer subtype latent variables based on the PLS projection model of each pre-trained EEG subtype. ;

[0099] The subtype fusion module from the perspective of subsymptoms is used to combine the posterior probabilities of each EEG subtype. The values ​​are converted into deterministic weights, and the third-level EEG feature latent variables are obtained based on these deterministic weights. .

[0100] The three-level subtype clustering module is constructed using the following method:

[0101] Based on a preset target cluster size range, secondary EEG latent variables of each brain region and the whole brain are utilized. Train Gaussian mixture models corresponding to different target cluster numbers in sequence;

[0102] Based on the Bayesian Information Criterion (BIC) value of each Gaussian mixture model, the optimal number of target clusters corresponding to each symptom in each brain region is obtained;

[0103] The Gaussian mixture model corresponding to the optimal target cluster number is used as a three-level subtype clustering module for inferring clustering results.

[0104] It should be noted that this clustering module uses latent variables of global symptom features and sub-symptom features to establish sub-type features for each sub-symptom. The purpose is to use data to drive the identification of EEG-related subtypes under a given symptom. For each symptom item, corresponding secondary EEG feature latent variables are used in each brain region and in the whole brain subset of EEG features. Features are subjected to Gaussian mixture clustering. The target number of clusters in the Gaussian mixture model is determined using the Bayesian Information Criterion (BIC). Specifically, within a fixed range (from 2 to the maximum candidate value), Gaussian mixture models with different target cluster numbers (number of clusters) are trained sequentially, and the BIC score of each model is recorded. Since a smaller BIC value indicates a better balance between goodness of fit and complexity, the model with the lowest BIC score and the corresponding number of clusters is selected as the optimal solution. After determining the optimal target cluster number, the Gaussian mixture model is trained again using the target cluster number as a parameter. This model is then used to infer the clustering results, including the final EEG physiological subtype (i.e., cluster label). And the posterior probability of each subtype (i.e., the probability of the clustering result). .

[0105] Furthermore, the construction of the latent variable extraction module from the perspective of subsymptom subtypes includes:

[0106] A secondary symptom-specific fitting model based on each symptom feature in each brain region and the whole brain. The loading matrix parameters for the corresponding second-order symptom-specific latent variables Perform an inverse transformation to obtain the reconstructed specific features. ;

[0107] Utilizing the first-order residual characteristics Reconstruction-specific features The difference is used as a second-order residual characteristic. Based on the aforementioned electrophysiological subtypes Extract the secondary residual feature samples corresponding to each cluster;

[0108] Based on the secondary residual feature samples, a corresponding subtype PLS projection model is established for each cluster using partial least squares method. Complete the construction of the latent variable extraction module from the perspective of the subsymptom subtype.

[0109] This embodiment establishes a latent variable extraction module from the perspective of subsymptom subtypes: based on clustering results, a PLS projection model is trained in each subtype to obtain the latent variable projection matrix. The aim is to better integrate subtype information into the feature extraction process.

[0110] For each symptom item, first-order residual features were used in the EEG feature sets for each brain region and the whole brain. Reconstruction-specific features The difference is used as a second-order residual characteristic. Because this feature subtracts components fully related to subsymptoms from the global symptom-related features, it includes specific components for each subsymptom subtype. Based on clustering labels. Extract the corresponding second-order residual features for each cluster. For each sample, a PLS projection model is built for each cluster. Cross-validation was used to determine the target projection dimension. For different subtypes of the same symptom subite in the same brain region, the maximum value of the target projection dimension of all subtypes was used as the target projection dimension of each subtype projection model to ensure the uniformity of feature dimensions of samples across different subtypes. Each subtype PLS projection model was trained and the subtype latent variables were inferred. .

[0111] Furthermore, the subtype fusion module from the subsymptom perspective obtains the three-level symptom subtype fusion features through the following method:

[0112] The posterior probabilities of each subtype The entropy transformed into a subtype probability distribution is expressed as:

[0113] ;

[0114] Calculate maximum entropy Where N represents the number of target subtypes;

[0115] use This normalizes the entropy into a deterministic fraction.

[0116] Based on the subtype latent variables The deterministic scores yielded the third-level EEG latent variables. .

[0117] This embodiment establishes a subtype fusion module from the perspective of subsymptoms, uses deterministic computation methods to enhance the feature effects of subtypes with high determinism, and weakens the influence of features of subtypes with low determinism.

[0118] Step S3: Based on the latent variables of EEG features at each level of multiple subjects, label the corresponding mood disorder diagnostic tags to construct a second training sample set; use the second training sample set to iteratively train the feature fusion module and the classification module;

[0119] Specifically, the feature fusion module merges the first to third level EEG feature latent variables corresponding to each brain region of each sample using a weighted fusion method, and merges the weighted fused EEG feature latent variables at the brain region level to obtain multi-level symptom-oriented fusion features.

[0120] A classification module, such as a machine learning or deep learning classifier, is constructed. Utilizing the multi-level symptom-oriented fusion features and corresponding mood disorder diagnostic labels, the feature fusion module and classification module are iteratively trained to obtain the optimal weights of the feature fusion module and a converged classification module. Based on the classification module, the mood disorder assessment result is output.

[0121] During training, the cross-entropy loss function can be used for iterative optimization, but this embodiment does not limit it.

[0122] Step S4: Based on the multi-level symptom-oriented feature extraction module, the convergent feature fusion module, and the classification module, a mood disorder assessment model is obtained.

[0123] The mood disorder assessment model constructed based on this embodiment realizes multi-level EEG feature extraction and fusion based on symptom information constraints, thereby improving the model's ability to assess mood disorders.

[0124] Another embodiment of the present invention also discloses a computer device, including at least one processor and at least one memory communicatively connected to said processor;

[0125] The memory stores instructions that can be executed by the processor to implement the aforementioned method for constructing a mood disorder assessment model that integrates EEG and symptom-oriented approaches.

[0126] In summary, the method and device for constructing a mood disorder assessment model that integrates EEG and symptom-guided modeling of the present invention employs a multi-level modeling strategy for EEG and symptom-guided fusion, and a brain region segmentation strategy. An independent EEG feature latent variable extraction model is established for each brain region. Each brain region uses features generated by EEG electrodes corresponding to its spatial location, and different levels of models are used for feature extraction to capture the unique functional patterns of each brain region. A fusion strategy is used for global coordination, achieving EEG feature extraction based on symptom information constraints, thus improving the model's assessment ability for mood disorders with different functional abnormalities. Furthermore, the present invention uses multi-dimensional clinical symptom indicators. Through a multi-level modeling strategy, not only is a global symptom fitting projection module established, but linear fitting models are also constructed for multi-dimensional clinical features. The projection matrix is ​​used as the feature matrix, and the fitting results for multiple clinical symptom dimensions are output, providing symptom references for doctors' diagnostic decisions. By establishing a mapping relationship between EEG features and the symptom space through the fitting projection mechanism, clinical indicators are only used to train the projection matrix, realizing the conversion from EEG to the symptom semantic space. This ensures that the final classification is mainly based on objective electrophysiological indicators, maintaining the objectivity of the diagnosis.

[0127] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0128] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for constructing a mood disorder assessment model that integrates electroencephalography (EEG) and symptom-guided approaches, characterized in that, include: Acquire multiple mood disorder-related symptom characteristics from multiple subjects; The electroencephalogram (EEG) characteristics of multiple brain regions of each subject were collected to construct the first training sample set; the symptom characteristics include the symptom assessment results of each symptom item obtained based on the mood disorder related assessment scale; Based on the first training sample set, a multi-level symptom-oriented feature extraction module is constructed, and multiple levels of symptom-oriented latent EEG feature variables are obtained. The multi-level symptom-oriented feature extraction module includes: a first-level symptom-oriented latent variable extraction module, which is constructed using partial least squares method based on the symptom features in the first training sample set and the EEG features corresponding to each brain region. This module is used to extract multiple first-level EEG feature latent variables that are most relevant to the global symptom features for each brain region and the whole brain. The secondary symptom-oriented latent variable extraction module is used to decouple the EEG characteristics of each brain region and the whole brain from the corresponding primary EEG latent variables, thereby obtaining secondary EEG latent variables containing symptom specificity for each brain region and the whole brain. The third-level subtype fusion latent variable extraction module is used for the extraction of the second-level EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region. Based on each EEG subtype, feature extraction and fusion were performed to obtain a three-level EEG feature latent variable of fused symptom subtype features. ; The first-level symptom-oriented latent variable extraction module is constructed using the following method: the mean of each symptom feature corresponding to each subject is used as the global symptom feature; based on the global symptom feature and the EEG features corresponding to each brain region and the whole brain, a first-level global regression model corresponding to each brain region and the whole brain is trained using partial least squares; a first-level latent variable screening unit is constructed to obtain multiple first-level EEG feature latent variables that are most correlated with the global symptom feature based on the projected weights W of the first-level global regression model. Based on the first-level global regression model and the first-level latent variable screening unit, the first-level symptom-oriented latent variable extraction module is obtained. Based on the latent variables of EEG features at each level corresponding to multiple subjects, corresponding mood disorder diagnostic labels are labeled to construct a second training sample set; and the feature fusion module and classification module are iteratively trained using the second training sample set. Based on the multi-level symptom-oriented feature extraction module, the convergent feature fusion module, and the classification module, a mood disorder assessment model is obtained.

2. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 1, characterized in that, The first-level global regression model is trained using the following method: S1: Standardize the various EEG characteristics and symptom characteristics, and represent them as follows: , ;initialization and ,make , ;in, The mean of the EEG characteristic X is... y represents the mean of the symptom characteristic. The standard deviation of EEG characteristics, The standard deviation of the symptom characteristics This represents the H-th update of EEG feature X, with latent variable dimension H as the target. Let y represent the symptom feature updated for the h-th time with the latent variable dimension H as the target. S2: with As the target variable, based on the formula By maximizing and linear combination and The covariance is calculated to obtain the h-th weight vector. , is represented as: ； S3: Based on and Obtain latent variables , is represented as: m is the number of samples; S4: Based on and latent variables The load of the h-th EEG feature was obtained. , is represented as: ; S5: Based on and latent variables To obtain the load of symptom characteristics , is represented as: ; S6: Receive the next round of input, where the next round of input is represented as: , Iteratively execute S2-S5 to obtain the first-level global regression model corresponding to each brain region and the whole brain. and latent variable matrix EEG feature load matrix and model weight matrix .

3. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 2, characterized in that, The secondary symptom-guided specific latent variable extraction module was constructed using the following method: Based on the EEG feature load matrix corresponding to each brain region and the whole brain Latent variables of primary EEG characteristics of each brain region and the whole brain Perform an inverse transformation to obtain the reconstructed original features. ; Based on the difference between the original EEG features corresponding to each brain region and the whole brain and the reconstructed original features, the first-order residual features corresponding to each brain region and the whole brain are obtained. ; For each symptom feature, the first-order residual feature is used separately. A corresponding second-order symptom-specific fitting model was established using partial least squares method. ; Using the aforementioned secondary symptom-specific fitting model Inferring the secondary symptom-specific latent variables corresponding to each symptom characteristic ; Construct fusion units to merge secondary symptom-specific latent variables corresponding to each symptom. And simultaneously merge the aforementioned primary EEG latent variables. Secondary EEG latent variables were obtained. ; Based on the secondary symptom-specific fitting model The fusion unit is used to obtain the secondary symptom-guided specific latent variable extraction module.

4. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 3, characterized in that, The three-level subtype fusion latent variable extraction module includes: The three-level subtype clustering module is used to perform clustering based on the secondary EEG feature latent variables. Clustering was performed to obtain multiple electroencephalographic subtypes corresponding to each brain region and the whole brain. and the posterior probability of each subtype. ; The latent variable extraction module from the perspective of subsymptom subtypes is used to infer subtype latent variables based on the PLS projection model of each pre-trained EEG subtype. ; The subtype fusion module from the perspective of subsymptoms is used to combine the posterior probabilities of each EEG subtype. The values ​​are converted into deterministic weights, and the third-level EEG feature latent variables are obtained based on these deterministic weights. .

5. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 4, characterized in that, The three-level subtype clustering module is constructed using the following method: Based on a preset target cluster size range, secondary EEG latent variables of each brain region and the whole brain are utilized. Train Gaussian mixture models corresponding to different target cluster numbers in sequence; The optimal number of target clusters is obtained based on the Bayesian Information Criterion (BIC) value of each Gaussian mixture model. The Gaussian mixture model corresponding to the optimal target cluster number is used as a three-level subtype clustering module for inferring clustering results.

6. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 5, characterized in that, The construction of the latent variable extraction module from the perspective of subsymptom subtypes includes: A secondary symptom-specific fitting model based on each symptom feature in each brain region and the whole brain. The loading matrix parameters for the corresponding second-order symptom-specific latent variables Perform an inverse transformation to obtain the reconstructed specific features. ; Utilizing the first-order residual characteristics Reconstruction-specific features The difference is used as a second-order residual characteristic. Based on the aforementioned electrophysiological subtypes Extract the secondary residual feature samples corresponding to each cluster; Based on the secondary residual feature samples, a corresponding subtype PLS projection model is established for each cluster using partial least squares method. Complete the construction of the latent variable extraction module from the perspective of the subsymptom subtype.

7. The method for constructing a mood disorder assessment model that integrates EEG and symptom-guided approaches according to claim 6, characterized in that, The subtype fusion module from the perspective of sub-symptoms obtains the three-level symptom subtype fusion features through the following method: The posterior probabilities of each subtype The entropy transformed into a subtype probability distribution is expressed as: ； Calculate maximum entropy Where N represents the number of target subtypes; use This normalizes the entropy into a deterministic fraction. Based on the subtype latent variables The deterministic scores yielded the third-level EEG latent variables. .

8. A computer device, characterized in that, It includes at least one processor and at least one memory communicatively connected to the processor; The memory stores instructions that can be executed by the processor to implement the method for constructing a mood disorder assessment model by fusion of EEG and symptom guidance as described in any one of claims 1-7.

Images