A processing method and device for brain efficiency prediction based on multi-dimensional indexes

Abstract

The embodiment of the present application relates to a kind of processing method and device for brain efficiency prediction based on multidimensional index, the method comprises: the multidimensional index set corresponding to multidimensional index data set for brain efficiency prediction is set to obtain;A deep learning model for brain efficiency prediction is constructed as corresponding brain efficiency prediction model;Through the way of Chinese and Western medicine examination and brain efficiency expert evaluation to pre-recruited volunteer population, model data set is constructed and is recorded as corresponding first data set;Brain efficiency prediction model is trained based on the first data set;After model training ends, brain efficiency index prediction is carried out according to the multidimensional index set of the measured person based on brain efficiency prediction model. Through the present application, the prediction efficiency and prediction accuracy can be improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a processing method and apparatus for predicting brain efficacy based on multidimensional indicators. Background Technology

[0002] Mental efficacy is a concept that comprehensively reflects the brain's efficiency and effectiveness in information processing, encompassing multiple dimensions such as cognition, emotion, physical function, and social participation. Traditional brain efficacy prediction mechanisms often focus on a single dimension, such as conducting isolated analyses using neuropsychological scales or neuroimaging indicators, and have not yet formed a comprehensive multi-dimensional prediction system covering physiological, psychological, and traditional Chinese medicine syndromes. Due to the limitations of single-dimensional data characteristics, the accuracy of traditional brain efficacy prediction schemes is difficult to improve further after reaching a certain level. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by providing a processing method, device, electronic device, and computer-readable storage medium for predicting brain efficacy based on multidimensional indicators. This invention first customizes a multidimensional (self-evaluation, neurological, cognitive psychology, and traditional Chinese medicine syndrome) indicator dataset for brain efficacy prediction and constructs a brain efficacy prediction model capable of predicting brain efficacy indices based on the multidimensional indicator set. Then, a model dataset is constructed by conducting traditional Chinese and Western medicine examinations and brain efficacy expert assessments on a pre-recruited volunteer population, and the brain efficacy prediction model is trained based on this dataset. After model training, the brain efficacy prediction model is used to predict the brain efficacy index of any subject. This invention improves feature richness and prediction accuracy through a multidimensional comprehensive prediction system integrating subjective and objective evaluation systems and traditional Chinese and Western medicine evaluation systems; it improves prediction performance and efficiency through the brain efficacy prediction model; and it enhances model robustness and generalization through a masking strategy during model training.

[0004] To achieve the above objectives, a first aspect of the present invention provides a method for predicting brain efficacy based on multidimensional indicators, the method comprising: A multidimensional indicator set is obtained by setting up a multidimensional indicator dataset for brain efficacy prediction; the multidimensional indicator set consists of four sub-datasets, namely, self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set and traditional Chinese medicine syndrome indicator set. A deep learning model for predicting brain efficacy is constructed as the corresponding brain efficacy prediction model; the brain efficacy prediction model is used to predict brain efficacy based on the multidimensional index set input to the model and output the corresponding predicted efficacy index. The model dataset was constructed by conducting both traditional Chinese and Western medicine examinations and brain function expert assessments on a pre-recruited volunteer population, and is denoted as the corresponding first dataset. The brain efficacy prediction model is trained based on the first dataset; After the model training is completed, the multidimensional index set of the test subject input by the user is input into the brain efficacy prediction model for prediction, and the predicted efficacy index output by this prediction is fed back to the current user.

[0005] Preferably, the self-evaluation index set consists of multiple categories of self-evaluation indicators, including at least: learning ability self-evaluation indicators, memory self-evaluation indicators, attention self-evaluation indicators, energy and physical strength self-evaluation indicators, emotional stability self-evaluation indicators, sleep recovery ability self-evaluation indicators, willpower self-evaluation indicators, control ability self-evaluation indicators, execution ability self-evaluation indicators, social vitality self-evaluation indicators, life vitality self-evaluation indicators, self-confidence self-evaluation indicators, and self-awareness self-evaluation indicators; each category of self-evaluation indicator is a self-evaluation score; all self-evaluation indicators have the same score range, which is a preset self-evaluation score threshold range; The set of neurological indicators comprises multiple categories of neurological indicators, including at least: brain region volume indicators, white matter integrity indicators, cortical thickness indicators, functional connectivity density indicators, neural oscillation synchronicity indicators, and causal interaction indicators. Specifically, the brain region volume indicator is the volume of the hippocampus; the white matter integrity indicator is the partial anisotropy score of the corpus callosum; the cortical thickness indicator is the thickness of the prefrontal cortex; the functional connectivity density indicator is the global connectivity strength of the default mode network; the neural oscillation synchronicity indicator is the alpha wave rhythm phase lock value on an electroencephalogram (EEG); and the causal interaction indicator is the average intensity of the causal flow from a pre-defined activated brain region to one or more other pre-defined brain regions. The cognitive psychological indicator set consists of multiple cognitive psychological indicators, including at least: MMSE assessment indicators, MoCA assessment indicators, AVLT assessment indicators, WMS assessment indicators, VFT assessment indicators, BNT assessment indicators, TMT-A assessment indicators, TMT-B assessment indicators, CDT assessment indicators, and Stroop. The assessment indicators include: ADL (Activities of Daily Living) assessment indicators, NPI (Natural Physical Scale) assessment indicators, HAMA (Hamiltonian Anxiety Scale) assessment indicators, and HAMD (Hamiltonian Depression Scale) assessment indicators. Specifically, the MMSE (Montreal Mental State Examination) assessment indicator is the score of the Mini-Mental State Examination; the MoCA (Montreal Cognitive Assessment) assessment indicator is the score of the Montreal Cognitive Assessment; the AVLT (Audio-Verbal Learning Test) assessment indicator is the score of the Auditory-Verbal Learning Test; the WMS (Wechsler Memory Scale) assessment indicator is the score of the Wechsler Memory Scale; the VFT (Verbal Fluency Test) assessment indicator is the score of the Verbal Fluency Test; the BNT (Boston Naming Test) assessment indicator is the score of the Boston Naming Test; the TMT-A and TB (Tracking Test A and B) assessment indicators are the scores of the Connecting Test A and B; the CDT (Clock Drawing Test) assessment indicator is the score of the Clock Drawing Test; the Stroop Test assessment indicator is the score of the Stroop Test; the ADL (Activities of Daily Living) assessment indicator is the score of the Daily Living Skills Scale; the NPI (Natural Physical Scale) assessment indicator is the score of the Neuropsychiatric Questionnaire; the HAMA (Hamiltonian Anxiety Scale) assessment indicator is the score of the Hamilton Anxiety Scale; and the HAMD (Hamiltonian Depression Scale) assessment indicator is the score of the Hamilton Depression Rating Scale. The TCM syndrome index set consists of multiple TCM syndrome indices, including at least: the syndrome index of gradual depletion of marrow and sea, the syndrome index of spleen and kidney deficiency, the syndrome index of qi and blood deficiency, the syndrome index of phlegm obstructing the orifices, the syndrome index of blood stasis obstructing the brain collaterals, the syndrome index of heart and liver fire excess, and the syndrome index of extreme deficiency due to toxicity. Each syndrome index is a binary index, specifically yes or no. The first dataset includes multiple first data records; each first data record includes a first training indicator set and a first label index; the first training indicator set is a multidimensional indicator set; the self-evaluation indicator set of the first training indicator set maintains data integrity and has no invalid indicators; if there are preset invalid indicators in the first training indicator set, the ratio of the total number of invalid indicators to the total number of all indicators satisfies a preset first masking rate, and all invalid indicators are randomly distributed in the neurological indicator set, the cognitive psychology indicator set, and the traditional Chinese medicine syndrome indicator set; the total number of invalid indicators is the total number of invalid indicators, and the total number of all indicators is the total number of indicators in the multidimensional indicator set.

[0006] Preferably, the model input terminal of the brain efficacy prediction model is used to receive the multidimensional index set, and the model output terminal is used to output the corresponding prediction efficacy index. The brain efficacy prediction model includes a preprocessing module, a self-evaluation feature encoding module, a neurological feature encoding module, a cognitive psychological feature encoding module, a traditional Chinese medicine syndrome feature encoding module, a Western medicine feature fusion module, a traditional Chinese and Western medicine feature fusion module, a subjective and objective feature fusion module, a feature mapping module, and an index prediction module. The input terminal of the preprocessing module is connected to the input terminal of the model, and the first, second, third, and fourth output terminals are respectively connected to the input terminals of the self-evaluation feature encoding module, the neurological feature encoding module, the cognitive psychological feature encoding module, and the traditional Chinese medicine syndrome feature encoding module; the output terminals of the neurological feature encoding module and the cognitive psychological feature encoding module are connected to the first and second input terminals of the Western medicine feature fusion module; the output terminals of the traditional Chinese medicine syndrome feature encoding module and the Western medicine feature fusion module are connected to the first and second input terminals of the traditional Chinese and Western medicine feature fusion module; the output terminals of the self-evaluation feature encoding module and the traditional Chinese and Western medicine feature fusion module are connected to the first and second input terminals of the subjective and objective feature fusion module; the output terminal of the subjective and objective feature fusion module is connected to the input terminal of the feature mapping module; the output terminal of the feature mapping module is connected to the input terminal of the index prediction module; and the output terminal of the index prediction module is connected to the model output terminal. The preprocessing module is used to take the self-evaluation index set, the neurological index set, the cognitive psychological index set, and the traditional Chinese medicine syndrome index set of the multidimensional index set as the corresponding self-evaluation index vector X. 1 Neurological index vector X 2 Cognitive psychological indicator vector X 3 and the TCM syndrome index vector X 4 The self-evaluation index vector X is sent to the corresponding self-evaluation feature encoding module, the neurological feature encoding module, the cognitive psychological feature encoding module, and the traditional Chinese medicine syndrome feature encoding module. 1 The shape is C1×1, and the neurological index vector X 2 The shape is C2×1, and the cognitive psychological index vector X 3 The shape is C3×1, and the TCM syndrome index vector X 4 The shape is C4×1, where C1, C2, C3, and C4 are the preset first, second, third, and fourth feature dimensions; The self-evaluation feature encoding module is implemented based on an MLP model; the self-evaluation feature encoding module is used to encode the self-evaluation index vector X. 1 The corresponding feature tensor H is obtained by performing feature encoding processing. 1 Send to the subjective and objective feature fusion module; The feature tensor H 1 The reasoning process is as follows: , , , ; Among them, W 11 W 12 Let b be the two weight matrix parameters of the self-evaluation feature encoding module. 11 b 12 h are the two offset vector parameters of the self-evaluation feature encoding module. 11 h 12 The two process feature vectors of the self-evaluation feature encoding module; GELU() is the GELU activation function; the weight matrix parameter W 11 The shape is C5×C1, and the offset vector parameter is b. 11 The shape is C5×1; the weight matrix parameter W 12 The shape is C6×C5, and the offset vector parameter is b. 12 The shape is C6×1; the process feature vector h 11 h 12 The shapes are C5×1 and C6×1; Reshape() is a vector shape reshaping function used to reshape the process feature vector h with a shape of C6×1. 12 This is converted into a feature tensor H of shape C1×C7. 1 C5, C6, and C7 are the preset fifth, sixth, and seventh feature dimensions, respectively, where C1 < C5 < C6 and C7 = C6 / C1. The neurological feature encoding module is implemented based on an MLP model; the neurological feature encoding module is used to process the neurological index vector X. 2 The corresponding feature tensor H is obtained by performing feature encoding processing. 2 Send to the Western medicine feature fusion module; The feature tensor H 2 The reasoning process is as follows: , , ; Among them, W 21 W 22 Let b be the two weight matrix parameters of the neurological feature encoding module. 21 b 22 h are the two offset vector parameters of the neurological feature encoding module. 21 h 22 The two process feature vectors of the neurological feature encoding module; weight matrix parameter W 21The shape is C8×C2, and the offset vector parameter is b. 21 The shape is C8×1; the weight matrix parameter W 22 The shape is C9×C8, and the offset vector parameter is b. 22 The shape is C9×1; the process feature vector h 21 h 22 The shapes are C8×1 and C9×1; the vector shape reshaping function Reshape() is used to reshape the process feature vector h with shape C9×1. 22 This is converted into a feature tensor H of shape C1×C7. 2 C8 and C9 are the preset eighth and ninth feature dimensions, respectively, where C2 < C8 < C9 and C9 = C1 × C7. The cognitive psychological feature encoding module is implemented based on an MLP model; the cognitive psychological feature encoding module is used to process the cognitive psychological index vector X. 3 The corresponding feature tensor H is obtained by performing feature encoding processing. 3 Send to the Western medicine feature fusion module; The feature tensor H 3 The reasoning process is as follows: , , ; Among them, W 31 W 32 Let b be the two weight matrix parameters of the cognitive psychological feature encoding module. 31 b 32 h are the two offset vector parameters of the cognitive psychological feature encoding module. 31 h 32 The two process feature vectors of the cognitive psychological feature encoding module; weight matrix parameter W 31 The shape is C 10 ×C3, offset vector parameter b 31 The shape is C 10 ×1; Weight matrix parameter W 32 The shape is C 11 ×C 10 offset vector parameter b 32 The shape is C 11 ×1; Process feature vector h 31 h 32 The shape is C 10 ×1、C 11 ×1; The vector shape reshaping function Reshape() is used to reshape the vector into a shape of C. 11 ×1 process feature vector h 32This is converted into a feature tensor H of shape C1×C7. 3 C 10 C 11 For the preset tenth and eleventh feature dimensions, C3 < C 10 <C 11 C 11 =C1×C7; The TCM syndrome feature encoding module is implemented based on an MLP model; the TCM syndrome feature encoding module is used to encode the TCM syndrome index vector X. 4 The corresponding feature tensor H is obtained by performing feature encoding processing. 4 Send to the Chinese and Western medicine feature fusion module; The feature tensor H 4 The reasoning process is as follows: , , ; Among them, W 41 W 42 b are the two weight matrix parameters of the TCM syndrome feature encoding module. 41 b 42 h are the two offset vector parameters of the TCM syndrome feature encoding module. 41 h 42 The two process feature vectors of the TCM syndrome feature encoding module; weight matrix parameter W 41 The shape is C 12 ×C4, offset vector parameter b 41 The shape is C 12 ×1; Weight matrix parameter W 42 The shape is C 13 ×C 12 offset vector parameter b 42 The shape is C 13 ×1; Process feature vector h 41 h 42 The shape is C 12 ×1、C 13 ×1; The vector shape reshaping function Reshape() is used to reshape the vector into a shape of C. 13 ×1 process feature vector h 42 This is converted into a feature tensor H of shape C1×C7. 4 C 12 C 13 For the preset twelfth and thirteenth feature dimensions, C4 < C 12 <C 13 C 13 =C1×C7; The Western medicine feature fusion module is used to perform fusion based on the feature tensor H. 2 Generate the corresponding query tensor Q, and based on the feature tensor H 3 Generate corresponding key-value tensors K and V; and perform attention feature extraction processing on the query, key, and value tensors Q, K, and V to obtain the corresponding attention tensor A; and process the feature tensor H 2 The corresponding feature tensor H is obtained by performing a residual connection with the attention tensor A. 5 Send to the Chinese and Western medicine feature fusion module; The feature tensor H 5 The reasoning process is as follows: , , , , ; Among them, W Q W K W V The three weight matrix parameters are: W, ... Q W K W V The shapes are all C7×C qkv Preset feature dimension C qkv =C7; the query tensor Q, the key tensor K, the value tensor V, the attention tensor A, and the feature tensor H 5 The shapes are all C1×C7; Softmax() is the Softmax function; The traditional Chinese and Western medicine feature fusion module is used to process the feature tensor H. 4 and the characteristic tensor H 5 Feature splicing yields a shape of C1×C 14 Feature tensor H 6 C 14 C is the preset fourteenth feature dimension. 14 =2×C7; and by applying the feature tensor H 6 A fully connected operation is performed to obtain a gated weight tensor G of shape C1×1; and C7 of these gated weight tensors G are combined to form a gated weight tensor G of shape C1×C7. ’ ; and based on the gated weight tensor G ’ The feature tensor H 4 and the characteristic tensor H 5 Feature fusion processing is performed to obtain the corresponding feature tensor H. 7 Send to the subjective and objective feature fusion module; The feature tensor H 7 The reasoning process is as follows: , ; Among them, W G b G The weight matrix parameters and offset vector parameters of the traditional Chinese and Western medicine feature fusion module are: weight matrix parameter W. G The shape is C 14 ×1, Offset vector parameter b G The shape of the gate weight tensor G is C1×1; Sigmoid() is the Sigmoid activation function; the shape of the gate weight tensor G is C1×1; I is a preset all-1 tensor with a shape of C1×C7; ⊙ is the Hadamard product; the feature tensor H 7 The shape is C1×C7; The subjective and objective feature fusion module is used to process the feature tensor H. 1 and the characteristic tensor H 7 Feature fusion processing is performed to obtain the corresponding feature tensor H. 8 Send to the feature mapping module; The feature tensor H 8 The reasoning process is as follows: ; The feature tensor H 8 The shape is C1×C7; The feature mapping module is used to map the feature tensor H using a 1×1 convolution kernel. 8 Feature reduction processing yields a shape of C1×C 15 Feature tensor H 9 C 15 As the preset fifteenth feature dimension, C1 < C 15 <C7; and for the characteristic tensor H 9 Global max pooling along the metric dimension yields a 1×C value. 15 eigenvector H 10 ; and for the feature vector H 10 Perform a fully connected operation to obtain the corresponding feature scalar H 11 Send to the index prediction module; The feature scalar H 11 The reasoning process is as follows: ; Among them, W 51 b 51 The weight matrix parameters and offset vector parameters of the feature mapping module; weight matrix parameter W51 The shape is C 15 ×C 16 Offset vector parameter b 51 The shape is 1×C 16 C 16 C is the preset sixteenth feature dimension. 16 =1; The index prediction module is used to predict the index based on a preset weight scalar W. 61 Offset scalar b 61 For the feature scalar H 11 Perform linear scaling to obtain the corresponding predicted performance index and output it; Predicted performance index = W 61 ×H 11 +b 61 .

[0007] Preferably, the model dataset constructed by conducting traditional Chinese and Western medicine examinations and brain efficacy expert assessments on a pre-recruited volunteer population is denoted as the corresponding first dataset, and specifically includes: Each volunteer in the volunteer group is taken as the corresponding current volunteer; The self-assessment indicators obtained by the current volunteer based on the preset brain efficacy self-assessment scale and corresponding scale self-assessment rules constitute a corresponding self-assessment indicator set; the brain efficacy self-assessment scale includes self-assessment items for learning ability, memory, attention, energy and physical strength, emotional stability, sleep recovery ability, willpower, control, executive function, social vitality, life vitality, self-confidence, and self-awareness; the self-assessment items of the brain efficacy self-assessment scale correspond one-to-one with the self-assessment indicators in the self-assessment indicator set; The self-assessment rules of the scale include self-assessment rules for learning ability, memory, attention, energy and physical strength, emotional stability, sleep recovery ability, willpower, control, executive function, social vitality, life vitality, self-confidence, and self-awareness. Each self-assessment rule corresponds one-to-one with the self-assessment items of the brain efficacy self-assessment scale. Each self-assessment rule is used to grade the score of the current self-assessment item and provides an explanation of the scoring rules corresponding to each grade. The first expert group, comprised of medical experts, conducted resting-state structural magnetic resonance imaging (SMRI), diffusion tensor imaging (DTI), and functional magnetic resonance imaging (fMRI) examinations on the current volunteers to obtain corresponding sMRI, DTI, and fMRI images. Based on the sMRI images, hippocampal brain regions were segmented and their volume measured using FreeSurfer, FSL, or SPM analysis tools, and the obtained hippocampal volume was used as the corresponding brain region volume index. Based on the sMRI images, the cerebral cortex surface was reconstructed using FreeSurfer or CAT12 analysis tools, and the prefrontal cortex thickness was measured on the reconstructed cortical surface, with the obtained prefrontal cortex thickness used as the corresponding cortical thickness index. Based on the DTI images, the partial anisotropy score of the corpus callosum was calculated using FSL analysis tools, and the calculated score was used as the corresponding white matter integrity index. Based on the fMRI images, the global connectivity strength of the brain's default mode network was analyzed using CONN or GRETNA analysis tools, and the analysis results were used as the corresponding functional connectivity density index. The first expert group was a medical imaging expert group. The pre-designated second expert group of medical experts will conduct an electroencephalogram (EEG) examination on the current volunteer to obtain the corresponding first EEG image; and based on the first EEG image, alpha wave rhythm phase lock value analysis will be performed using EEGLAB, FieldTrip, or Chronux analysis tools, and the analysis results will be used as the corresponding neural oscillation synchronicity index; the second expert group is a neurophysiology and neuroscience expert group; A pre-designated third expert group of medical experts performed transcranial magnetic stimulation (TMS) on the activated brain regions of the current volunteer. Simultaneously, electroencephalography (EEG) was performed on the activated brain regions and one or more other brain regions to obtain corresponding second and third EEG maps. Following Granger causality analysis, the causal flow intensity between the activated brain regions and each of the other brain regions was analyzed based on the second and third EEG maps. The mean of all obtained causal flow intensities was calculated, and the result was used as the corresponding causal interaction index. The second EEG map corresponds to the activated brain region; the third EEG map set consists of one or more third EEG maps, each corresponding one-to-one with the other brain regions; the third expert group is a neurophysiology and neuroscience expert group. The cognitive psychological indicators of the current volunteer will be evaluated by medical experts from a pre-designated fourth expert group; the fourth expert group consists of experts in neuropsychology, behavioral neurology, psychiatry, and neurology. The medical experts of the pre-designated fifth expert group will evaluate all the TCM syndrome indicators of the current volunteer's TCM syndrome indicator set; the fifth expert group is an expert group in the field of TCM. A joint expert group composed of the first, second, third, fourth and fifth expert groups will assess and score the current volunteer's brain efficacy status and use the obtained assessment score as the corresponding first label index; A set of neurological indicators is formed by the brain region volume index, white matter integrity index, cortical thickness index, functional connectivity density index, neural oscillation synchronicity index, and causal interaction index corresponding to the current volunteer; and a set of training indicators is formed by the self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set, and traditional Chinese medicine syndrome indicator set corresponding to the current volunteer. Based on the first mask rate and the invalid index, the indicators of the neurological indicator set, the cognitive psychological indicator set, and the traditional Chinese medicine syndrome indicator set of the current first training indicator set are randomly replaced once to obtain a derived first training indicator set, and the random replacement process is repeated N times to obtain N derived first training indicator sets; N is a preset positive integer. The first data record is composed of each set of the first training indicators and the corresponding first label index corresponding to the current volunteer; and the first data record set is composed of N+1 first data records corresponding to the current volunteer. The first dataset is composed of all the first data records corresponding to all volunteers in the volunteer population.

[0008] Preferably, training the brain efficacy prediction model based on the first dataset specifically includes: Step 51: Based on a preset first segmentation ratio, the first dataset is divided into two sub-datasets, denoted as the first training set and the first evaluation set. Wherein, both the first training set and the first evaluation set are composed of multiple first data records; the ratio of the total number of records in the first training set and the first evaluation set satisfies the first segmentation ratio; Step 52: Perform a round of traversal on all the first data records in the first training set; during this round of traversal, take the currently traversed first data record as the corresponding current training record; and input the first training index set of the current training record as the current multidimensional index set into the brain efficacy prediction model for prediction, and take the prediction efficacy index output by this prediction as the corresponding first prediction index; and form a corresponding first prediction-label pair with the first prediction index and the first label index of the current training record; and at the end of this round of traversal, substitute all the first prediction-label pairs obtained in this round of traversal into the preset first model loss function to calculate the corresponding first loss value; The first model loss function is implemented based on the L1 loss function or the L2 loss function; Step 53: Identify whether the first loss value meets the preset first loss value range; if it does, proceed to step 54; if it does not, modulate the model parameters of the brain efficacy prediction model in one round based on the preset first model optimizer in the direction of minimizing the first model loss function, and return to step 52 to continue training when the first round of modulation ends. The first model optimizer includes at least the Adam optimizer and the SGD optimizer; Step 54: Perform a traversal of all the first data records in the first evaluation set; during this traversal, the currently traversed first data record is taken as the corresponding current evaluation record; the first training index set of the current evaluation record is taken as the current multidimensional index set and input into the brain efficacy prediction model for prediction; the prediction efficacy index output by this prediction is taken as the corresponding second prediction index; the second prediction index and the first label index of the current evaluation record form a corresponding second prediction-label pair; at the end of this traversal, all the second prediction-label pairs obtained in this traversal are input into a preset first model evaluation function to calculate the corresponding first evaluation value; The first model evaluation function is implemented based on the MAE function, MSE function, or RMSE function. Step 55: Identify whether the first evaluation value meets the preset first evaluation value range; if not, return to step 52 to continue training; if yes, confirm that the model training is over.

[0009] A second aspect of the present invention provides an apparatus for implementing the processing method for predicting brain efficacy based on multidimensional indicators as described in the first aspect above. The apparatus includes: a multidimensional indicator system construction module, a model construction module, a dataset construction module, a model training module, and a model prediction module. The multidimensional indicator system construction module is used to set up the multidimensional indicator dataset for brain efficacy prediction to obtain the corresponding multidimensional indicator set; the multidimensional indicator set consists of four sub-datasets, namely, self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set and traditional Chinese medicine syndrome indicator set. The model building module is used to build a deep learning model for brain efficacy prediction as the corresponding brain efficacy prediction model; the brain efficacy prediction model is used to predict brain efficacy based on the multidimensional index set input to the model and output the corresponding predicted efficacy index. The dataset construction module is used to construct a model dataset by conducting Chinese and Western medicine examinations and brain efficacy expert evaluations on a pre-recruited volunteer population, denoted as the corresponding first dataset. The model training module is used to train the brain efficacy prediction model based on the first dataset; The model prediction module is used to input the multidimensional index set of the test subject input by the user into the brain efficacy prediction model after the model training is completed, and to feed back the predicted efficacy index output by the current prediction to the current user.

[0010] A third aspect of the present invention provides an electronic device, including: a memory, a processor, and a transceiver; The processor is used to couple with the memory, read and execute instructions in the memory to implement the steps of the method described in the first aspect above; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

[0011] A fourth aspect of the present invention provides a computer-readable storage medium storing computer instructions that, when executed by a computer, cause the computer to perform the instructions described in the first aspect.

[0012] This invention provides a method, apparatus, electronic device, and computer-readable storage medium for predicting brain efficacy based on multidimensional indicators. As described above, this invention first customizes a multidimensional (self-evaluation, neurological, cognitive psychology, and traditional Chinese medicine syndrome) indicator dataset for brain efficacy prediction and constructs a brain efficacy prediction model capable of predicting brain efficacy indices based on the multidimensional indicator set. Then, a model dataset is constructed by conducting traditional Chinese and Western medicine examinations and brain efficacy expert assessments on a pre-recruited volunteer population, and the brain efficacy prediction model is trained based on this dataset. After model training, the brain efficacy prediction model is used to predict the brain efficacy index of any subject. This invention improves data feature richness and prediction accuracy through a multidimensional comprehensive prediction system integrating subjective and objective evaluation systems and traditional Chinese and Western medicine evaluation systems; it improves prediction performance and efficiency through the use of a brain efficacy prediction model; and it enhances the robustness and generalization of the model through a masking strategy set during model training. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of a brain efficacy prediction method based on multidimensional indicators provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the data structure of the multidimensional index set provided in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the modules of the brain efficacy prediction model provided in Embodiment 1 of the present invention; Figure 4 This is a module structure diagram of a processing device for predicting brain efficacy based on multidimensional indicators, provided in Embodiment 2 of the present invention. Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0015] Embodiment 1 of the present invention provides a processing method for predicting brain efficacy based on multidimensional indicators, such as... Figure 1 The schematic diagram shows a processing method for predicting brain efficacy based on multidimensional indicators provided in Embodiment 1 of the present invention. The method mainly includes the following steps: Step 1: Set up the multidimensional indicator dataset for brain performance prediction to obtain the corresponding multidimensional indicator set.

[0016] Here, the multidimensional indicator set in this embodiment of the invention consists of four sub-data sets: a self-evaluation indicator set, a neurological indicator set, a cognitive psychological indicator set, and a traditional Chinese medicine syndrome indicator set, such as... Figure 2 The data structure diagram of the multidimensional index set provided in Embodiment 1 of the present invention is shown.

[0017] The self-evaluation index set in this invention consists of multiple categories of self-evaluation indicators, including at least: learning ability self-evaluation indicators, memory self-evaluation indicators, attention self-evaluation indicators, energy and physical strength self-evaluation indicators, emotional stability self-evaluation indicators, sleep recovery ability self-evaluation indicators, willpower self-evaluation indicators, control ability self-evaluation indicators, execution ability self-evaluation indicators, social vitality self-evaluation indicators, life vitality self-evaluation indicators, self-confidence self-evaluation indicators, and self-awareness self-evaluation indicators. Each category of self-evaluation indicator represents a self-evaluation score; all self-evaluation indicators have the same score range, which is a preset self-evaluation score threshold range. This self-evaluation score threshold range is a pre-set score range, for example, a score range of 0-10.

[0018] The neurological indicator set of this invention consists of multiple types of neurological indicators, including at least: brain region volume indicators, white matter integrity indicators, cortical thickness indicators, functional connectivity density indicators, neural oscillation synchronicity indicators, and causal interaction indicators. Specifically, the brain region volume indicator is the volume of the hippocampus; the white matter integrity indicator is the partial anisotropy (FA) score of the corpus callosum; the cortical thickness indicator is the thickness of the prefrontal cortex; the functional connectivity density indicator is the global connectivity strength of the Default Mode Network (DMN); the neural oscillation synchronicity indicator is the alpha wave rhythm phase locking value (PLV) of an electroencephalogram (EEG); and the causal interaction indicator is the average intensity of the causal flow from a pre-defined activated brain region to one or more other pre-defined brain regions.

[0019] The cognitive psychological indicator set consists of multiple cognitive psychological indicators, including at least: MMSE assessment indicators, MoCA assessment indicators, AVLT assessment indicators, WMS assessment indicators, VFT assessment indicators, BNT assessment indicators, TMT-A assessment indicators, TMT-B assessment indicators, CDT assessment indicators, Stroop test assessment indicators, ADL assessment indicators, NPI assessment indicators, HAMA assessment indicators, and HAMD assessment indicators. Specifically, the MMSE assessment indicator is the Mini-Mental State Examination (MMSE) score; the MoCA assessment indicator is the Montreal Cognitive Assessment (MoCA) score; the AVLT assessment indicator is the Auditory Verbal Learning Test (AVLT) score; the WMS assessment indicator is the Wechsler Memory Scale (WMS) score; the VFT assessment indicator is the Verbal Fluency Test (VFT) score; the BNT assessment indicator is the Boston Naming Test (BNT) score; the TMT-A and TMT-B assessment indicators are the Trail Making Tests A and B (TMT-A and TMT-A and TMT-B) scores; and the CDT assessment indicator is the Clock Drawing Test. The assessment indicators are as follows: Stroop Test (CDT) score; ADL score; Neuropsychiatric Inventory (NPI) score; Hamilton Anxiety Rating Scale (HAMA) score; and Hamilton Depression Rating Scale (HAMD) score.

[0020] The TCM syndrome indicator set consists of multiple TCM syndrome indicators, including at least: the syndrome indicators for gradual depletion of marrow and sea, the syndrome indicators for deficiency of both spleen and kidney, the syndrome indicators for deficiency of qi and blood, the syndrome indicators for phlegm obstructing the orifices, the syndrome indicators for blood stasis obstructing the brain collaterals, the syndrome indicators for excessive heart and liver fire, and the syndrome indicators for extreme deficiency due to excessive toxicity; each syndrome indicator is a binary indicator, specifically yes or no.

[0021] Step 2: Construct a deep learning model for predicting brain efficacy as the corresponding brain efficacy prediction model.

[0022] Here, the brain efficacy prediction model of this invention is used to predict brain efficacy based on the multidimensional index set input to the model and output the corresponding predicted efficacy index.

[0023] like Figure 3 As shown in the schematic diagram of the brain efficacy prediction model provided in Embodiment 1 of the present invention, the model input end of the brain efficacy prediction model is used to receive a multi-dimensional index set, and the model output end is used to output the corresponding prediction efficacy index.

[0024] The brain efficacy prediction model consists of the following components: a preprocessing module, a self-evaluation feature encoding module, a neurological feature encoding module, a cognitive psychological feature encoding module, a traditional Chinese medicine syndrome feature encoding module, a Western medicine feature fusion module, a traditional Chinese and Western medicine feature fusion module, a subjective and objective feature fusion module, a feature mapping module, and an index prediction module.

[0025] The connection relationships of the brain efficacy prediction model components are as follows: the input end of the preprocessing module is connected to the model input end; the first, second, third, and fourth output ends are connected to the input ends of the self-evaluation feature encoding module, the neurological feature encoding module, the cognitive psychological feature encoding module, and the traditional Chinese medicine syndrome feature encoding module, respectively; the output ends of the neurological feature encoding module and the cognitive psychological feature encoding module are connected to the first and second input ends of the Western medicine feature fusion module; the output ends of the traditional Chinese medicine syndrome feature encoding module and the Western medicine feature fusion module are connected to the first and second input ends of the traditional Chinese and Western medicine feature fusion module; the output ends of the self-evaluation feature encoding module and the traditional Chinese and Western medicine feature fusion module are connected to the first and second input ends of the subjective and objective feature fusion module; the output end of the subjective and objective feature fusion module is connected to the input end of the feature mapping module; the output end of the feature mapping module is connected to the input end of the index prediction module; and the output end of the index prediction module is connected to the model output end.

[0026] The model components of the brain efficacy prediction model are shown below.

[0027] 1) Preprocessing module: The preprocessing module is used to take the self-evaluation index set, neurological index set, cognitive psychological index set, and traditional Chinese medicine syndrome index set of the multidimensional index set as the corresponding self-evaluation index vector X. 1 Neurological index vector X 2 Cognitive psychological indicator vector X 3 and the TCM syndrome index vector X 4 Send to the corresponding self-evaluation feature coding module, neurological feature coding module, cognitive psychological feature coding module, and traditional Chinese medicine syndrome feature coding module.

[0028] Here, the self-evaluation index vector X1 The shape is C1×1, and the neurological index vector X 2 The shape is C2×1, and the cognitive psychological index vector X 3 The shape is C3×1, and the TCM syndrome index vector X 4 The shape is C4×1; where C1, C2, C3, and C4 are the preset first, second, third, and fourth feature dimensions, for example, C1=13, C2=6, C3=14, and C4=7.

[0029] 2) Self-evaluation feature encoding module: The self-evaluation feature encoding module is implemented based on an MLP model; the self-evaluation feature encoding module is used to encode the self-evaluation index vector X. 1 The corresponding feature tensor H is obtained by performing feature encoding processing. 1 Send to the subjective and objective feature fusion module.

[0030] Here, the feature tensor H 1 The reasoning process is as follows: , , , ; Among them, W 11 W 12 b are the two weight matrix parameters of the self-evaluation feature encoding module. 11 b 12 These are the two offset vector parameters of the self-evaluation feature encoding module; h 11 h 12 These are the two process feature vectors of the self-evaluation feature encoding module; GELU() is the GELU activation function; the weight matrix parameter W 11 The shape is C5×C1, and the offset vector parameter is b. 11 The shape is C5×1; the weight matrix parameter W 12 The shape is C6×C5, and the offset vector parameter is b. 12 The shape is C6×1; the process feature vector h 11 h 12 The shapes are C5×1 and C6×1; Reshape() is a vector shape reshaping function used to reshape the process feature vector h with a shape of C6×1. 12 Transform it into a feature tensor H of shape C1×C7 1 C5, C6, and C7 are the preset fifth, sixth, and seventh feature dimensions, respectively, with C1 < C5 < C6 and C7 = C6 / C1. For example, C5 = 256, C6 = 832, and C7 = 64.

[0031] 3) Neurological Feature Encoding Module: The neurological feature encoding module is implemented based on an MLP model; the neurological feature encoding module is used to encode the neurological index vector X. 2 The corresponding feature tensor H is obtained by performing feature encoding processing. 2 Send to the Western medicine feature fusion module.

[0032] Here, the feature tensor H 2 The reasoning process is as follows: , , ; Among them, W 21 W 22 Here are the two weight matrix parameters for the neurological feature encoding module, b 21 b 22 These are the two offset vector parameters of the neurological feature encoding module; h 21 h 22 These are the two process feature vectors of the neurological feature encoding module; the weight matrix parameter W 21 The shape is C8×C2, and the offset vector parameter is b. 21 The shape is C8×1; the weight matrix parameter W 22 The shape is C9×C8, and the offset vector parameter is b. 22 The shape is C9×1; the process feature vector h 21 h 22 The shapes are C8×1 and C9×1; the vector shape reshaping function Reshape() is used to reshape the process feature vector h with a shape of C9×1. 22 Transform it into a feature tensor H of shape C1×C7 2 C8 and C9 are the preset eighth and ninth feature dimensions, C2 < C8 < C9, C9 = C1 × C7, for example C8 = 256, C9 = 832.

[0033] 4) Cognitive Psychological Feature Encoding Module: The cognitive psychological feature encoding module is implemented based on the MLP model; the cognitive psychological feature encoding module is used to encode the cognitive psychological index vector X. 3 The corresponding feature tensor H is obtained by performing feature encoding processing. 3 Send to the Western medicine feature fusion module; Here, the feature tensor H 3 The reasoning process is as follows: , , ; Among them, W31 W 32 b are the two weight matrix parameters of the cognitive psychological feature encoding module. 31 b 32 These are the two offset vector parameters for the cognitive psychological feature encoding module; h 31 h 32 The two process feature vectors of the cognitive psychological feature encoding module; weight matrix parameter W 31 The shape is C 10 ×C3, offset vector parameter b 31 The shape is C 10 ×1; Weight matrix parameter W 32 The shape is C 11 ×C 10 offset vector parameter b 32 The shape is C 11 ×1; Process feature vector h 31 h 32 The shape is C 10 ×1、C 11 ×1; The vector shape reshaping function Reshape() is used to reshape a vector into a shape of C. 11 ×1 process feature vector h 32 Transform it into a feature tensor H of shape C1×C7 3 C 10 C 11 For the preset tenth and eleventh feature dimensions, C3 < C 10 <C 11 C 11 =C1×C7, for example, C 10 =256、C 11 =832.

[0034] 5) Traditional Chinese Medicine Syndrome Characteristic Coding Module: The TCM syndrome feature encoding module is implemented based on an MLP model; the TCM syndrome feature encoding module is used to encode the TCM syndrome indicator vector X. 4 The corresponding feature tensor H is obtained by performing feature encoding processing. 4 Send to the TCM and Western medicine feature fusion module.

[0035] Here, the feature tensor H 4 The reasoning process is as follows: , , ; Among them, W 41 W 42 For the two weight matrix parameters of the TCM syndrome feature encoding module, b 41 b42 These are the two offset vector parameters of the TCM syndrome feature encoding module; h 41 h 42 These are the two process feature vectors of the TCM syndrome feature encoding module; the weight matrix parameter W 41 The shape is C 12 ×C4, offset vector parameter b 41 The shape is C 12 ×1; Weight matrix parameter W 42 The shape is C 13 ×C 12 offset vector parameter b 42 The shape is C 13 ×1; Process feature vector h 41 h 42 The shape is C 12 ×1、C 13 ×1; The vector shape reshaping function Reshape() is used to reshape a vector into a shape of C. 13 ×1 process feature vector h 42 Transform it into a feature tensor H of shape C1×C7 4 C 12 C 13 For the preset twelfth and thirteenth feature dimensions, C4 < C 12 <C 13 C 13 =C1×C7, for example, C 12 =256、C 13 =832.

[0036] 6) Western Medicine Feature Integration Module: The Western medicine feature fusion module is used for feature tensor H 2 Generate the corresponding query tensor Q and feature tensor H. 3 Generate the corresponding key-value tensors K and V; and perform attention feature extraction processing based on the query, key, and value tensors Q, K, and V to obtain the corresponding attention tensor A; and process the feature tensor H 2 The corresponding feature tensor H is obtained by performing a residual connection with the attention tensor A. 5 Send to the TCM and Western medicine feature fusion module.

[0037] Here, the feature tensor H 5 The reasoning process is as follows: , , , , ; Among them, W Q WK W V These are the three weight matrix parameters for the Western medicine feature fusion module; weight matrix parameter W Q W K W V The shapes are all C7×C qkv Preset feature dimension C qkv =C7; Query tensor Q, key tensor K, value tensor V, attention tensor A, and feature tensor H. 5 The shapes are all C1×C7; Softmax() is the Softmax function.

[0038] 7) Integration of Traditional Chinese and Western Medicine Features Module: The Traditional Chinese Medicine and Western Medicine Feature Fusion Module is used to process the feature tensor H 4 and feature tensor H 5 Feature splicing yields a shape of C1×C 14 Feature tensor H 6 C 14 C is the preset fourteenth feature dimension. 14 =2×C7; and by applying the characteristic tensor H 6 Performing a fully connected operation yields a gated weight tensor G of shape C1×1; and C7 gated weight tensors G are combined to form a gated weight tensor G of shape C1×C7. ’ ; and based on the gated weight tensor G ’ Feature tensor H 4 and feature tensor H 5 Feature fusion processing is performed to obtain the corresponding feature tensor H. 7 Send to the subjective and objective feature fusion module.

[0039] Here, the feature tensor H 7 The reasoning process is as follows: , ; Among them, W G b G For the weight matrix parameters and offset vector parameters of the traditional Chinese and Western medicine feature fusion module; weight matrix parameter W G The shape is C 14 ×1, Offset vector parameter b G The shape is C1×1; Sigmoid() is the Sigmoid activation function; the shape of the gate weight tensor G is C1×1; I is a preset all-1 tensor with a shape of C1×C7; ⊙ is the Hadamard product; the feature tensor H 7 The shape is C1×C7.

[0040] 8) Subjective and objective feature fusion module: The subjective and objective feature fusion module is used to process the feature tensor H 1 and feature tensor H 7 Feature fusion processing is performed to obtain the corresponding feature tensor H. 8 Send to the feature mapping module.

[0041] Here, the feature tensor H 8 The reasoning process is as follows: ; Among them, the feature tensor H 8 The shape is C1×C7.

[0042] 9) Feature mapping module: The feature mapping module is used to map the feature tensor H using a 1×1 convolution kernel. 8 Feature reduction processing yields a shape of C1×C 15 Feature tensor H 9 C 15 As the preset fifteenth feature dimension, C1 < C 15 <C7; and for the characteristic tensor H 9 Global max pooling along the metric dimension yields a 1×C value. 15 eigenvector H 10 ; and for the eigenvector H 10 Perform a fully connected operation to obtain the corresponding feature scalar H 11 Send to the index prediction module.

[0043] Here, the characteristic scalar H 11 The reasoning process is as follows: ; Among them, W 51 b 51 The weight matrix parameters and offset vector parameters are for the feature mapping module; the weight matrix parameter W 51 The shape is C 15 ×C 16 Offset vector parameter b 51 The shape is 1×C 16 C 16 C is the preset sixteenth feature dimension. 16 =1.

[0044] 10) Index Prediction Module: The index prediction module is used to predict the index based on a preset weight scalar W. 61 Offset scalar b 61 For the characteristic scalar H 11 Linear scaling is performed to obtain the corresponding prediction performance index and output it.

[0045] Here, the predictive performance index = W 61 ×H 11 +b 61 ; Weight scalar W 61 Offset scalar b 61 These are two learnable scalar parameters for the exponential prediction module.

[0046] Step 3: Construct a model dataset by conducting Chinese and Western medicine examinations and brain function expert assessments on the pre-recruited volunteer population, and denote it as the corresponding first dataset.

[0047] Here, the pre-recruited volunteer population covers a wide range of age groups; and during the recruitment of volunteers, their brain efficacy was pre-assessed, and the number of volunteers in the four categories of pre-assessed volunteers (volunteers with suspected severe decline in brain efficacy, volunteers with suspected moderate decline in brain efficacy, volunteers with suspected mild decline in brain efficacy, and volunteers with suspected good brain efficacy) was balanced so that the ratio of the four categories of pre-assessed volunteers was approximately 1:1:1:1.

[0048] The first dataset includes multiple first data records; each first data record includes a first training indicator set and a first label index; the first training indicator set is a multidimensional indicator set; the self-evaluation indicator set of the first training indicator set maintains data integrity and has no invalid indicators; if there are preset invalid indicators in the first training indicator set, the ratio of the total number of invalid indicators to the total number of all indicators satisfies the preset first masking rate, and all invalid indicators are randomly distributed in the neurological indicator set, the cognitive psychology indicator set, and the traditional Chinese medicine syndrome indicator set; the total number of invalid indicators is the total number of invalid indicators, and the total number of all indicators is the total number of indicators in the multidimensional indicator set; the first masking rate is a preset ratio parameter, and the maximum value of this ratio parameter cannot exceed 40%, for example, 30% or 40%.

[0049] The current step 3 specifically includes: Step 3-1: Identify each volunteer in the volunteer group as the current volunteer. Step 3-2, and the current volunteer conducts a self-evaluation based on the preset brain efficacy self-evaluation scale and the corresponding scale self-evaluation rules, and all the self-evaluation indicators obtained are combined into a corresponding self-evaluation indicator set. Here, the brain efficacy self-assessment scale of this invention is a custom assessment scale composed of multiple self-assessment items, namely: learning ability self-assessment item, memory self-assessment item, attention self-assessment item, energy and physical strength self-assessment item, emotional stability self-assessment item, sleep repair ability self-assessment item, willpower self-assessment item, control ability self-assessment item, executive ability self-assessment item, social vitality self-assessment item, life vitality self-assessment item, self-confidence self-assessment item, and self-awareness self-assessment item; the self-assessment items of this brain efficacy self-assessment scale correspond one-to-one with the self-assessment indicators in the self-assessment indicator set; The self-assessment rules for the Brain Efficacy Self-Rating Scale consist of multiple self-assessment rules, namely: self-assessment rules for learning ability, memory, attention, energy and physical strength, emotional stability, sleep recovery ability, willpower, control, executive function, social vitality, daily life vitality, self-confidence, and self-awareness. Each self-assessment rule corresponds one-to-one with the self-assessment items of the Brain Efficacy Self-Rating Scale. The self-assessment rules of this scale are used to set the scoring levels for the current self-assessment item and to explain the scoring rules corresponding to each level of score. Step 3-3 involves pre-designated medical experts from the first expert group performing resting-state structural magnetic resonance imaging (sMRI), diffusion tensor imaging (DTI), and functional magnetic resonance imaging (fMRI) on the current volunteers. Resonance imaging (fMRI) was used to obtain corresponding sMRI, DTI, and fMRI images. Based on the sMRI images, hippocampal brain regions were segmented and their volume measured using FreeSurfer, FSL, or SPM analysis tools, and the obtained hippocampal volume was used as the corresponding brain region volume index. Based on the sMRI images, the surface of the cerebral cortex was reconstructed using FreeSurfer or CAT12 analysis tools, and the thickness of the prefrontal cortex was measured on the reconstructed surface, with the obtained prefrontal cortex thickness used as the corresponding cortical thickness index. Based on the DTI images, the partial anisotropy score of the corpus callosum was calculated using FSL analysis tools, and the calculated score was used as the corresponding white matter integrity index. Finally, based on the fMRI images, the global connectivity strength of the brain's default mode network was analyzed using CONN or GRETNA analysis tools, and the analysis results were used as the corresponding functional connectivity density index. Here, the first expert group in this embodiment of the invention is a medical imaging expert group; Steps 3-4 involve medical experts from a pre-designated second expert group performing an electroencephalogram (EEG) on the current volunteer to obtain the corresponding first EEG image; and then using EEGLAB, FieldTrip, or Chronux analysis tools to perform alpha wave rhythm phase lock value analysis based on the first EEG image, and using the analysis results as the corresponding neural oscillation synchronicity index. Here, the second expert group in this embodiment of the invention is a neuroelectrophysiology and neuroscience expert group; Steps 3-5 involve a pre-selected third expert group of medical experts performing transcranial magnetic stimulation (TMS) on the activated brain regions of the current volunteer. Simultaneously, electroencephalography (EEG) is performed on the activated brain regions and one or more other brain regions of the current volunteer to obtain corresponding second and third EEG sets. Following Granger causality analysis, the causal flow intensity between the activated brain regions of the current volunteer and other brain regions is analyzed based on the second and third EEG sets. The mean of all obtained causal flow intensities is calculated, and the calculation results are used as the corresponding causal interaction indicators. Here, the second EEG map corresponds to the activated brain region; the third EEG map set consists of one or more third EEG maps, and the third EEG map corresponds one-to-one with other brain regions; The third expert group in this embodiment of the invention is a neuroelectrophysiology and neuroscience expert group; Steps 3-6, and the medical experts of the pre-designated fourth expert group will evaluate all cognitive psychological indicators of the current volunteers' cognitive psychological indicator set; Here, the fourth expert group in this embodiment of the invention is a group of experts in neuropsychology, behavioral neurology, psychiatry, and neurology; Steps 3-7, and the medical experts of the pre-designated fifth expert group will evaluate all the TCM syndrome indicators of the current volunteers' TCM syndrome indicator set; Here, the fifth expert group in this embodiment of the invention is an expert group in the field of traditional Chinese medicine; Steps 3-8, and a joint expert group composed of the first, second, third, fourth and fifth expert groups will assess and score the current brain efficacy status of the volunteers and use the obtained assessment scores as the corresponding first label index; Steps 3-9 are used to form a set of neurological indicators, which consists of brain region volume, white matter integrity, cortical thickness, functional connectivity density, neural oscillation synchronicity, and causal interaction indicators corresponding to the current volunteer; and a set of first training indicators is formed by the set of self-evaluation indicators, neurological indicators, cognitive psychology indicators, and traditional Chinese medicine syndrome indicators corresponding to the current volunteer. Steps 3-10: Based on the first mask rate and invalid indicators, the indicators of the neurological indicator set, cognitive psychology indicator set and traditional Chinese medicine syndrome indicator set of the current first training indicator set are randomly replaced once to obtain a derived first training indicator set. The random replacement process is repeated N times to obtain N derived first training indicator sets. Here, N is a preset positive integer; the random replacement process in this embodiment of the invention is actually a random masking mechanism; based on this random masking mechanism, on the one hand, the sample quantity expansion effect of data augmentation is achieved, and on the other hand, the diversity of test data is improved. Training the model based on such data helps to improve the model's prediction performance, prediction robustness and model generalization. Steps 3-11 are performed, and a corresponding first data record is formed by each first training indicator set and the corresponding first label index corresponding to the current volunteer; and a corresponding first data record set is formed by N+1 first data records corresponding to the current volunteer. Steps 3-12, and the corresponding first dataset is composed of all the first data records corresponding to all volunteers in the volunteer population.

[0050] Step 4: Train the brain efficacy prediction model based on the first dataset; Specifically, it includes: Step 41, dividing the first dataset into two sub-datasets based on a preset first segmentation ratio, denoted as the corresponding first training set and first evaluation set; Wherein, the first segmentation ratio is a pre-set ratio parameter, such as 8:2; both the first training set and the first evaluation set consist of multiple first data records; the ratio of the total number of records in the first training set and the first evaluation set satisfies the first segmentation ratio; Step 42: Perform a round of traversal on all first data records in the first training set; during this round of traversal, take the first data record currently traversed as the corresponding current training record; input the first training index set of the current training record as the current multidimensional index set into the brain efficacy prediction model for prediction, and take the prediction efficacy index output by this prediction as the corresponding first prediction index; and form a corresponding first prediction-label pair by the first prediction index and the first label index of the current training record; and at the end of this round of traversal, substitute all the first prediction-label pairs obtained in this round of traversal into the preset first model loss function to calculate the corresponding first loss value; The first model loss function is implemented based on either the L1 loss function or the L2 loss function. Step 43: Identify whether the first loss value meets the preset first loss value range; if it does, proceed to step 44; if it does not, modulate the model parameters of the brain efficacy prediction model in one round based on the preset first model optimizer in the direction of minimizing the first model loss function, and return to step 42 to continue training when the first round of modulation ends. Here, the first loss value range is a pre-set numerical range; the first model optimizer includes at least the Adam optimizer and the SGD optimizer; Step 44: Perform a round of traversal on all first data records in the first evaluation set; during this round of traversal, take the currently traversed first data record as the corresponding current evaluation record; input the first training index set of the current evaluation record as the current multidimensional index set into the brain efficacy prediction model for prediction, and take the prediction efficacy index output by this prediction as the corresponding second prediction index; and form a corresponding second prediction-label pair with the first label index of the current evaluation record; and at the end of this round of traversal, input all the second prediction-label pairs obtained in this round of traversal into the preset first model evaluation function to calculate the corresponding first evaluation value; The first model evaluation function is implemented based on the MAE function, MSE function, or RMSE function. Step 45: Identify whether the first evaluation value meets the preset first evaluation value range; if not, return to step 42 to continue training; if yes, confirm that the model training is complete.

[0051] Here, the first evaluation value range is a pre-set numerical range.

[0052] Step 5: After the model training is completed, the multidimensional index set of the test subjects input by the user is input into the brain efficacy prediction model for prediction, and the predicted efficacy index output by this prediction is fed back to the current user.

[0053] Here, in the multidimensional indicator set of the test subjects, the self-evaluation indicator set must maintain data integrity and have no invalid indicators; the neurological indicator set, cognitive psychology indicator set, and traditional Chinese medicine syndrome indicator set can have invalid indicators, as long as the ratio of the total number of invalid indicators to the total number of indicators does not exceed the first mask rate.

[0054] It should be noted that, for the multidimensional indicator set in this embodiment of the invention, indicators can also be added or deleted in each sub-indicator set (self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set, and traditional Chinese medicine syndrome indicator set) based on actual application needs. After each indicator addition or deletion operation, it is necessary to adaptively adjust the training data in the first dataset, adaptively adjust the input / output structure of each level module of the brain efficacy prediction model, and retrain the adjusted brain efficacy prediction model based on the adjusted first dataset. This will result in a new prediction model adapted to the new version of the multidimensional indicator set.

[0055] Figure 4 This is a module structure diagram of a processing device for predicting brain efficacy based on multidimensional indicators, provided in Embodiment 2 of the present invention. This device can be a terminal device or server implementing the aforementioned method embodiments, or it can be a device that enables the aforementioned terminal device or server to implement the aforementioned method embodiments. For example, the device can be a device or chip system of the aforementioned terminal device or server. Figure 4 As shown, the device includes: a multi-dimensional indicator system construction module 201, a model construction module 202, a dataset construction module 203, a model training module 204, and a model prediction module 205.

[0056] The multidimensional indicator system construction module 201 is used to set up the multidimensional indicator dataset for brain efficacy prediction to obtain the corresponding multidimensional indicator set; the multidimensional indicator set consists of four sub-datasets, namely the self-evaluation indicator set, the neurological indicator set, the cognitive psychology indicator set, and the traditional Chinese medicine syndrome indicator set.

[0057] The model building module 202 is used to build a deep learning model for brain efficacy prediction as the corresponding brain efficacy prediction model; the brain efficacy prediction model is used to predict brain efficacy based on the multidimensional index set of model input and output the corresponding predicted efficacy index.

[0058] The dataset construction module 203 is used to construct a model dataset by conducting Chinese and Western medicine examinations and brain efficacy expert evaluations on a pre-recruited volunteer population, denoted as the corresponding first dataset.

[0059] The model training module 204 is used to train the brain efficacy prediction model based on the first dataset.

[0060] The model prediction module 205 is used to input the multidimensional index set of the test subjects input by the user into the brain efficacy prediction model after the model training is completed, and to feed back the prediction efficacy index output by this prediction to the current user.

[0061] The present invention provides a processing device for predicting brain efficacy based on multidimensional indicators, which can execute the method steps in the above method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0062] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. For example, the multi-dimensional index system construction module can be a separate processing element, or it can be integrated into a chip in the above device. Alternatively, it can be stored as program code in the memory of the above device, and called and executed by a processing element of the device. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.

[0063] For example, these modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). As another example, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a System-on-a-Chip (SOC).

[0064] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the foregoing method embodiments are generated. The computer described above can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The aforementioned computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the aforementioned computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, Bluetooth, microwave, etc.) means. The aforementioned computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The aforementioned available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0065] Figure 5 This is a schematic diagram of an electronic device provided in Embodiment 3 of the present invention. This electronic device can be a terminal device or server implementing the methods of the aforementioned embodiments, or it can be a terminal device or server connected to the aforementioned terminal device or server implementing the methods of the aforementioned embodiments. Figure 5 As shown, the electronic device may include: a processor 301 (e.g., CPU), a memory 302, and a transceiver 303; the transceiver 303 is coupled to the processor 301, and the processor 301 controls the transmission and reception operations of the transceiver 303. The memory 302 may store various instructions for performing various processing functions and implementing the processing steps described in the foregoing embodiments. Preferably, the electronic device according to the embodiments of the present invention further includes: a power supply 304, a system bus 305, and a communication port 306. The system bus 305 is used to realize communication connections between components. The communication port 306 is used for communication between the electronic device and other peripherals.

[0066] exist Figure 5The system bus 305 mentioned can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, it is represented by only one thick line in the figure, but this does not indicate that there is only one bus or one type of bus. The communication interface is used to enable communication between the database access device and other devices (e.g., clients, read-write libraries, and read-only libraries). Memory may include Random Access Memory (RAM) and may also include Non-Volatile Memory, such as at least one disk storage device.

[0067] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), graphics processing units (GPUs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0068] It should be noted that the embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when run on a computer, cause the computer to perform the methods and processes provided in the above embodiments.

[0069] This invention provides a method, apparatus, electronic device, and computer-readable storage medium for predicting brain efficacy based on multidimensional indicators. As described above, this invention first customizes a multidimensional (self-evaluation, neurological, cognitive psychology, and traditional Chinese medicine syndrome) indicator dataset for brain efficacy prediction and constructs a brain efficacy prediction model capable of predicting brain efficacy indices based on the multidimensional indicator set. Then, a model dataset is constructed by conducting traditional Chinese and Western medicine examinations and brain efficacy expert assessments on a pre-recruited volunteer population, and the brain efficacy prediction model is trained based on this dataset. After model training, the brain efficacy prediction model is used to predict the brain efficacy index of any subject. This invention improves data feature richness and prediction accuracy through a multidimensional comprehensive prediction system integrating subjective and objective evaluation systems and traditional Chinese and Western medicine evaluation systems; it improves prediction performance and efficiency through the use of a brain efficacy prediction model; and it enhances the robustness and generalization of the model through a masking strategy set during model training.

[0070] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0071] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A processing method for predicting brain efficacy based on multidimensional indicators, characterized in that, The method includes: A multidimensional indicator set is obtained by setting up a multidimensional indicator dataset for brain efficacy prediction; the multidimensional indicator set consists of four sub-datasets, namely, self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set and traditional Chinese medicine syndrome indicator set. A deep learning model for predicting brain efficacy is constructed as the corresponding brain efficacy prediction model; the brain efficacy prediction model is used to predict brain efficacy based on the multidimensional index set input to the model and output the corresponding predicted efficacy index. The model dataset was constructed by conducting both traditional Chinese and Western medicine examinations and brain function expert assessments on a pre-recruited volunteer population, and is denoted as the corresponding first dataset. The brain efficacy prediction model is trained based on the first dataset; After the model training is completed, the multidimensional index set of the test subject input by the user is input into the brain efficacy prediction model for prediction, and the prediction efficacy index output by this prediction is fed back to the current user. The brain efficacy prediction model includes a preprocessing module, a self-evaluation feature encoding module, a neurological feature encoding module, a cognitive psychological feature encoding module, a traditional Chinese medicine syndrome feature encoding module, a Western medicine feature fusion module, a traditional Chinese and Western medicine feature fusion module, a subjective and objective feature fusion module, a feature mapping module, and an index prediction module. The input of the preprocessing module is connected to the model input, and the first, second, third, and fourth outputs are respectively connected to the inputs of the self-evaluation feature encoding module, the neurological feature encoding module, the cognitive psychological feature encoding module, and the traditional Chinese medicine syndrome feature encoding module. The outputs of the neurological feature encoding module and the cognitive psychological feature encoding module are connected to the first and second inputs of the Western medicine feature fusion module. The outputs of the traditional Chinese medicine syndrome feature encoding module and the Western medicine feature fusion module are connected to the first and second inputs of the traditional Chinese and Western medicine feature fusion module. The outputs of the self-evaluation feature encoding module and the traditional Chinese and Western medicine feature fusion module are connected to the first and second inputs of the subjective and objective feature fusion module. The output of the subjective and objective feature fusion module is connected to the input of the feature mapping module. The output of the feature mapping module is connected to the input of the index prediction module. The output of the index prediction module is connected to the model output.

2. The processing method for predicting brain efficacy based on multidimensional indicators according to claim 1, characterized in that, The self-evaluation indicator set consists of multiple categories of self-evaluation indicators, including at least: learning ability self-evaluation indicators, memory self-evaluation indicators, attention self-evaluation indicators, energy and physical strength self-evaluation indicators, emotional stability self-evaluation indicators, sleep recovery ability self-evaluation indicators, willpower self-evaluation indicators, control ability self-evaluation indicators, execution ability self-evaluation indicators, social vitality self-evaluation indicators, life vitality self-evaluation indicators, self-confidence self-evaluation indicators, and self-awareness self-evaluation indicators; each category of self-evaluation indicator is a self-evaluation score; all self-evaluation indicators have the same score range, which is a preset self-evaluation score threshold range; The set of neurological indicators comprises multiple categories of neurological indicators, including at least: brain region volume indicators, white matter integrity indicators, cortical thickness indicators, functional connectivity density indicators, neural oscillation synchronicity indicators, and causal interaction indicators. Specifically, the brain region volume indicator is the volume of the hippocampus; the white matter integrity indicator is the partial anisotropy score of the corpus callosum; the cortical thickness indicator is the thickness of the prefrontal cortex; the functional connectivity density indicator is the global connectivity strength of the default mode network; the neural oscillation synchronicity indicator is the alpha wave rhythm phase lock value on an electroencephalogram (EEG); and the causal interaction indicator is the average intensity of the causal flow from a pre-defined activated brain region to one or more other pre-defined brain regions. The cognitive psychological indicator set consists of multiple cognitive psychological indicators, including at least: MMSE assessment indicators, MoCA assessment indicators, AVLT assessment indicators, WMS assessment indicators, VFT assessment indicators, BNT assessment indicators, TMT-A assessment indicators, TMT-B assessment indicators, CDT assessment indicators, and Stroop. The assessment indicators include: ADL (Activities of Daily Living) assessment indicators, NPI (Natural Physical Scale) assessment indicators, HAMA (Hamiltonian Anxiety Scale) assessment indicators, and HAMD (Hamiltonian Depression Scale) assessment indicators. Specifically, the MMSE (Montreal Mental State Examination) assessment indicator is the score of the Mini-Mental State Examination; the MoCA (Montreal Cognitive Assessment) assessment indicator is the score of the Montreal Cognitive Assessment; the AVLT (Audio-Verbal Learning Test) assessment indicator is the score of the Auditory-Verbal Learning Test; the WMS (Wechsler Memory Scale) assessment indicator is the score of the Wechsler Memory Scale; the VFT (Verbal Fluency Test) assessment indicator is the score of the Verbal Fluency Test; the BNT (Boston Naming Test) assessment indicator is the score of the Boston Naming Test; the TMT-A and TB (Tracking Test A and B) assessment indicators are the scores of the Connecting Test A and B; the CDT (Clock Drawing Test) assessment indicator is the score of the Clock Drawing Test; the Stroop Test assessment indicator is the score of the Stroop Test; the ADL (Activities of Daily Living) assessment indicator is the score of the Daily Living Skills Scale; the NPI (Natural Physical Scale) assessment indicator is the score of the Neuropsychiatric Questionnaire; the HAMA (Hamiltonian Anxiety Scale) assessment indicator is the score of the Hamilton Anxiety Scale; and the HAMD (Hamiltonian Depression Scale) assessment indicator is the score of the Hamilton Depression Rating Scale. The TCM syndrome index set consists of multiple TCM syndrome indices, including at least: the syndrome index of gradual depletion of marrow and sea, the syndrome index of spleen and kidney deficiency, the syndrome index of qi and blood deficiency, the syndrome index of phlegm obstructing the orifices, the syndrome index of blood stasis obstructing the brain collaterals, the syndrome index of heart and liver fire excess, and the syndrome index of extreme deficiency due to toxicity. Each syndrome index is a binary index, specifically yes or no. The first dataset includes multiple first data records; each first data record includes a first training indicator set and a first label index; the first training indicator set is a multidimensional indicator set; the self-evaluation indicator set of the first training indicator set maintains data integrity and has no invalid indicators; if there are preset invalid indicators in the first training indicator set, the ratio of the total number of invalid indicators to the total number of all indicators satisfies a preset first masking rate, and all invalid indicators are randomly distributed in the neurological indicator set, the cognitive psychology indicator set, and the traditional Chinese medicine syndrome indicator set; the total number of invalid indicators is the total number of invalid indicators, and the total number of all indicators is the total number of indicators in the multidimensional indicator set.

3. The processing method for predicting brain efficacy based on multidimensional indicators according to claim 1, characterized in that, The model input terminal of the brain efficacy prediction model is used to receive the multidimensional index set, and the model output terminal is used to output the corresponding prediction efficacy index. The preprocessing module is used to take the self-evaluation index set, the neurological index set, the cognitive psychological index set, and the traditional Chinese medicine syndrome index set of the multidimensional index set as the corresponding self-evaluation index vector X. 1 Neurological index vector X 2 Cognitive psychological indicator vector X 3 and the TCM syndrome index vector X 4 The self-evaluation index vector X is sent to the corresponding self-evaluation feature encoding module, the neurological feature encoding module, the cognitive psychological feature encoding module, and the traditional Chinese medicine syndrome feature encoding module. 1 The shape is C1×1, and the neurological index vector X 2 The shape is C2×1, and the cognitive psychological index vector X 3 The shape is C3×1, and the TCM syndrome index vector X 4 The shape is C4×1, where C1, C2, C3, and C4 are the preset first, second, third, and fourth feature dimensions; The self-evaluation feature encoding module is implemented based on an MLP model; the self-evaluation feature encoding module is used to encode the self-evaluation index vector X. 1 The corresponding feature tensor H is obtained by performing feature encoding processing. 1 Send to the subjective and objective feature fusion module; The feature tensor H 1 The reasoning process is as follows: ， ， ， ； Among them, W 11 W 12 Let b be the two weight matrix parameters of the self-evaluation feature encoding module. 11 b 12 h are the two offset vector parameters of the self-evaluation feature encoding module. 11 h 12 The two process feature vectors of the self-evaluation feature encoding module; GELU() is the GELU activation function; the weight matrix parameter W 11 The shape is C5×C1, and the offset vector parameter is b. 11 The shape is C5×1; the weight matrix parameter W 12 The shape is C6×C5, and the offset vector parameter is b. 12 The shape is C6×1; the process feature vector h 11 h 12 The shapes are C5×1 and C6×1; Reshape() is a vector shape reshaping function used to reshape the process feature vector h with a shape of C6×1. 12 This is converted into a feature tensor H of shape C1×C7. 1 C5, C6, and C7 are the preset fifth, sixth, and seventh feature dimensions, respectively, where C1 < C5 < C6 and C7 = C6 / C1. The neurological feature encoding module is implemented based on an MLP model; the neurological feature encoding module is used to process the neurological index vector X. 2 The corresponding feature tensor H is obtained by performing feature encoding processing. 2 Send to the Western medicine feature fusion module; The feature tensor H 2 The reasoning process is as follows: ， ， ； Among them, W 21 W 22 Let b be the two weight matrix parameters of the neurological feature encoding module. 21 b 22 h are the two offset vector parameters of the neurological feature encoding module. 21 h 22 The two process feature vectors of the neurological feature encoding module; weight matrix parameter W 21 The shape is C8×C2, and the offset vector parameter is b. 21 The shape is C8×1; the weight matrix parameter W 22 The shape is C9×C8, and the offset vector parameter is b. 22 The shape is C9×1; the process feature vector h 21 h 22 The shapes are C8×1 and C9×1; the vector shape reshaping function Reshape() is used to reshape the process feature vector h with shape C9×1. 22 This is converted into a feature tensor H of shape C1×C7. 2 C8 and C9 are the preset eighth and ninth feature dimensions, respectively, where C2 < C8 < C9 and C9 = C1 × C7. The cognitive psychological feature encoding module is implemented based on an MLP model; the cognitive psychological feature encoding module is used to process the cognitive psychological index vector X. 3 The corresponding feature tensor H is obtained by performing feature encoding processing. 3 Send to the Western medicine feature fusion module; The feature tensor H 3 The reasoning process is as follows: ， ， ； Among them, W 31 W 32 Let b be the two weight matrix parameters of the cognitive psychological feature encoding module. 31 b 32 h are the two offset vector parameters of the cognitive psychological feature encoding module. 31 h 32 The two process feature vectors of the cognitive psychological feature encoding module; weight matrix parameter W 31 The shape is C 10 ×C3, offset vector parameter b 31 The shape is C 10 ×1; Weight matrix parameter W 32 The shape is C 11 ×C 10 offset vector parameter b 32 The shape is C 11 ×1; Process feature vector h 31 h 32 The shape is C 10 ×1、C 11 ×1; The vector shape reshaping function Reshape() is used to reshape the vector into a shape of C. 11 ×1 process feature vector h 32 This is converted into a feature tensor H of shape C1×C7. 3 C 10 C 11 For the preset tenth and eleventh feature dimensions, C3 < C 10 <C 11 C 11 =C1×C7; The TCM syndrome feature encoding module is implemented based on an MLP model; the TCM syndrome feature encoding module is used to encode the TCM syndrome index vector X. 4 The corresponding feature tensor H is obtained by performing feature encoding processing. 4 Send to the Chinese and Western medicine feature fusion module; The feature tensor H 4 The reasoning process is as follows: ， ， ； Among them, W 41 W 42 b are the two weight matrix parameters of the TCM syndrome feature encoding module. 41 b 42 h are the two offset vector parameters of the TCM syndrome feature encoding module. 41 h 42 The two process feature vectors of the TCM syndrome feature encoding module; weight matrix parameter W 41 The shape is C 12 ×C4, offset vector parameter b 41 The shape is C 12 ×1; Weight matrix parameter W 42 The shape is C 13 ×C 12 offset vector parameter b 42 The shape is C 13 ×1; Process feature vector h 41 h 42 The shape is C 12 ×1、C 13 ×1; The vector shape reshaping function Reshape() is used to reshape the vector into a shape of C. 13 ×1 process feature vector h 42 This is converted into a feature tensor H of shape C1×C7. 4 C 12 C 13 For the preset twelfth and thirteenth feature dimensions, C4 < C 12 <C 13 C 13 =C1×C7; The Western medicine feature fusion module is used to perform fusion based on the feature tensor H. 2 Generate the corresponding query tensor Q, and based on the feature tensor H 3 Generate corresponding key-value tensors K and V; and perform attention feature extraction processing on the query, key, and value tensors Q, K, and V to obtain the corresponding attention tensor A; and process the feature tensor H 2 The corresponding feature tensor H is obtained by performing a residual connection with the attention tensor A. 5 Send to the Chinese and Western medicine feature fusion module; The feature tensor H 5 The reasoning process is as follows: ， ， ， ， ； Among them, W Q W K W V The three weight matrix parameters are: W, ... Q W K W V The shapes are all C7×C qkv Preset feature dimension C qkv =C7; the query tensor Q, the key tensor K, the value tensor V, the attention tensor A, and the feature tensor H 5 The shapes are all C1×C7; Softmax() is the Softmax function; The traditional Chinese and Western medicine feature fusion module is used to process the feature tensor H. 4 and the characteristic tensor H 5 Feature splicing yields a shape of C1×C 14 Feature tensor H 6 C 14 C is the preset fourteenth feature dimension. 14 =2×C7; and by applying the feature tensor H 6 A fully connected operation is performed to obtain a gated weight tensor G of shape C1×1; and C7 of these gated weight tensors G are combined to form a gated weight tensor G of shape C1×C7. ’ ; and based on the gated weight tensor G ’ The feature tensor H 4 and the characteristic tensor H 5 Feature fusion processing is performed to obtain the corresponding feature tensor H. 7 Send to the subjective and objective feature fusion module; The feature tensor H 7 The reasoning process is as follows: ， ； Among them, W G b G The weight matrix parameters and offset vector parameters of the traditional Chinese and Western medicine feature fusion module are: weight matrix parameter W. G The shape is C 14 ×1, Offset vector parameter b G The shape of the gate weight tensor G is C1×1; Sigmoid() is the Sigmoid activation function; the shape of the gate weight tensor G is C1×1; I is a preset all-1 tensor with a shape of C1×C7; ⊙ is the Hadamard product; the feature tensor H 7 The shape is C1×C7; The subjective and objective feature fusion module is used to process the feature tensor H. 1 and the characteristic tensor H 7 Feature fusion processing is performed to obtain the corresponding feature tensor H. 8 Send to the feature mapping module; The feature tensor H 8 The reasoning process is as follows: ； The feature tensor H 8 The shape is C1×C7; The feature mapping module is used to map the feature tensor H using a 1×1 convolution kernel. 8 Feature reduction processing yields a shape of C1×C 15 Feature tensor H 9 C 15 As the preset fifteenth feature dimension, C1 < C 15 <C7; and for the characteristic tensor H 9 Global max pooling along the metric dimension yields a 1×C value. 15 eigenvector H 10 ; and for the feature vector H 10 Perform a fully connected operation to obtain the corresponding feature scalar H 11 Send to the index prediction module; The feature scalar H 11 The reasoning process is as follows: ； Among them, W 51 b 51 The weight matrix parameters and offset vector parameters of the feature mapping module; weight matrix parameter W 51 The shape is C 15 ×C 16 Offset vector parameter b 51 The shape is 1×C 16 C 16 C is the preset sixteenth feature dimension. 16 =1; The index prediction module is used to predict the index based on a preset weight scalar W. 61 Offset scalar b 61 For the feature scalar H 11 Perform linear scaling to obtain the corresponding predicted performance index and output it; Predicted performance index = W 61 ×H 11 +b 61 .

4. The processing method for predicting brain efficacy based on multidimensional indicators according to claim 2, characterized in that, The model dataset, denoted as the first dataset, is constructed by conducting traditional Chinese and Western medicine examinations and brain function expert assessments on a pre-recruited volunteer population. Specifically, it includes: Each volunteer in the volunteer group is taken as the corresponding current volunteer; The self-assessment indicators obtained by the current volunteer based on the preset brain efficacy self-assessment scale and corresponding scale self-assessment rules constitute a corresponding self-assessment indicator set; the brain efficacy self-assessment scale includes self-assessment items for learning ability, memory, attention, energy and physical strength, emotional stability, sleep recovery ability, willpower, control, executive function, social vitality, life vitality, self-confidence, and self-awareness; the self-assessment items of the brain efficacy self-assessment scale correspond one-to-one with the self-assessment indicators in the self-assessment indicator set; The self-assessment rules of the scale include self-assessment rules for learning ability, memory, attention, energy and physical strength, emotional stability, sleep recovery ability, willpower, control, executive function, social vitality, life vitality, self-confidence, and self-awareness. Each self-assessment rule corresponds one-to-one with the self-assessment items of the brain efficacy self-assessment scale. Each self-assessment rule is used to grade the score of the current self-assessment item and provides an explanation of the scoring rules corresponding to each grade. The first expert group, comprised of medical experts, conducted resting-state structural magnetic resonance imaging (SMRI), diffusion tensor imaging (DTI), and functional magnetic resonance imaging (fMRI) examinations on the current volunteers to obtain corresponding sMRI, DTI, and fMRI images. Based on the sMRI images, hippocampal brain regions were segmented and their volume measured using FreeSurfer, FSL, or SPM analysis tools, and the obtained hippocampal volume was used as the corresponding brain region volume index. Based on the sMRI images, the cerebral cortex surface was reconstructed using FreeSurfer or CAT12 analysis tools, and the prefrontal cortex thickness was measured on the reconstructed cortical surface, with the obtained prefrontal cortex thickness used as the corresponding cortical thickness index. Based on the DTI images, the partial anisotropy score of the corpus callosum was calculated using FSL analysis tools, and the calculated score was used as the corresponding white matter integrity index. Based on the fMRI images, the global connectivity strength of the brain's default mode network was analyzed using CONN or GRETNA analysis tools, and the analysis results were used as the corresponding functional connectivity density index. The first expert group was a medical imaging expert group. The pre-designated second expert group of medical experts will conduct an electroencephalogram (EEG) examination on the current volunteer to obtain the corresponding first EEG image; and based on the first EEG image, alpha wave rhythm phase lock value analysis will be performed using EEGLAB, FieldTrip, or Chronux analysis tools, and the analysis results will be used as the corresponding neural oscillation synchronicity index; the second expert group is a neurophysiology and neuroscience expert group; A pre-designated third expert group of medical experts performed transcranial magnetic stimulation (TMS) on the activated brain regions of the current volunteer. Simultaneously, electroencephalography (EEG) was performed on the activated brain regions and one or more other brain regions to obtain corresponding second and third EEG maps. Following Granger causality analysis, the causal flow intensity between the activated brain regions and each of the other brain regions was analyzed based on the second and third EEG maps. The mean of all obtained causal flow intensities was calculated, and the result was used as the corresponding causal interaction index. The second EEG map corresponds to the activated brain region; the third EEG map set consists of one or more third EEG maps, each corresponding one-to-one with the other brain regions; the third expert group is a neurophysiology and neuroscience expert group. The cognitive psychological indicators of the current volunteer will be evaluated by medical experts from a pre-designated fourth expert group; the fourth expert group consists of experts in neuropsychology, behavioral neurology, psychiatry, and neurology. The medical experts of the pre-designated fifth expert group will evaluate all the TCM syndrome indicators of the current volunteer's TCM syndrome indicator set; the fifth expert group is an expert group in the field of TCM. A joint expert group composed of the first, second, third, fourth and fifth expert groups will assess and score the current volunteer's brain efficacy status and use the obtained assessment score as the corresponding first label index; A set of neurological indicators is formed by the brain region volume index, white matter integrity index, cortical thickness index, functional connectivity density index, neural oscillation synchronicity index, and causal interaction index corresponding to the current volunteer; and a set of training indicators is formed by the self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set, and traditional Chinese medicine syndrome indicator set corresponding to the current volunteer. Based on the first mask rate and the invalid index, the indicators of the neurological indicator set, the cognitive psychological indicator set, and the traditional Chinese medicine syndrome indicator set of the current first training indicator set are randomly replaced once to obtain a derived first training indicator set, and the random replacement process is repeated N times to obtain N derived first training indicator sets; N is a preset positive integer. The first data record is composed of each set of the first training indicators and the corresponding first label index corresponding to the current volunteer; and the first data record set is composed of N+1 first data records corresponding to the current volunteer. The first dataset is composed of all the first data records corresponding to all volunteers in the volunteer population.

5. The processing method for predicting brain efficacy based on multidimensional indicators according to claim 2, characterized in that, The training of the brain efficacy prediction model based on the first dataset specifically includes: Step 51: Based on a preset first segmentation ratio, the first dataset is divided into two sub-datasets, denoted as the first training set and the first evaluation set. Wherein, both the first training set and the first evaluation set are composed of multiple first data records; the ratio of the total number of records in the first training set and the first evaluation set satisfies the first segmentation ratio; Step 52: Perform a round of traversal on all the first data records in the first training set; during this round of traversal, take the currently traversed first data record as the corresponding current training record; and input the first training index set of the current training record as the current multidimensional index set into the brain efficacy prediction model for prediction, and take the prediction efficacy index output by this prediction as the corresponding first prediction index; and form a corresponding first prediction-label pair with the first prediction index and the first label index of the current training record; and at the end of this round of traversal, substitute all the first prediction-label pairs obtained in this round of traversal into the preset first model loss function to calculate the corresponding first loss value; The first model loss function is implemented based on the L1 loss function or the L2 loss function; Step 53: Identify whether the first loss value meets the preset first loss value range; if it does, proceed to step 54; if it does not, modulate the model parameters of the brain efficacy prediction model in one round based on the preset first model optimizer in the direction of minimizing the first model loss function, and return to step 52 to continue training when the first round of modulation ends. The first model optimizer includes at least the Adam optimizer and the SGD optimizer; Step 54: Perform a traversal of all the first data records in the first evaluation set; during this traversal, the currently traversed first data record is taken as the corresponding current evaluation record; the first training index set of the current evaluation record is taken as the current multidimensional index set and input into the brain efficacy prediction model for prediction; the prediction efficacy index output by this prediction is taken as the corresponding second prediction index; the second prediction index and the first label index of the current evaluation record form a corresponding second prediction-label pair; at the end of this traversal, all the second prediction-label pairs obtained in this traversal are input into a preset first model evaluation function to calculate the corresponding first evaluation value; The first model evaluation function is implemented based on the MAE function, MSE function, or RMSE function. Step 55: Identify whether the first evaluation value meets the preset first evaluation value range; if not, return to step 52 to continue training; if yes, confirm that the model training is over.

6. An apparatus for performing the processing method for predicting brain efficacy based on multidimensional indicators as described in any one of claims 1-5, characterized in that, The device includes: a multi-dimensional indicator system construction module, a model construction module, a dataset construction module, a model training module, and a model prediction module; The multidimensional indicator system construction module is used to set up the multidimensional indicator dataset for brain efficacy prediction to obtain the corresponding multidimensional indicator set; the multidimensional indicator set consists of four sub-datasets, namely, self-evaluation indicator set, neurological indicator set, cognitive psychology indicator set and traditional Chinese medicine syndrome indicator set. The model building module is used to build a deep learning model for brain efficacy prediction as the corresponding brain efficacy prediction model; the brain efficacy prediction model is used to predict brain efficacy based on the multidimensional index set input to the model and output the corresponding predicted efficacy index. The dataset construction module is used to construct a model dataset by conducting Chinese and Western medicine examinations and brain efficacy expert evaluations on a pre-recruited volunteer population, denoted as the corresponding first dataset. The model training module is used to train the brain efficacy prediction model based on the first dataset; The model prediction module is used to input the multidimensional index set of the test subject input by the user into the brain efficacy prediction model after the model training is completed, and to feed back the predicted efficacy index output by the current prediction to the current user.

7. An electronic device, characterized in that, include: Memory, processor, and transceiver; The processor is configured to be coupled to the memory, read and execute instructions in the memory to implement the method according to any one of claims 1-5; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1-5.

Images