An intelligent psychiatric treatment effect evaluation system

Abstract

The present invention discloses an intelligent psychiatric treatment effect evaluation system, including a data acquisition module, a data preprocessing module, a psychiatric treatment effect evaluation model building module and a psychiatric treatment effect evaluation module. The present invention belongs to the field of effect evaluation technology, and specifically refers to an intelligent psychiatric treatment effect evaluation system. This scheme provides a buffer for uncertainty by defining an estimation range to cover the real improvement evaluation level; introduces an adaptive Lagrange multiplier to allow the system to more strongly punish predictions with large errors, and punishes the prediction intervals that fail to capture the real treatment effect, thereby reducing the wrong prediction of the evaluation results; combines random factors with time parameters to construct a diverse system parameter adjustment initialization strategy, increase global randomness, and make the convergence of the system adjustment parameters stronger, thereby generating more robust prediction results and improving the system's prediction efficiency for the evaluation results.

Description

Technical Field

[0001] The present invention relates to the technical field of effect evaluation, and in particular to an intelligent psychiatric treatment effect evaluation system. Background Art

[0002] The psychiatric treatment effect evaluation system is a comprehensive system used to evaluate the treatment effect of patients during psychiatric treatment. Based on the patient's treatment process data, behavior and psychological state changes, the system uses data analysis and machine learning technologies to provide quantitative treatment effect evaluation to help doctors judge the patient's degree of improvement, the effectiveness of treatment and future treatment plans. However, the general psychiatric treatment effect evaluation system has the problem that the original evaluation data has samples that are difficult to capture, which leads to large system prediction errors and large uncertainty fluctuations, resulting in poor adaptability of the system to the evaluation data, and thus a large number of false predictions; the general psychiatric treatment effect evaluation system has insufficient parameter convergence, which leads to slow system training speed and weak generalization ability, which leads to poor efficiency of system evaluation prediction. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent psychiatric treatment effect evaluation system. In view of the fact that the original evaluation data of the general psychiatric treatment effect evaluation system has samples that are difficult to capture, which leads to large system prediction errors, large uncertainty fluctuations, and poor adaptability of the system to the evaluation data, and thus a large number of erroneous predictions, this solution provides a buffer for uncertainty by defining an estimation range to cover the actual improvement evaluation level; introducing capture rate, smooth capture rate and adaptive Lagrange multiplier, allowing the system to more strongly punish predictions with large errors, and punish the prediction interval that fails to capture the real treatment effect, thereby reducing erroneous predictions of the evaluation results and providing more accurate evaluation results; in view of the fact that the general psychiatric treatment effect evaluation system has insufficient parameter convergence, which leads to slow system training speed and weak generalization ability, and thus leads to poor efficiency of system evaluation prediction, this solution combines random factors with time parameters to construct a diverse system parameter initialization strategy, so that the initialized individual position is adjusted as the number of iterations changes, increases global randomness, and makes the convergence of the system adjustment parameters stronger, thereby generating more robust prediction results and improving the system's prediction efficiency for evaluation results.

[0004] The technical solution adopted by the present invention is as follows: an intelligent psychiatric treatment effect evaluation system provided by the present invention includes a data acquisition module, a data preprocessing module, a psychiatric treatment effect evaluation model establishment module and a psychiatric treatment effect evaluation module;

[0005] The data collection module collects historical psychiatric treatment process data;

[0006] The data preprocessing module performs data cleaning, data conversion and data set division on the collected data;

[0007] The psychiatric treatment effect evaluation model establishment module provides a buffer for uncertainty by defining an estimation range to cover the actual improvement evaluation level; introduces capture rate, smooth capture rate and adaptive Lagrange multiplier to allow the system to more strongly penalize predictions with large errors and penalize prediction intervals that fail to capture the actual treatment effect, and based on the system parameter adjustment initialization strategy, realizes the establishment of a psychiatric treatment effect evaluation model;

[0008] The psychiatric treatment effect evaluation module realizes psychiatric treatment effect evaluation based on the established psychiatric treatment effect evaluation model.

[0009] Furthermore, in the data collection module, the historical psychiatric treatment process data includes physiological data, psychological data, drug response data, social activity data, environmental information, time data and improvement assessment level; the improvement assessment level is used as a data label.

[0010] Furthermore, in the data preprocessing module, the data cleaning is to process missing values ​​and duplicate values; the data conversion is to convert the data into a vector form and perform standardization to obtain a time series data set; the data set partitioning is to divide the time series data set into a test set and a training set.

[0011] Furthermore, the psychiatric treatment effect evaluation model establishment module specifically includes the following contents:

[0012] Model architecture design unit; the established psychiatric treatment effect evaluation model includes: multi-layer one-dimensional convolutional layer, extracting time series features from pre-processed historical psychiatric treatment process data and adapting to changes at different time scales; pooling layer, reducing dimensions and extracting main features to avoid overfitting; fully connected layer, fusing multi-scale features and generating predicted values ​​of improvement evaluation level; output layer, outputting predicted improvement evaluation level;

[0013] Multi-layer one-dimensional convolutional layer design unit; using different convolution kernels, each layer of convolution kernel has a different window size, the convolution operation formula is expressed as follows: ; The feature map generated by each convolutional layer describes the local pattern of the historical psychiatric treatment process data. The feature maps from different convolutional layers are concatenated to form a multi-scale feature matrix, which represents the combination of short-term fluctuations and long-term trends; where F t is the output of the convolution operation; X t+m-1 is a local segment of the input sequence data starting from time t; W m is the weight of the convolution kernel; M is the window size of the convolution kernel, and m is the convolution kernel window size index;

[0014] Loss function design unit; specifically includes:

[0015] Define an estimation range. Generate an estimation range for improving the assessment level. Define the upper bound of the estimation range. and lower bound ; and ensure the maximum number of real improvement assessment levels It falls within the estimated range and is expressed as: ; Introduce capture indicator k i , used to indicate whether each estimated range captures the true improvement assessment level, expressed as: ; Calculate the estimated number of successfully captured ranges c, expressed as: ; Where Pr(·) is the probability; α is the uncertainty parameter; n is the number of samples, and i is the sample index;

[0016] Define the interval estimation width; calculate the proportion PP of the estimated range that successfully captures the true treatment effect, expressed as: ; and calculate the average estimated width PW of the estimated range that captures the observed values, expressed as: ;

[0017] Construct the initial loss function; introduce adaptive Lagrange multipliers , for each sample, when its prediction error is larger, the adaptive Lagrange multiplier is larger, and the penalty for wrong prediction is increased; it is expressed as: ; ;in, is the initial multiplier; γ is the adaptive growth factor;

[0018] Define the smooth snap vector; the smooth snap vector is expressed as: ; ; Get the smooth capture rate PPs, expressed as: ; and the softened average estimated width PWs, expressed as: ;in, is the sigmoid function; s i is the scaling factor;

[0019] Define the error calibration term; used to penalize the estimated range that fails to capture the true treatment effect, expressed as: ; Where C is the error calibration term;

[0020] Construct the final loss function; obtain the final loss function LQ applied to the psychiatric treatment effect evaluation model, expressed as: ;

[0021] Model judgment unit; when the loss of the psychiatric treatment effect evaluation model for the training set converges, the training of the psychiatric treatment effect evaluation model is completed; a prediction threshold is set in advance, and when the prediction accuracy of the trained psychiatric treatment effect evaluation model for the test set is higher than the prediction threshold, the psychiatric treatment effect evaluation model is established; if it is not higher than the prediction threshold or the maximum number of training times is reached, it is transferred to the parameter optimization unit for parameter optimization and the data set is re-divided to train the psychiatric treatment effect evaluation model;

[0022] Parameter optimization unit; construct parameter optimization space based on uncertainty parameters, initial multipliers, adaptive growth factors and initial weights and biases of psychiatric treatment effect evaluation model; initialize and optimize individual positions in the population, use the prediction accuracy of the psychiatric treatment effect evaluation model for the test set based on the individual position as the individual fitness value, and use the particle swarm search algorithm when updating the position; when there is an individual fitness value higher than the prediction threshold, the individual position is the optimized parameter setting, and the established psychiatric treatment effect evaluation model is obtained; otherwise, re-initialize the individual position in the population for parameter optimization; the individual position in the initialized optimization population is expressed as:

[0023] ; Among them, X i3j3 is the initialization position of individual i3 in the j3th dimension, t1 is the number of iterations, and t=0 at initialization; X minj3 and X maxj3 are the minimum and maximum boundaries of the j3rd dimension respectively; r i3j3 is a uniformly distributed random number ranging from 0 to 1; T is the maximum number of iterations.

[0024] Furthermore, the psychiatric treatment effect evaluation module is based on the established psychiatric treatment effect evaluation model, and collects physiological data, psychological data, drug response data, social activity data, environmental information and time data in real time; after preprocessing, it is input into the psychiatric treatment effect evaluation model, and the improvement evaluation level output by the model is used as the psychiatric treatment effect evaluation result.

[0025] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0026] (1) In view of the problem that the original evaluation data of general psychiatric treatment effect evaluation systems have samples that are difficult to capture, which leads to large system prediction errors. The large uncertainty fluctuations lead to poor adaptability of the system to the evaluation data, and thus there are a large number of erroneous predictions. This scheme provides a buffer for uncertainty by defining an estimation range to cover the actual improvement evaluation level; introducing capture rate, smooth capture rate and adaptive Lagrange multiplier, allowing the system to more strongly penalize predictions with large errors and penalize prediction intervals that fail to capture the true treatment effect, thereby reducing erroneous predictions of the evaluation results and providing more accurate evaluation results.

[0027] (2) In order to address the problem that general psychiatric treatment effect evaluation systems have insufficient convergence of parameter adjustments, which leads to slow system training speed and weak generalization ability, and in turn leads to poor efficiency of system evaluation and prediction, this scheme combines random factors with time parameters to construct a diverse system parameter initialization strategy, so that the initialized individual positions are adjusted as the number of iterations changes, increasing global randomness and making the system adjustment parameters more convergent, thereby generating more robust prediction results and improving the system's prediction efficiency for evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the process of an intelligent psychiatric treatment effect evaluation system provided by the present invention;

[0029] Figure 2 Schematic diagram of the process of establishing modules for the psychiatric treatment effectiveness evaluation model;

[0030] Figure 3 Schematic diagram of the process of designing a unit for the loss function.

[0031] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0034] Example 1, see Figure 1 , the present invention provides an intelligent psychiatric treatment effect evaluation system, including a data acquisition module, a data preprocessing module, a psychiatric treatment effect evaluation model establishment module and a psychiatric treatment effect evaluation module;

[0035] The data acquisition module collects historical psychiatric treatment process data; and sends the data to the data preprocessing module;

[0036] The data preprocessing module receives the data sent by the data acquisition module; performs data cleaning, data conversion and data set division on the collected data; and sends the data to the psychiatric treatment effect evaluation model establishment module;

[0037] The psychiatric treatment effect evaluation model establishment module receives data sent by the data preprocessing module; defines an estimation range to cover the actual improvement evaluation level, providing a buffer for uncertainty; introduces capture rate, smooth capture rate and adaptive Lagrange multiplier, allowing the system to more strongly penalize predictions with large errors, and penalizes the prediction interval that fails to capture the actual treatment effect, and based on the system parameter adjustment initialization strategy, realizes the establishment of the psychiatric treatment effect evaluation model; and sends the data to the psychiatric treatment effect evaluation module;

[0038] The psychiatric treatment effect evaluation module receives data sent by the psychiatric treatment effect evaluation model establishment module; and implements psychiatric treatment effect evaluation based on the established psychiatric treatment effect evaluation model.

[0039] Example 2, see Figure 1 This embodiment is based on the above embodiment. In the data collection module, the historical psychiatric treatment process data includes physiological data, psychological data, drug response data, social activity data, environmental information, time data and improvement assessment level; the improvement assessment level includes no obvious improvement, slight improvement, moderate improvement, significant improvement and complete improvement; the improvement assessment level is used as a data label.

[0040] Example 3, see Figure 1This embodiment is based on the above embodiment. In the data preprocessing module, data cleaning is to process missing values ​​and duplicate values; data conversion is to convert data into vector form and perform standardization to obtain a time series data set; data set division is to divide the time series data set into a test set and a training set.

[0041] Example 4, see Figure 1 and Figure 2 Based on the above embodiment, this embodiment includes the following contents:

[0042] Model architecture design unit; the established psychiatric treatment effect evaluation model includes: multi-layer one-dimensional convolutional layer, extracting time series features from pre-processed historical psychiatric treatment process data and adapting to changes at different time scales; pooling layer, reducing dimensions and extracting main features to avoid overfitting; fully connected layer, fusing multi-scale features and generating predicted values ​​of improvement evaluation level; output layer, outputting predicted improvement evaluation level;

[0043] Multi-layer one-dimensional convolutional layer design unit; using different convolution kernels, each layer of convolution kernel has a different window size, the convolution operation formula is expressed as follows: ; The feature map generated by each convolutional layer describes the local pattern of the historical psychiatric treatment process data. The feature maps from different convolutional layers are concatenated to form a multi-scale feature matrix, which represents the combination of short-term fluctuations and long-term trends; where F t is the output of the convolution operation; X t+m-1 is a local segment of the input sequence data starting from time t; W m is the weight of the convolution kernel; M is the window size of the convolution kernel, and m is the convolution kernel window size index;

[0044] Loss function design unit;

[0045] Model judgment unit; when the loss of the psychiatric treatment effect evaluation model for the training set converges, the training of the psychiatric treatment effect evaluation model is completed; a prediction threshold is set in advance, and when the prediction accuracy of the trained psychiatric treatment effect evaluation model for the test set is higher than the prediction threshold, the psychiatric treatment effect evaluation model is established; if it is not higher than the prediction threshold or the maximum number of training times is reached, it is transferred to the parameter optimization unit for parameter optimization and the data set is re-divided to train the psychiatric treatment effect evaluation model;

[0046] Parameter optimization unit; construct parameter optimization space based on uncertainty parameters, initial multipliers, adaptive growth factors and initial weights and biases of psychiatric treatment effect evaluation model; initialize and optimize individual positions in the population, use the prediction accuracy of the psychiatric treatment effect evaluation model for the test set based on the individual position as the individual fitness value, and use the particle swarm search algorithm when updating the position; when there is an individual fitness value higher than the prediction threshold, the individual position is the optimized parameter setting, and the established psychiatric treatment effect evaluation model is obtained; otherwise, re-initialize the individual position in the population for parameter optimization; the individual position in the initialized optimization population is expressed as:

[0047] ; Among them, X i3j3 is the initialization position of individual i3 in the j3th dimension, t1 is the number of iterations, and t=0 at initialization; X minj3 and X maxj3 are the minimum and maximum boundaries of the j3rd dimension respectively; r i3j3 is a uniformly distributed random number ranging from 0 to 1; T is the maximum number of iterations.

[0048] By performing the above operations, this solution combines random factors with time parameters to construct diverse system parameter initialization strategies, so that the initialized individual positions are adjusted with the number of iterations, increasing global randomness and making the system adjustment parameters more convergent, thereby generating more robust prediction results and improving the system's prediction efficiency for evaluation results.

[0049] Example 5, see Figure 2 and Figure 3 This embodiment is based on the above embodiment. In the psychiatric treatment effect evaluation model establishment module, the loss function design unit specifically includes the following contents:

[0050] Define an estimation range. Generate an estimation range for improving the assessment level. Define the upper bound of the estimation range. and lower bound ; and ensure the maximum number of real improvement assessment levels It falls within the estimated range and is expressed as: ; Introduce capture indicator k i , used to indicate whether each estimated range captures the true improvement assessment level, expressed as: ; Calculate the estimated number of successfully captured ranges c, expressed as: ; Where Pr(·) is the probability; α is the uncertainty parameter; n is the number of samples, and i is the sample index;

[0051] Define the interval estimation width; calculate the proportion PP of the estimated range that successfully captures the true treatment effect, expressed as: ; and calculate the average estimated width PW of the estimated range that captures the observed values, expressed as: ;

[0052] Construct the initial loss function; introduce adaptive Lagrange multipliers , for each sample, when its prediction error is larger, the adaptive Lagrange multiplier is larger, and the penalty for wrong prediction is increased; it is expressed as: ; ;in, is the initial multiplier; γ is the adaptive growth factor;

[0053] Define the smooth snap vector; the smooth snap vector is expressed as: ; ; Get the smooth capture rate PPs, expressed as: ; and the softened average estimated width PWs, expressed as: ;in, is the sigmoid function; s i is the scaling factor;

[0054] Define the error calibration term; used to penalize the estimated range that fails to capture the true treatment effect, expressed as: ; Where C is the error calibration term;

[0055] Construct the final loss function; obtain the final loss function LQ applied to the psychiatric treatment effect evaluation model, expressed as: .

[0056] By performing the above operations, the general psychiatric treatment effect evaluation system has the problem that the original evaluation data has samples that are difficult to capture, which leads to large system prediction errors, large uncertainty fluctuations, poor system adaptability to the evaluation data, and a large number of erroneous predictions. This solution provides a buffer for uncertainty by defining an estimation range to cover the actual improvement evaluation level; introducing capture rate, smooth capture rate and adaptive Lagrange multiplier, allowing the system to more strongly penalize predictions with large errors and penalize prediction intervals that fail to capture the true treatment effect, thereby reducing erroneous predictions of evaluation results and providing more accurate evaluation results.

[0057] Example 6, see Figure 1This embodiment is based on the above embodiment. The psychiatric treatment effect evaluation module is based on the established psychiatric treatment effect evaluation model, and collects physiological data, psychological data, drug response data, social activity data, environmental information and time data in real time; after preprocessing, it is input into the psychiatric treatment effect evaluation model, and the improvement evaluation level output by the model is used as the psychiatric treatment effect evaluation result.

[0058] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0059] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0060] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An intelligent psychiatric treatment effect evaluation system, characterized by: The system includes a data acquisition module, a data preprocessing module, a psychiatric treatment effect evaluation model building module and a psychiatric treatment effect evaluation module; The data collection module collects historical psychiatric treatment process data; The data preprocessing module performs data cleaning, data conversion and data set division on the collected data; The psychiatric treatment effect evaluation model establishment module provides a buffer for uncertainty by defining an estimation range to cover the actual improvement evaluation level; introduces capture rate, smooth capture rate and adaptive Lagrange multiplier to allow the system to more strongly penalize predictions with large errors and penalize prediction intervals that fail to capture the actual treatment effect, and based on the system parameter adjustment initialization strategy, realizes the establishment of a psychiatric treatment effect evaluation model; The psychiatric treatment effect evaluation module implements psychiatric treatment effect evaluation based on the established psychiatric treatment effect evaluation model; The psychiatric treatment effect evaluation model establishment module specifically includes the following contents: Model architecture design unit; the established psychiatric treatment effect evaluation model includes: multi-layer one-dimensional convolutional layers, which extract time series features from pre-processed historical psychiatric treatment process data and adapt to changes at different time scales; pooling layers, which reduce dimensions and extract main features; fully connected layers, which fuse multi-scale features and generate predicted values ​​of improvement evaluation levels; output layers, which output predicted improvement evaluation levels; Multi-layer one-dimensional convolutional layer design unit; using different convolution kernels, each layer of convolution kernel has a different window size, the convolution operation formula is expressed as follows: ; The feature map generated by each convolutional layer describes the local pattern of the historical psychiatric treatment process data. The feature maps from different convolutional layers are concatenated to form a multi-scale feature matrix, which represents the combination of short-term fluctuations and long-term trends; where F t is the output of the convolution operation; X t+m-1 is a local segment of the input sequence data starting from time t; W m is the weight of the convolution kernel; M is the window size of the convolution kernel, and m is the convolution kernel window size index; Loss function design unit; specifically includes the following contents: Define an estimation range. Generate an estimation range for improving the assessment level. Define the upper bound of the estimation range. and the lower bound ; and ensure the maximum number of real improvement assessment levels It falls within the estimated range and is expressed as: ; Introduce capture indicator k i , used to indicate whether each estimated range captures the true improvement assessment level, expressed as: ; Calculate the estimated number of successfully captured ranges c, expressed as: ; Where Pr(·) is the probability; α is the uncertainty parameter; n is the number of samples, and i is the sample index; Define the interval estimation width; calculate the proportion PP of the estimated range that successfully captures the true treatment effect, expressed as: ; and calculate the average estimated width PW of the estimated range that captures the observed values, expressed as: ; Construct the initial loss function; introduce adaptive Lagrange multipliers , for each sample, when its prediction error is larger, the adaptive Lagrange multiplier is larger, and the penalty for wrong prediction is increased; it is expressed as: ; ;in, is the initial multiplier; γ is the adaptive growth factor; Define the smooth snap vector; the smooth snap vector is expressed as: ; ; Get the smooth capture rate PPs, expressed as: ; and the softened average estimated width PWs, expressed as: ;in, is the sigmoid function; s i is the scaling factor; Define the error calibration term; used to penalize the estimated range that fails to capture the true treatment effect, expressed as: ; Where C is the error calibration term; Construct the final loss function; obtain the final loss function LQ applied to the psychiatric treatment effect evaluation model, expressed as: ； Model judgment unit; when the loss of the psychiatric treatment effect evaluation model for the training set converges, the training of the psychiatric treatment effect evaluation model is completed; a prediction threshold is set in advance, and when the prediction accuracy of the trained psychiatric treatment effect evaluation model for the test set is higher than the prediction threshold, the psychiatric treatment effect evaluation model is established; if it is not higher than the prediction threshold or the maximum number of training times is reached, it is transferred to the parameter optimization unit for parameter optimization and the data set is re-divided to train the psychiatric treatment effect evaluation model; Parameter optimization unit; construct parameter optimization space based on uncertainty parameters, initial multipliers, adaptive growth factors and initial weights and biases of psychiatric treatment effect evaluation model; initialize and optimize individual positions in the population, use the prediction accuracy of the psychiatric treatment effect evaluation model for the test set based on the individual position as the individual fitness value, and use the particle swarm search algorithm when updating the position; when there is an individual fitness value higher than the prediction threshold, the individual position is the optimized parameter setting, and the established psychiatric treatment effect evaluation model is obtained; otherwise, re-initialize the individual position in the population for parameter optimization; the individual position in the initialized optimization population is expressed as: ; Among them, X i3j3 is the initialization position of individual i3 in the j3th dimension, t1 is the number of iterations, and t=0 at initialization; X minj3 and X maxj3 are the minimum and maximum boundaries of the j3rd dimension respectively; r i3j3 is a uniformly distributed random number ranging from 0 to 1; T is the maximum number of iterations.

2. The intelligent psychiatric treatment effect evaluation system according to claim 1, characterized in that: In the data collection module, the historical psychiatric treatment process data includes physiological data, psychological data, drug response data, social activity data, environmental information, time data and improvement assessment level; Improve the evaluation level as data label.

3. The intelligent psychiatric treatment effect evaluation system according to claim 1, characterized in that: The psychiatric treatment effect evaluation module is based on the established psychiatric treatment effect evaluation model, and collects physiological data, psychological data, drug response data, social activity data, environmental information and time data in real time; After preprocessing, the data are input into the psychiatric treatment effect evaluation model, and the improvement assessment level output by the model is used as the psychiatric treatment effect evaluation result.

4. The intelligent psychiatric treatment effect evaluation system according to claim 1, characterized in that: In the data preprocessing module, the data cleaning is to process missing values ​​and duplicate values; the data conversion is to convert the data into vector form and perform standardization to obtain a time series data set; the data set partitioning is to divide the time series data set into a test set and a training set.

Images