Depression assessment method based on grey correlation analysis
Through a method based on gray correlation analysis, multiple characteristics of patients with depression are evaluated, which solves the problem of difficulty in accurately assessing depression in the prior art, and achieves the effectiveness of efficiently evaluating the changes in depression and the treatment plan under a small sample.
Patent Information
- Application Number
- CN202311064599.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-22
- Publication Date
- 2025-02-25
AI Technical Summary
The prior art is difficult to accurately evaluate depression and it is difficult to establish a large sample library to build accurate predictive models.
A method based on gray correlation analysis was used to score 28 depression characteristics in self-evaluation, psychological state, social support, physiological signs, living environment, etc., and a gray correlation algorithm was used to calculate the correlation coefficient of each depression characteristic value, and then the depression status of the tester was evaluated, and the historical data was compared to evaluate the depression dynamics.
A gray depression assessment model was established under a small sample, which could efficiently evaluate the changes in depression, and was used to evaluate the effectiveness of treatment plans and achieve efficient and individualized precise treatment.
Smart Images
Figure CN119517389A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of mental health detection, and in particular to a depression assessment method based on grey correlation analysis. Background Art
[0002] Depression is a common clinical mental illness with persistent low mood and loss of interest in life as core symptoms, and may be accompanied by cognition, sleep, appetite, weight changes or suicide, self-harm and other symptoms. According to the "2022 China Depression Blue Book", my country recognizes that the lifetime prevalence of depressive disorders is 6.8%, of which depression is 3.4%. It is estimated that 95 million people in my country suffer from depression; about 280,000 people commit suicide each year, of which 40% suffer from depression. Depression is a high-burden disease that requires long-term drug treatment and psychological intervention. It is difficult and long to cure, and it brings huge losses to society and families. The prevention, detection and diagnosis of depression are current research hotspots.
[0003] At present, the main clinical diagnostic methods for depression are: relying on the subjective judgment of doctors, that is, doctors use depression assessment scales for screening; using biological information for evaluation, such as electroencephalogram (EEG), functional magnetic resonance imaging (fMRI), electrocardiogram (ECG) and other technologies; although there are scales that can evaluate depressive disorders, depression is complex and the onset time is difficult to predict, so the evaluation results of a single scale are not ideal. In recent years, data mining technology has been widely used in many fields; there are still few studies on the use of data mining technology in the evaluation and prediction of psychological states. Due to the complexity of people's cognition and the diversity of living environments, the number of similar samples is not large, and intelligent algorithms such as deep learning and machine learning require a large amount of training data to build a good prediction and evaluation model. The advantage of gray system theory is to achieve the establishment of prediction models, prediction of development trends, decision-making of things, and effective and reasonable evaluation of system control and status through data-based processing and phenomenon analysis under the uncertainty background of less data and less information.
[0004] In order to solve the problems that a single scale is difficult to accurately evaluate depression and it is difficult to establish a large sample library and then use artificial intelligence methods to build an accurate prediction model, the present invention proposes a gray depression assessment model based on a small sample using gray system theory, and uses this model to evaluate the effectiveness of treatment plans based on the patient's psychological state. Summary of the invention
[0005] The present invention provides a psychological depression assessment method based on grey correlation analysis. The method first scores 28 depression characteristics in terms of self-evaluation, psychological state, social support, physiological signs, living environment, etc., and then uses a grey correlation algorithm to calculate the correlation coefficient of each depression characteristic value. The correlation coefficient represents the depression state of the tester, and the current correlation coefficient is compared with the historical correlation coefficient to further evaluate the depression dynamics of the tester.
[0006] See the following description for details: First, from the aspects of self-evaluation, psychological state, social support, physiological signs, living environment, etc., the selected depression characteristics are not limited to: age, gender, education, sadness, insomnia, memory loss, suicidal impulse, suicidal behavior, economic pressure, self-image, childhood trauma, anxiety, weight change, hypochondria, substance dependence, chronic disease, life interest, physical disease, nightmare, pessimism, failure experience, warm feeling of family or friends, interest in social activities, fatigue, mental labor blockage, concentration, pleasure, living area, and the depression characteristics are implanted into the questionnaire, and the test subjects are asked to answer the questionnaire and obtain the scores of each characteristic. The score set is X i =[x1, x2…x 28 ] T , the depression state is Y; Then collect training data to form a comparison matrix X, that is Where n is the number of training data sets; Depressive state Y = [y1…y n ], where n is the number of training data sets; The test subjects' scores and judged depression status are analyzed. Set the coefficient of each factor W = [w1, w2…w 28 ] Then there is Y=WX (2) Using the least squares method, we can find W=YX -1 (3) Among them, X -1 is the generalized inverse matrix of X.
[0007] Then a score set is formed according to the patient's score measured each time, and its Z j =[z 1，j , z 2，j …z 28，j ] T Where j is the number of measurements, j = 1, 2...n. The reference sequence of its evaluation standard data is Z0=[z 0，1, z 0，2 …z 0，28 ] T (4) Among them, z 0，1 =max(z 1，j ), j = 1, 2…28, Further, if the index is dimensionless, we have in, Its correlation coefficient is Here, ρ is the resolution coefficient, which is used to increase the significance of the difference between the correlation coefficients. Here, ρ=0.5. Its correlation is Then γ i The value can represent the depression level at the time of the i-th test, γ i The larger the value, the deeper the depression.
[0007] The system workflow is as follows: First, the least squares method is used to find the coefficients of each indicator based on the depression training data. During the test, each patient's condition is tested each time. Each patient's historical measurement set from the first time to the present is involved in calculating the correlation. According to the solved correlation γ i The value of γ is the same as the last time i -1 value comparison, if γ i >γ i-1 If γ i <γ i-1 Depression is relieved, γ i =γ i-1 No improvement.
[0008] The beneficial effects of the technical solution provided by the present invention are: The present invention is designed to determine changes in depression in patients by calculating the correlation between the patient's current and historical measurement factors, which can be used to evaluate the effectiveness of depression treatment plans, thereby achieving efficient, individualized and precise treatment. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0010] Figure 1This is a schematic diagram of the factors associated with depression;
[0011] Figure 2 This is a flowchart of the depression assessment method based on grey relational analysis. DETAILED DESCRIPTION
[0012] The purpose of the present invention is to provide a psychological depression assessment method based on grey correlation analysis, which first scores 28 depression characteristics in terms of self-evaluation, psychological state, social support, physiological signs, living environment, etc., and then uses the grey correlation algorithm to calculate the correlation coefficient of each depression characteristic value, and the correlation coefficient characterizes the depression state of the tester, and compares the current correlation coefficient with the historical correlation coefficient, further evaluates the depression dynamics of the tester, and then uses it to evaluate the effectiveness of the current treatment plan to achieve efficient, personalized and precise treatment. In order to make the purpose, technical solution and advantages of the present invention clearer, the following is a further detailed description of the embodiments of the present invention. like Figure 1 As shown, the depression characteristics selected from self-evaluation, psychological state, social support, physiological signs, living environment, etc. are not limited to: age, gender, education level, sadness, insomnia, memory loss, suicidal impulses, suicidal behavior, economic pressure, self-image, childhood trauma, anxiety, weight change, hypochondria, substance dependence, chronic disease, life interests, physical diseases, nightmares, pessimism, failure experiences, warm feelings from family or friends, interest in social activities, fatigue, mental labor blockage, concentration, pleasure, and residential area. The depression characteristics are also embedded in the questionnaire.
[0013] At the same time, based on the patients' answers and the judgment of their depression status by several senior experts, 20 experts were invited to judge 50 depressed patients. Each expert scored their depression status with a ratio of 1-10 points. The average value of its solution is the final depression state Y of the training set; Let the testers answer the questionnaire and obtain the scores of each feature. The score set is X i =[x1, x2…x 28 ] T , the depression state is Y; Then collect training data to form a comparison matrix X, that is Where n is the number of training data sets, n = 50; Depressive state Y = [y1…y n ], where n is the number of training data sets, n = 50; The test subjects' scores and judged depression status are analyzed. Set the coefficient of each factor W = [w1, w2…w28 ] Then there is Y=WX (2) Using the least squares method, we can find W=YX -1 (3) X -1 is the generalized inverse matrix of X.
[0014] like Figure 2 As shown, if the coefficients of each factor have been solved, the score set Z is formed according to the scores of each patient measurement j =[z 1，j , z 2，j …z 28，j ] T , where j is the number of measurements, j = 1, 2…n. The reference sequence of its evaluation standard data is Z0=[z 0，1 , z 0，2 …z 0，28 ] T (4) Among them, z 0，1 =max(z 1，j ), j = 1, 2…28, Further, if the index is dimensionless, we have Its association The coefficient is Here, ρ is the resolution coefficient, which is used to improve the significance of the difference between the correlation coefficients. Here, ρ=0.5, and the correlation degree is Then γ i The value can represent the depression level at the time of the i-th test, γ i The larger the value, the deeper the depression.
[0015] The system workflow is as follows: First, the least squares method is used to find the coefficients of each indicator based on the depression training data. During the test, each patient's condition is tested each time. Each patient's historical measurement set from the first time to the present is involved in calculating the correlation. According to the solved correlation γ i The value of γ is the same as the last time i-1 Value comparison, if γ i >γ i-1 If γ i <γ i-1 Depression is relieved, γ i =γ i-1 No improvement.
[0016] The beneficial effects of the technical solution provided by the present invention are:
[0017] The present invention is designed to determine changes in depression in patients by calculating the correlation between the patient's current and historical measurement factors, which can be used to evaluate the effectiveness of depression treatment plans, thereby achieving efficient, individualized and precise treatment.
[0018] Unless otherwise specified, the models of the components in the embodiments of the present invention are not limited, and any device that can perform the above functions may be used.
[0019] Those skilled in the art will appreciate that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0020] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. The present invention provides a depression assessment method based on grey correlation analysis, characterized in that: Scores are given for 28 depressive characteristics in terms of self-evaluation, psychological state, social support, physiological signs, living environment, etc., and then the grey correlation algorithm is used to calculate the correlation coefficient of each depression characteristic value. The correlation coefficient represents the depressive state of the tester, and the current correlation coefficient is compared with the historical correlation coefficient to further evaluate the tester's depression dynamics.
2. According to claim 1, the present invention provides a depression assessment method based on grey relational analysis, characterized in that: The evaluation factor set includes self-evaluation, psychological state, social support, physiological signs, living environment and other aspects. The selected depressive characteristics are not limited to: age, gender, education level, sadness, insomnia, memory loss, suicidal impulses, suicidal behavior, economic pressure, self-image, childhood trauma, anxiety, weight change, hypochondria, substance dependence, chronic diseases, life interests, physical diseases, nightmares, pessimism, failure experiences, warm feelings from family or friends, interest in social activities, fatigue, mental labor blockage, concentration, pleasure, and residential area.
3. According to claim 1, the present invention provides a depression assessment method based on grey relational analysis, characterized in that: The method for obtaining the coefficients of each factor is as follows: Let the trainees answer the questionnaire and obtain the scores of each feature. The score set is X i =[x1, x2…x 28 ] T , the depression state is Z; Then collect training data to form a comparison matrix X, that is Where n is the number of training data sets; Depressive state Y=[y1…y n ] Where n is the number of training data sets; The test subjects' scores and judged depression status are analyzed. Set the coefficient of each factor W = [w1, w2…w 28 ] Then there is Y=WX Using the least squares method, we can find W = YX -1。 4. According to claim 1, the present invention provides a depression assessment method based on grey relational analysis, characterized in that: The method for solving the current and historical depression correlation coefficients is: Then the score set Z is formed according to the patient's score measured each time j =[z 1，j , …z 28，j ] T , where j is the number of measurements, j = 1, 2…n. The reference sequence of its evaluation standard data is Z0=[z 0，1 ,With 0，2 …With 0，28 ] T Among them, z 0，1 =max(z 1，j ), j = 1, 2…28. Further, if the index is dimensionless, we have in, Its correlation coefficient is Here, ρ is the resolution coefficient, which is used to increase the significance of the differences between the correlation coefficients. Here, the value is ρ = 0.
5. Its correlation is Then γ i The value can represent the depression level at the time of the i-th test, γ i The larger the value, the deeper the depression.