Obsessive-compulsive disorder diagnosis and symptom degree evaluation system

Abstract

The invention relates to an obsessive-compulsive disorder diagnosis and symptom degree evaluation system, and belongs to the technical field of medical instruments. The evaluation system comprises a knowledge base module used for storing evaluation questions and used for responding to an evaluation request of a testee and randomly calling the evaluation questions stored in the knowledge base module, the question and answer evaluation module is used for acquiring face video data and electroencephalogram data of the testee in a detection time period and calling a next round of questions according to each round of voice or character answer information of the testee; the acquisition module is used for acquiring face video data and electroencephalogram data of the testee in the detection time period; and the analysis and evaluation module is used for analyzing the compulsive behavior and the compulsive thinking degree of the detected person in the detection time period. According to the invention, optical flow analysis and 3D face reconstruction sentiment analysis and evaluation are carried out based on face video data, the analysis and evaluation process is more efficient, and the analysis and evaluation result is more accurate; according to the method, multi-modal data such as electroencephalogram signals and 3D reconstruction are combined, and the emotion recognition effect can be further improved.

Description

Technical Field

[0001] The invention relates to an obsessive-compulsive disorder diagnosis and symptom degree assessment system, belonging to the technical field of medical devices. Background Art

[0002] Obsessive-compulsive disorder is a common mental disorder characterized by severe, recurring obsessions and / or compulsive behaviors that lead to impairment of the patient's social functioning; obsessions refer to persistent, unwanted thoughts, impulses, or images that an individual experiences, knowing that they are unnecessary but cannot control; these thoughts, impulses, or images are intrusive and cause significant anxiety or distress; compulsive behaviors refer to repetitive behaviors or ritualistic actions that an individual performs to reduce the anxiety caused by obsessions.

[0003] The diagnosis of obsessive-compulsive disorder usually needs to be carried out by a professional psychiatrist or psychologist, who will use standardized diagnostic tools to assess the presence of symptoms and their impact on the individual's daily life. In the prior art, such as the Chinese patent, publication number CN116312970B discloses an intelligent interactive method and device for psychological assessment, wherein the method is performed by an intelligent robot, including: responding to the user's assessment request, opening an intelligent question-and-answer session, and asking questions to the user; generating the next round of questions based on the voice answer information of each round of users; collecting the user's face video data during the entire assessment process, generating the user's psychological assessment report based on the face video data and historical question-and-answer information, and generating the user's psychological assessment report based on the face video data and historical question-and-answer information by algorithm means, which can accurately judge the user's psychological problems such as depression and anxiety, assist doctors in diagnosis, reduce the doctor's face-to-face consultation burden, and avoid the absolute subjectivity of doctors' diagnosis; however, this method of extracting facial features by regional segmentation of face video data is relatively complex and easily affected by factors such as lighting and posture, resulting in low diagnostic accuracy. In addition, the latest International Diagnostic System for Disease (ICD-11) removes OCD from the spectrum of anxiety disorders and classifies it as a large category with impulse disorders. The new diagnostic classification emphasizes the impulsivity and repetitiveness of OCD. Therefore, the original diagnostic system that focuses on facial expressions is not suitable for the new diagnostic system and is more likely to lead to misdiagnosis. Therefore, the technical field urgently needs a new OCD diagnosis and symptom assessment system that focuses more on analyzing the degree of compulsive behavior and obsessive thinking symptoms. It can assist doctors in more accurately analyzing compulsive behavior and assessing the degree of obsessive thinking symptoms by integrating and analyzing information data from different sources. Summary of the invention

[0004] The purpose of the present invention is to solve the technical problem of how to assist doctors in more accurately analyzing compulsive behaviors and evaluating the degree of obsessive thinking symptoms by integrating and analyzing information data from different sources.

[0005] In order to achieve the purpose of solving the above problems, the technical solution adopted by the present invention is to provide an obsessive-compulsive disorder diagnosis and symptom degree assessment system, the assessment system comprising:

[0006] Knowledge base module, used to store evaluation questions;

[0007] The question-answering evaluation module is used to respond to the evaluation request of the test subject, randomly retrieve the evaluation questions stored in the knowledge base module, and retrieve the next round of questions according to the voice or text answer information of the test subject in each round;

[0008] The acquisition module is used to collect face video data and EEG data of the subject during the detection period;

[0009] The analysis and evaluation module is used to analyze the subject's compulsive behavior and compulsive thinking level during the test period, including voice or text answer duration analysis, voice or text answer content analysis, face video data analysis and EEG data analysis; the face video data analysis refers to optical flow analysis and 3D facial reconstruction emotion analysis evaluation based on face video data.

[0010] Preferably, the EEG data analysis comprises the following steps:

[0011] S301, converting a time domain signal into a frequency domain signal;

[0012] S302. Observe the energy distribution of different frequency bands by calculating the power spectrum density:

[0013] PSD(f)=|X(f)| 2

[0014] Where X(f) is the Fourier transform value of the signal x(t) at frequency f;

[0015] S303, calculate the relative energy ratio of α wave (8-12 Hz) and β wave (13-30 Hz):

[0016]

[0017] Among them, R α / β A lower value means that the brain is in a higher state of alertness, which is consistent with the common EEG characteristics of patients with obsessive-compulsive disorder;

[0018] S304. Obtain analysis results according to the relative energy ratio.

[0019] Preferably, the optical flow analysis includes analyzing the motion vectors of pixels between video frames to obtain changes in facial expressions, and judging the compulsive behavior based on the frequency and intensity of changes in facial expressions:

[0020] The basic formula of optical flow analysis:

[0021] in, is the gradient of the image in the x direction, is the gradient of the image in the y direction, is the rate of change of the image in the time t direction, u is the speed of the pixel in the x direction, and v is the speed of the pixel in the y direction;

[0022] Set up linear equation analysis:

[0023]

[0024] Among them, I x Represents the gradient of the image in the x direction, I y Represents the gradient of the image in the y direction, I t is the rate of change of the image in the time t direction, w(x,y) is the weight function used to weight the influence of adjacent pixels, is the optical flow vector, which indicates the speed of the pixel in the x and y directions;

[0025] Assume there are two adjacent frames I1 and I2, and calculate and Estimate the optical flow of point (x0, y0), assuming that the gradients at point (x0, y0) are I x (x0,y0)=5,I y (x0, y0) = 3, and the rate of change is I t (x0, y0) = -2, establish the following equation:

[0026]

[0027] Solving it, we get 5u+3v=2

[0028] In order to solve u and v, we need to construct a linear equation system, which can be achieved by integration or weighted averaging in a small area: Assume that the optical flow is consistent in a 3X3 area around the point (x0, y0);

[0029] Suppose that in a small area around the point (x0, y0), I x ,I y with I t The values ​​are:

[0030]

[0031] Construct the following system of equations:

[0032]

[0033] Compute the matrix:

[0034]

[0035] ∑ (x,y) I x I y =5·3+4·2+…+5·3+4·2+3·1=63

[0036] ∑ (x,y) I x I t =5·(-2)+4(·-1)+…+5(·-2)+4(·-1)+3·0=-45

[0037] Σ (x,y) I y I t =3·(-2)+2·(-1)+…+3(·-2)+2(·-1)+1·0=-18

[0038] Construct the system of equations:

[0039]

[0040] Solve the equations to get the values ​​of u and v, calculate the optical flow vector (u, v) at a certain point, and then analyze the pixel movement between video frames and the changes in facial expressions.

[0041] Preferably, the 3D facial reconstruction emotion analysis evaluation comprises the following steps:

[0042] S601, feature point detection: detecting facial feature points in the frame image of the face video data;

[0043] S602, 3D model fitting: matching the detected feature points with the 3D template model to obtain the 3D shape of the face;

[0044] S603, Expression parameterization: Use parameterized models to represent expression changes and capture subtle changes in the face.

[0045] Preferably, the voice or text answer duration analysis includes:

[0046] Calculate the average time T for answering avg To see if there is a significant difference:

[0047]

[0048] Among them, T i is the duration of the ith answer, and n is the total number of answers;

[0049] Calculate the standard deviation S of the answer time t , get the range of answer time:

[0050]

[0051] A larger standard deviation indicates greater fluctuation in response time, reflecting uncertainty or delays in the respondent's thinking process when answering.

[0052] Preferably, the voice or text answer content analysis includes the following steps:

[0053] S801, pre-processing the voice or text answer information of each round of users;

[0054] S802, using LDA (Latent Dirichlet Allocation) topic modeling to identify topics in the information;

[0055] S803, clustering similar contents together using a clustering algorithm;

[0056] S804. Count recurring themes or words and analyze the order of occurrence of words or phrases to identify specific patterns;

[0057] S805. Calculate the frequency of occurrence of each topic or repeated word, and observe the changing trend of the topic over time.

[0058] Preferably, the EEG data collection comprises the following steps:

[0059] S201, preparation stage: clean the subject's scalp, put the electrode cap on the subject's head, apply conductive paste, and use the EEG data acquisition software to set the sampling rate parameters;

[0060] S202, acquisition phase: let the subject complete the corresponding task according to the instructions or keep a natural state, start recording the EEG signal, adjust the electrode position or re-apply the conductive paste when necessary to ensure the quality of the monitoring signal;

[0061] S203, data processing: after stopping recording, save the EEG data, and perform data cleaning and filtering.

[0062] Preferably, the data cleaning in step S203 includes the following steps:

[0063] S901, bad track detection, check whether there is a constant value or extremely high impedance in each channel;

[0064] S902, eliminate anomalies, and remove data points or channels that are obviously beyond the normal range;

[0065] S903, interpolation repair, uses interpolation technology to fill the gaps left after removing outliers.

[0066] The present invention provides a storage device, wherein the storage device stores a plurality of instructions, and the instructions are used to execute the evaluation work of the above-mentioned obsessive-compulsive disorder diagnosis and symptom degree evaluation system.

[0067] The present invention provides an embedded device, comprising: a storage device for storing a computer program; a processor for executing the computer program to implement the evaluation work of the above-mentioned obsessive-compulsive disorder diagnosis and symptom degree evaluation system.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. Compared with the prior art which only performs regional segmentation on face video data, the present invention adopts optical flow analysis and 3D facial reconstruction emotion analysis and evaluation based on face video data. The analysis and evaluation process is more efficient and the analysis and evaluation results are more accurate, especially in large data sets and complex lighting scenes.

[0070] 2. The analysis and evaluation process of the present invention combines multimodal data such as EEG signals and 3D reconstruction to further improve the effects of emotion recognition and emotion intensity judgment.

[0071] 3. The present invention can greatly improve the accuracy and efficiency of OCD diagnosis by integrating information from different sources and utilizing modern information technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A structural block diagram of the obsessive-compulsive disorder diagnosis and symptom severity assessment system provided by the present invention;

[0073] Figure 2 A flowchart of the EEG data collection process provided by the present invention;

[0074] Figure 3 A flowchart of the EEG data analysis provided by the present invention;

[0075] Figure 4 This is a flowchart of the face video data collection process provided by the present invention. DETAILED DESCRIPTION

[0076] In order to make the present invention more clearly understood, a preferred embodiment is described in detail with reference to the accompanying drawings as follows:

[0077] like Figure 1-4 As shown, the present invention provides an obsessive-compulsive disorder diagnosis and symptom degree assessment system, the system comprising:

[0078] Knowledge base module, used to store evaluation questions;

[0079] The question-and-answer evaluation module is used to respond to the evaluation request of the test subject, randomly retrieve the evaluation questions stored in the knowledge base module, and retrieve the next round of questions based on the voice or text answer information of the test subject in each round;

[0080] The acquisition module is used to collect face video data and EEG data of the subject during the detection period;

[0081] The analysis and evaluation module is used to analyze the subject's compulsive behavior and compulsive thinking during the test period, including voice or text answer duration analysis, voice or text answer content analysis, face video data analysis, and EEG data analysis;

[0082] Among them, the facial video data analysis refers to optical flow analysis and 3D facial reconstruction emotion analysis evaluation based on facial video data.

[0083] The EEG data collection includes the following steps:

[0084] S201, preparation stage: clean the subject's scalp, put the electrode cap on the subject's head, apply conductive paste, and use the EEG data acquisition software to set the sampling rate parameters;

[0085] S202, acquisition phase: let the subject complete the corresponding task according to the instructions or keep a natural state, start recording the EEG signal, adjust the electrode position or re-apply the conductive paste when necessary to ensure the quality of the monitoring signal;

[0086] S203, data processing: after stopping recording, save the EEG data, and perform data cleaning and filtering.

[0087] The EEG data analysis comprises the following steps:

[0088] S301, converting a time domain signal into a frequency domain signal;

[0089] S302, observing energy distribution of different frequency bands by calculating power spectrum density;

[0090] S303, calculating the relative energy ratio of α wave (8-12 Hz) and β wave (13-30 Hz);

[0091] S304. Obtain analysis results according to the relative energy ratio.

[0092] The face video data collection includes the following steps:

[0093] S401, preparation stage: set the position and angle of the camera, ensure appropriate lighting, and use video recording software to set recording parameters;

[0094] S402, recording stage: start the video recording software, inform the subject to start recording after getting ready, and keep the background quiet during the recording process;

[0095] S403, storage: saving the video file after completing the recording.

[0096] The optical flow analysis includes analyzing the motion vectors of pixels between video frames to obtain changes in facial expressions, and judging the degree of pain caused by compulsive behavior and compulsive thinking based on the frequency and intensity of changes in facial expressions.

[0097] The 3D facial reconstruction emotion analysis evaluation comprises the following steps:

[0098] S601, feature point detection: detecting facial feature points in the frame image of the face video data;

[0099] S602, 3D model fitting: matching the detected feature points with the 3D template model to obtain the 3D shape of the face;

[0100] S603, Expression parameterization: Use parameterized models to represent expression changes and capture subtle changes in the face.

[0101] The voice or text answer duration analysis includes:

[0102] Calculate the average length of time to answer to see if there is a significant difference. If the standard deviation is large, it means that the fluctuation in the length of the answer is also large, reflecting the uncertainty of the respondents when answering or the delay in the thinking process.

[0103] The analysis of the content of the voice or text answer includes:

[0104] S801, pre-processing the voice or text answer information of each round of users;

[0105] S802, using LDA topic modeling to identify topics in the information;

[0106] S803, clustering similar contents together using a clustering algorithm;

[0107] S804. Count recurring themes or words and analyze the order of occurrence of words or phrases to identify specific patterns;

[0108] S805. Calculate the frequency of occurrence of each topic or repeated word, and observe the changing trend of the topic over time.

[0109] The data cleaning comprises the following steps:

[0110] S901, bad track detection, check whether there is a constant value or extremely high impedance in each channel;

[0111] S902, eliminate anomalies, and remove data points or channels that are obviously beyond the normal range;

[0112] S903, interpolation repair, uses interpolation technology to fill the gaps left after removing outliers.

[0113] Example

[0114] like Figure 1 As shown, the present invention provides an obsessive-compulsive disorder diagnosis and symptom assessment system comprising:

[0115] 1. Knowledge base module, used to store evaluation questions;

[0116] 2. Question and answer evaluation module, which is used to respond to the evaluation request of the test subject, randomly retrieve the evaluation questions stored in the knowledge base module, and then retrieve the next round of questions based on the voice or text answer information of the test subject in each round;

[0117] 3. Collection module, used to collect face video data and EEG data of the subject during the detection period. The collection module includes a face capture camera and an EEG (Electroencephalogram, EEG) collection device;

[0118] 4. Analysis and evaluation module, used to analyze the compulsive behavior of the subject during the detection period, including analysis of the duration of voice or text answers, voice or text answer content, face video data and EEG data, and perform optical flow analysis and 3D facial reconstruction emotion analysis evaluation based on face video data.

[0119] like Figure 2 As shown, EEG data collection includes the following steps:

[0120] S201, preparation stage: clean the subject's scalp to reduce impedance, put the electrode cap on the subject's head, apply conductive paste, and use the EEG data acquisition software to set the sampling rate parameters;

[0121] S202, acquisition phase: let the subject complete the corresponding task according to the instructions or maintain a natural state, start recording the EEG signal, adjust the electrode position or re-apply the conductive paste when necessary to ensure the quality of the monitoring signal.

[0122] S203, data processing: after stopping recording, save the EEG data, and perform data cleaning and filtering.

[0123] EEG data cleaning includes the following steps:

[0124] S901, bad track detection, check whether there is a constant value or extremely high impedance in each channel;

[0125] S902, eliminate anomalies, and remove data points or channels that are obviously beyond the normal range;

[0126] S903, interpolation repair, uses interpolation technology to fill the gaps left after removing outliers.

[0127] Filtering: Use a bandpass filter to select an appropriate frequency range (e.g., 0.5-50 Hz) to retain useful information in the EEG signal, which can provide high-quality data support for further analysis and application;

[0128] Furthermore, if Figure 3 As shown, EEG data analysis includes the following steps:

[0129] S301, converting a time domain signal into a frequency domain signal;

[0130] S302. Observe the energy distribution of different frequency bands by calculating the power spectrum density:

[0131] PSD(f)=|X(f)| 2

[0132] Where X(f) is the Fourier transform value of the signal x(t) at frequency f;

[0133] S303, calculate the relative energy ratio of α wave (8-12 Hz) and β wave (13-30 Hz):

[0134]

[0135] Among them, R α / β A lower value means that the brain is in a higher state of alertness, which is consistent with the common EEG characteristics of patients with obsessive-compulsive disorder;

[0136] S304. Obtain analysis results according to the relative energy ratio.

[0137] like Figure 4 As shown, face video data collection includes the following steps:

[0138] S401, preparation stage: set the position and angle of the camera, ensure appropriate lighting, and use video recording software to set recording parameters such as resolution, frame rate, etc.;

[0139] S402, recording stage: start the video recording software, inform the subject to start recording after getting ready, and keep the background quiet during recording to reduce interference;

[0140] S403, storage: saving the video file after completing the recording.

[0141] Furthermore, optical flow analysis includes analyzing the motion vectors of pixels between video frames, obtaining changes in facial expressions, and judging compulsive behavior based on the frequency and intensity of changes in facial expressions:

[0142] The basic formula of optical flow analysis:

[0143] in, is the gradient of the image in the x direction, is the gradient of the image in the y direction, is the rate of change of the image in the time t direction, u is the speed of the pixel in the x direction, and v is the speed of the pixel in the y direction;

[0144] Set up linear equation analysis:

[0145]

[0146] Among them, I x Represents the gradient of the image in the x direction, I y Represents the gradient of the image in the y direction, I t is the rate of change of the image in the time t direction, w(x,y) is the weight function used to weight the influence of adjacent pixels, is the optical flow vector, which indicates the speed of the pixel in the x and y directions;

[0147] Assume there are two adjacent frames I1 and I2, and calculate and Estimate the optical flow of point (x0, y0), assuming that the gradients at point (x0, y0) are I x (x0,y0)=5,I y (x0, y0) = 3, and the rate of change is I t (x0, y0) = -2, establish the following equation:

[0148]

[0149] Solving it, we get 5u+3v=2

[0150] In order to solve u and v, we need to construct a linear equation system, which can be achieved by integration or weighted averaging in a small area: Assume that the optical flow is consistent in a 3X3 area around the point (x0, y0);

[0151] Suppose that in a small area around the point (x0, y0), I x ,I y with I t The values ​​are:

[0152]

[0153] Construct the following system of equations:

[0154]

[0155] Compute the matrix:

[0156]

[0157] ∑ (x,y) I x I y =5·3+4·2+…+5·3+4·2+3·1=63

[0158] Σ (x,y) I x I t =5·(-2)+4·(-1)+…+5·(-2)+4·(-1)+3·0=-45

[0159] ∑ (x,y) I y I t =3·(-2)+2·(-1)+…+3·(-2)+2·(-1)+1·0=-18

[0160] Construct the system of equations:

[0161]

[0162] By solving the equations to get the values ​​of u and v, we can calculate the optical flow vector (u, v) at a certain point, and then analyze the pixel movement between video frames, making the analysis of facial expression changes more accurate.

[0163] Furthermore, 3D facial reconstruction emotion analysis includes the following steps:

[0164] S601, feature point detection: detecting facial feature points in the frame image of the face video data;

[0165] S602, 3D model fitting: matching the detected feature points with the 3D template model to obtain the 3D shape of the face;

[0166] S603, Expression parameterization: Use parameterized models to represent expression changes and capture subtle changes in the face.

[0167] Voice or text answer duration analysis includes:

[0168] Calculate the average time T for answering avg To see if there is a significant difference:

[0169]

[0170] Among them, T iis the duration of the ith answer, and n is the total number of answers;

[0171] In addition, the standard deviation S of the answer time is calculated t , get the range of answer time:

[0172]

[0173] A larger standard deviation indicates greater fluctuation in response time, which may reflect uncertainty or delays in the respondent's thinking process when answering.

[0174] Content analysis of spoken or written responses includes:

[0175] S801, pre-processing the voice or text answer information of each round of users;

[0176] S802, using LDA (Latent Dirichlet Allocation) topic modeling to identify topics in the information;

[0177] S803, clustering similar contents together using a clustering algorithm;

[0178] S804. Count recurring themes or words and analyze the order of occurrence of words or phrases to identify specific patterns;

[0179] S805. Calculate the frequency of occurrence of each topic or repeated word, and observe the changing trend of the topic over time.

[0180] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0181] The above is only a preferred embodiment of the present invention, and is not any formal or substantial limitation of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and supplements without departing from the present invention, and these improvements and supplements should also be regarded as the protection scope of the present invention. Any technician familiar with this profession, without departing from the spirit and scope of the present invention, can make some changes, modifications and equivalent changes made by using the technical content disclosed above, which are equivalent embodiments of the present invention; at the same time, any changes, modifications and evolutions of any equivalent changes made to the above embodiments based on the essential technology of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An obsessive-compulsive disorder diagnosis and symptom severity assessment system, characterized in that: The evaluation system comprises: Knowledge base module, used to store evaluation questions; The question-answering evaluation module is used to respond to the evaluation request of the test subject, randomly retrieve the evaluation questions stored in the knowledge base module, and retrieve the next round of questions according to the voice or text answer information of the test subject in each round; The acquisition module is used to collect face video data and EEG data of the subject during the detection period; The analysis and evaluation module is used to analyze the subject's compulsive behavior and compulsive thinking level during the test period, including voice or text answer duration analysis, voice or text answer content analysis, face video data analysis and EEG data analysis; the face video data analysis refers to optical flow analysis and 3D facial reconstruction emotion analysis evaluation based on face video data.

2. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The EEG data analysis comprises the following steps: S301, converting a time domain signal into a frequency domain signal; S302. Observe the energy distribution of different frequency bands by calculating the power spectrum density: PSD(f)=|X(f)|2 Where X(f) is the Fourier transform value of the signal x(t) at frequency f; S303, calculate the relative energy ratio of alpha wave 8-12 Hz and beta wave 13-30 Hz: Among them, Rα / β A lower value means that the brain is in a higher state of alertness, which is consistent with the common EEG characteristics of patients with obsessive-compulsive disorder; S304. Obtain analysis results according to the relative energy ratio.

3. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The optical flow analysis includes analyzing the motion vectors of pixels between video frames, obtaining changes in facial expressions, and judging compulsive behaviors based on the frequency and intensity of changes in facial expressions: The basic formula of optical flow analysis: in, is the gradient of the image in the x direction, is the gradient of the image in the y direction, is the rate of change of the image in the time t direction, u is the speed of the pixel in the x direction, and v is the speed of the pixel in the y direction; Establish linear equation analysis: Among them, I x Represents the gradient of the image in the x direction, I y Represents the gradient of the image in the y direction, I t is the rate of change of the image in the time t direction, w(x, y) is the weight function used to weight the influence of adjacent pixels, is the optical flow vector, which indicates the speed of the pixel in the x and y directions; Assume there are two adjacent frames I1 and I2, and calculate and Estimate the optical flow of point (x0, y0), assuming that the gradients at point (x0, y0) are I x (x0, y0) = 5, I y (x0, y0) = 3, and the rate of change is t (x0, y0) = -2, establish the following equation: Solving it, we get 5u+3v=2 In order to solve u and v, we need to construct a linear equation system, which can be achieved by integration or weighted averaging in a small area: Assume that the optical flow is consistent in a 3×3 area around the point (x0, y0); Suppose that in a small area around the point (x0, y0), I x ,I y with I t The values ​​are: Construct the following system of equations: Compute the matrix: ∑ (x，y) I x I y =5·3+4·2+...+5·3+4·2+3·1=63 ∑ (x，y) I x I t =5·(-2)+4·(-1)+...+5·(-2)+4·(-1)+3·0=-45 ∑ (x，y) I y I t =3·(-2)+2·(-1)+...+3·(-2)+2·(-1)+1·0=-18 Construct the system of equations: Solve the equations to get the values ​​of u and v, calculate the optical flow vector (u, v) at a certain point, and then analyze the pixel movement between video frames and the changes in facial expressions.

4. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The 3D facial reconstruction emotion analysis evaluation comprises the following steps: S601, feature point detection: detecting facial feature points in the frame image of the face video data; S602, 3D model fitting: matching the detected feature points with the 3D template model to obtain the 3D shape of the face; S603, Expression parameterization: Use parameterized models to represent expression changes and capture subtle changes in the face.

5. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The voice or text answer duration analysis includes: Calculate the average time T for answering avg Let's see if there are any significant differences: Among them, T i is the duration of the ith answer, and n is the total number of answers; Calculate the standard deviation S of the answer time t , get the range of answer time: A larger standard deviation indicates greater fluctuation in response time, reflecting uncertainty or delays in the respondent's thinking process when answering.

6. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The voice or text answer content analysis includes the following steps: S801, pre-processing the voice or text answer information of each round of users; S802, using LDA Latent Dirichlet Allocation topic modeling to identify topics in the information; S803, clustering similar contents together using a clustering algorithm; S804. Count recurring themes or words and analyze the order of occurrence of words or phrases to identify specific patterns; S805. Calculate the frequency of occurrence of each topic or repeated word, and observe the changing trend of the topic over time.

7. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 1, characterized in that: The EEG data collection includes the following steps: S201, preparation stage: clean the subject's scalp, put the electrode cap on the subject's head, apply conductive paste, and use the EEG data acquisition software to set the sampling rate parameters; S202, acquisition phase: let the subject complete the corresponding task according to the instructions or keep a natural state, start recording the EEG signal, adjust the electrode position or re-apply the conductive paste when necessary to ensure the quality of the monitoring signal; S203, data processing: after stopping recording, save the EEG data, and perform data cleaning and filtering.

8. The obsessive-compulsive disorder diagnosis and symptom assessment system according to claim 7, characterized in that: The step S203 of data cleaning includes the following steps: S901, bad track detection, check whether there is a constant value or extremely high impedance in each channel; S902, eliminate anomalies, and remove data points or channels that are obviously beyond the normal range; S903, interpolation repair, uses interpolation technology to fill the gaps left after removing outliers.

9. A storage device, characterized in that: The storage device stores a plurality of instructions, and the instructions are used to execute the evaluation work of the evaluation system according to any one of claims 1-8.

10. An embedded device, characterized in that: include: A storage device for storing a computer program; A processor is used to execute the computer program to implement the evaluation work of the obsessive-compulsive disorder diagnosis and symptom severity evaluation system as described in claims 1-8.

Images