Health assessment method and system for rheumatoid arthritis synovial injury DBA / 1 mouse

Abstract

The invention relates to the field of data image processing, and discloses a health assessment method and system for rheumatoid arthritis synovial injury DBA / 1 mice, and the method comprises the steps: obtaining the image data of each labeled mouse in a preset time period; based on the image data, extracting an activity track, a health state and a recovery state of each marked mouse, and determining a treatment score of each marked mouse according to the activity track, the health state and the recovery state; acquiring a treatment score mean value among the marked mice, and screening out experimental mice with high treatment scores among the treatment scores based on the treatment score mean value; and acquiring a treatment mode and recovery data of the experimental mice with high treatment scores, determining a recommended treatment method for the experimental mice in the batch, and determining a recovery level of the recommended treatment method according to a relationship between the experimental mice with high treatment scores and preset treatment scores. According to the method, the optimal treatment scheme effect can be determined by regularly obtaining the treatment scores of the laboratory mice.

Description

Technical Field

[0001] This application relates to the field of data image processing. Specifically, it relates to a health assessment method and system for DBA / 1 mice with rheumatoid arthritis synovial injury. Background Art

[0002] In the study of rheumatoid arthritis (RA), DBA / 1 mice are a commonly used animal model to simulate the pathological characteristics of human RA for studying disease mechanisms and evaluating treatment regimens. Currently, the health assessment of RA synovial injury mice mainly relies on manual observation and subjective scoring, such as evaluating based on the walking posture, joint swelling degree, and behavioral changes of the mice. However, this method has large subjective errors and is difficult to achieve efficient and accurate quantitative analysis, thus affecting the reliability and repeatability of experimental results.

[0003] The traditional assessment methods mainly have the following defects: First, manual observation is easily affected by the experience and subjective judgment of the observer, resulting in inconsistent assessment results; second, manual scoring is difficult to capture the subtle changes in the behavior and physiological state of the mice, and may ignore important health indicators; third, manual assessment is time-consuming and laborious, difficult to meet the needs of large-scale experiments, and limits the research efficiency.

[0004] With the development of computer vision and data analysis technologies, health assessment methods based on image data have gradually become an important tool in biomedical research. Through automated data collection and analysis, manual intervention can be reduced and experimental efficiency can be improved. For example, using computer vision technology to extract the activity trajectories, behavioral patterns, and physiological states of experimental mice, and combining data processing algorithms to calculate health scores can more accurately reflect the treatment response and recovery of the mice. In addition, the analysis method based on big data can also identify the effects of different treatment methods on the health status of the mice, thereby optimizing the experimental design.

[0005] However, the existing health assessment methods based on image data still have some limitations. For example, the image preprocessing technology is not perfect enough, which may lead to unstable data quality; the accuracy of target detection and tracking algorithms in complex environments needs to be improved; the health status scoring model is too simple to comprehensively reflect the health status of the mice; there is a lack of quantitative assessment of treatment effects and optimization recommendation mechanisms, etc.

[0006] In view of the above problems, there is an urgent need for improvement in the existing technology. Summary of the Invention

[0007] In view of this, the present invention proposes a health assessment method and system for DBA / 1 mice with rheumatoid arthritis synovial injury, which has the advantages of improving the assessment accuracy, reducing subjective errors, and enhancing experimental efficiency.

[0008] The present application provides a health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury. The technical solution is as follows: It includes: marking each experimental mouse, and obtaining the image data of each marked mouse within a preset time period; extracting the activity trajectory, health status, and recovery status of each marked mouse based on the image data, and determining the treatment score of each marked mouse according to the activity trajectory, health status, and recovery status; obtaining the average treatment score among each marked mouse, and screening out the experimental mice with high treatment scores among each treatment score based on the average treatment score; obtaining the treatment methods and recovery data of the experimental mice with high treatment scores, determining the recommended treatment method for this batch of experimental mice, and determining the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score.

[0009] Further, the present application also proposes that when obtaining the image data of each marked mouse within a preset time period, it includes: performing image preprocessing on the obtained image data; generating a unique identification symbol for the marked mouse based on the target detection technology, and obtaining the image information of the marked mouse in each time frame of the preprocessed image data based on the identification symbol.

[0010] Further, the present application also proposes that when performing image preprocessing on the obtained image data, it includes: performing histogram equalization processing on the obtained image data; establishing an image segmentation model based on the Gaussian mixture model, and segmenting the image data according to the image segmentation; removing the interference background image data after segmentation, and merging according to the remaining image data; performing illumination normalization processing on the merged image data based on local adaptive histogram equalization.

[0011] Further, the present application also proposes that when establishing an image segmentation model based on the Gaussian mixture model and segmenting the image data according to the image segmentation, it includes:

[0012] Performing feature extraction on the obtained image data to obtain the color feature, spatial information, and texture feature of the pixel points;

[0013] Representing the image data as a Gaussian mixture distribution, and calculating the probability that the pixel point \(X_i\) belongs to different Gaussian components based on the following formula:

[0014]

[0015] where \(\gamma_{i k}\) represents the probability that the pixel point \(X_i\) belongs to the \(k\)th Gaussian component, \(\pi\) k is the mixing weight of the \(k\)th Gaussian component, is the Gaussian probability density function corresponding to this component;

[0016] The Huya Expectation-Maximization (EM) algorithm iteratively optimizes the GMM parameters and classifies each pixel based on the Maximum A Posteriori (MAP) rule to segment the foreground and background;

[0017] Morphological processing is performed on the segmentation result, including erosion, dilation, connected component analysis, and Gaussian smoothing, to remove noise and optimize the foreground object contour.

[0018] Furthermore, when extracting the activity trajectories, health status, and recovery status of each labeled mouse based on the image data, the present application includes: determining the activity trajectory and activity route of the labeled mouse within a preset period according to the labeled mouse image information; obtaining the movement direction, speed, and behavior pattern of the labeled mouse within the preset period based on the optical flow method, running trajectory, and activity route; and determining the health status and recovery status of the labeled mouse according to the movement direction, speed, and behavior model.

[0019] Furthermore, when determining the treatment score of each labeled mouse according to the activity trajectory, health status, and recovery status, the present application includes:

[0020] Determining the health status score of the labeled mouse according to the activity trajectory of the labeled mouse within the preset period;

[0021] Determining the recovery score according to the relationship between the health status score and the health status score of the adjacent historical period, and determining the treatment score of the labeled mouse according to the relationship between the recovery score and the health status score:

[0022] T S = α·H S + β·R S + γ·f(H S ,R S );

[0023] where H S is the health status score, R S is the recovery score, f(HS,RS) is a non-linear function between the recovery score and the health status score, α, β, and γ are weight coefficients, and the sum of α, β, and γ is 1.

[0024] Furthermore, the present application also proposes that when determining the health status score of the labeled mouse according to the activity trajectory of the labeled mouse within a preset time period, it includes: smoothing the mouse movement trajectory based on the optical flow method and Kalman filtering to calculate the speed change per unit time, and determining the movement ability of the labeled mouse based on the degree of deviation of the speed from the optimal movement speed; measuring the activity range of the mouse based on the trajectory coverage area calculation method, determining the smoothness of the movement trajectory of the labeled mouse based on the curvature change rate, and using an exponential decay function to perform a trajectory distortion score on the degree of trajectory distortion; obtaining the proportion of the mouse's stationary time, and evaluating the movement activity of the mouse based on the ratio of the stationary time to the total activity time, and determining the health status score of the labeled mouse based on weighted calculation according to the movement ability, activity range, trajectory distortion score, and movement activity.

[0025] Furthermore, the present application also proposes that when determining the recovery score according to the relationship between the health status score and the health status score of the adjacent historical period, it includes: obtaining the score difference between the health status score and the health status score of the adjacent historical period, and determining the recovery score according to the relationship between the score difference and the preset first preset score difference and second preset score difference: when the score difference is lower than the first preset score difference, then determine the recovery score as D1; when the score difference is higher than or equal to the first preset score difference and the score difference is lower than the second preset score difference, then determine the recovery score as D2; when the score difference is higher than or equal to the second preset score difference, then determine the recovery score as D3; where the first preset score difference is less than the second preset score difference, and D1 = 0, D1 < D2 < D3, Rs = D1, D2, and D3.

[0026] Furthermore, the present application also proposes that when determining the recovery level of the recommended treatment method according to the relationship between the high-treatment-score experimental mouse and the preset treatment score, it includes: obtaining the treatment score difference between the high-treatment-score experimental mouse and the preset treatment score, and determining the recovery level of the recommended treatment method according to the relationship between the treatment score difference and the pre-configured first preset treatment score difference and second preset treatment score difference; when the treatment score difference is lower than the first preset treatment score difference, then determine the recovery level of the recommended treatment method as the low level; when the treatment score difference is greater than or equal to the first preset treatment score difference and the treatment score difference is lower than the second preset treatment score difference, then determine the recovery level of the recommended treatment method as the medium level; when the treatment score difference is greater than or equal to the second preset treatment score difference, then determine the recovery level of the recommended treatment method as the high level; where the first preset treatment score difference is less than the second preset treatment score difference, and the recovery levels of the recommended treatment method are sorted from high to low as the low level, medium level, and high level.

[0027] Furthermore, the present application also proposes a system for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury, adopting the above-mentioned health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury, including: an image acquisition module configured to label each experimental mouse and obtain image data of each labeled mouse within a preset period; the image acquisition module is also configured to extract the activity trajectory, health status, and recovery status of each labeled mouse based on the image data, and determine the treatment score of each labeled mouse according to the activity trajectory, health status, and recovery status; a central control module electrically connected to the image acquisition module, the central control module is configured to obtain the average treatment score among each labeled mouse, and screen out the experimental mice with high treatment scores among the treatment scores based on the average treatment score; an output module electrically connected to the central control module, the output module is configured to obtain the treatment method and recovery data of the experimental mice with high treatment scores, determine the recommended treatment method for this batch of experimental mice, and determine the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score.

[0028] As can be seen from the above, a health assessment method and system for DBA / 1 mice with rheumatoid arthritis synovial injury provided by the present application include labeling each experimental mouse, obtaining image data, extracting the activity trajectory, health status, and recovery status, determining the treatment score, screening out the experimental mice with high treatment scores, and determining the recommended treatment method and recovery level. Through automated image processing and data analysis, an objective and accurate assessment of the health status of mice is achieved, reducing the subjective error of manual observation, improving the experimental efficiency, and having the advantages of improving the assessment accuracy, reducing the subjective error, and enhancing the experimental efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0030] Figure 1 is a flowchart of the health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury provided by an embodiment of the present invention;

[0031] Figure 2 is a functional block diagram of the system for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the embodiments.

[0033] In rheumatoid arthritis research, DBA / 1 mice are a commonly used animal model for simulating the pathological characteristics of human rheumatoid arthritis to study disease mechanisms and evaluate treatment regimens. Existing health assessment methods mainly rely on manual observation and subjective scoring, such as evaluating based on the walking posture, joint swelling degree, and behavioral changes of mice. However, this method has a large subjective error and is difficult to achieve efficient and accurate quantitative analysis, thus affecting the reliability and repeatability of experimental results.

[0034] The present invention proposes a health assessment method based on image data. By marking experimental mice and collecting image data, the activity trajectory, health status, and recovery status of the mice are extracted, solving the problem of how to accurately assess the health status of DBA / 1 mice with rheumatoid arthritis synovial injury. By analyzing the image data and extracting the activity trajectory, health status, and recovery status of the mice, the treatment response of the mice can be objectively reflected. Based on these data, the treatment score of the mice is determined, and mice with high treatment scores are screened out to further optimize the treatment regimen and evaluate the effectiveness of the treatment method. Finally, by comparing the relationship between the mice with high treatment scores and the preset treatment score, the recovery level of the recommended treatment method is determined, thereby providing a scientific data support, optimizing the treatment regimen, and improving the accuracy and objectivity of health assessment.

[0035] In the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury, there is a problem of how to improve the accuracy and objectivity of the assessment. The traditional manual observation method has subjective errors and is difficult to achieve efficient and accurate quantitative analysis. With the development of computer vision and data analysis technologies, the health assessment method based on image data has gradually become an important tool in biomedical research. By automating data collection and analysis, manual intervention can be reduced and experimental efficiency can be improved. For example, using computer vision technology to extract the activity trajectory, behavior pattern, and physiological state of experimental mice, and combining data processing algorithms to calculate the health score can more accurately reflect the treatment response and recovery of the mice. In addition, the analysis method based on big data can also identify the effects of different treatment methods on the health status of mice, thereby optimizing the experimental design.

[0036] Such asFigure 1 As shown in Figure 1 , the health assessment method of the present invention includes the following steps: marking each experimental mouse, and acquiring the image data of each marked mouse within a preset time period; extracting the activity trajectory, health status, and recovery status of each marked mouse based on the image data, and determining the treatment score of each marked mouse according to the activity trajectory, health status, and recovery status; obtaining the average treatment score among each marked mouse, and screening out the experimental mice with high treatment scores among each treatment score based on the average treatment score; obtaining the treatment methods and recovery data of the experimental mice with high treatment scores, determining the recommended treatment method for this batch of experimental mice, and determining the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score.

[0037] For example, by marking and collecting image data of experimental mice, extracting the activity trajectory, health status, and recovery status of the mice can objectively reflect the treatment response of the mice. According to these data, the treatment score of the mice is determined, and the mice with high treatment scores are screened out to further optimize the treatment plan and evaluate the effect of the treatment method. Finally, by comparing the relationship between the mice with high treatment scores and the preset treatment score, the recovery level of the recommended treatment method is determined, so as to provide a scientific data support, optimize the treatment plan, and improve the accuracy and objectivity of health assessment.

[0038] Among them, the acquisition and processing of image data are the key steps to realize health assessment. By performing image preprocessing on the acquired image data, including histogram equalization processing, establishing an image segmentation model with a Gaussian mixture model, removing interference background image data, and illumination normalization processing, the quality and accuracy of the image data can be improved. Then, by extracting the activity trajectory, health status, and recovery status of each marked mouse based on the image data, the treatment response and recovery situation of the mice can be reflected more accurately.

[0039] Furthermore, by calculating the treatment score and screening out the mice with high treatment scores, the treatment plan can be optimized and the effect of the treatment method can be evaluated. For example, according to the activity trajectory of the marked mouse within the preset time period, the health status score of the marked mouse is determined; according to the relationship between the health status score and the health status score in the adjacent historical time period, the recovery score is determined, and according to the relationship between the recovery score and the health status score, the treatment score of the marked mouse is determined. Finally, by comparing the relationship between the mice with high treatment scores and the preset treatment score, the recovery level of the recommended treatment method is determined, so as to provide a scientific data support, optimize the treatment plan, and improve the accuracy and objectivity of health assessment.

[0040] Compared with the prior art, the health assessment method of the present invention has the following advantages: First, through automated data collection and analysis, manual intervention can be reduced and the experimental efficiency can be improved; Second, through the health assessment method based on image data, the treatment response and recovery of mice can be more accurately reflected; Finally, by screening mice with high treatment scores, the treatment plan can be optimized and the effectiveness of the treatment method can be evaluated.

[0041] The present invention marks and collects image data of experimental mice, extracts the activity trajectories, health status and recovery status of the mice, and solves the problem of how to accurately evaluate the health status of DBA / 1 mice with rheumatoid arthritis synovial injury. By analyzing the image data and extracting the activity trajectories, health status and recovery status of the mice, the treatment response of the mice can be objectively reflected. According to these data, the treatment scores of the mice are determined, and the mice with high treatment scores are screened out to further optimize the treatment plan and evaluate the effectiveness of the treatment method. Finally, by comparing the relationship between the mice with high treatment scores and the preset treatment scores, the recovery level of the recommended treatment method is determined, so as to provide a scientific data support, optimize the treatment plan, and improve the accuracy and objectivity of health assessment.

[0042] Furthermore, the present application also proposes to obtain the image data of each marked mouse within a preset time period, including performing image preprocessing on the obtained image data, generating a unique identification symbol for the marked mouse based on the target detection technology, and obtaining the image information of the marked mouse in each time frame of the preprocessed image data based on the identification symbol.

[0043] The present application includes performing image preprocessing on the obtained image data, generating a unique identification symbol for the marked mouse based on the target detection technology, and obtaining the image information of the marked mouse in each time frame of the preprocessed image data based on the identification symbol. Image preprocessing can improve the quality of the image data, reduce noise and interference, and generating a unique identification symbol can ensure that each mouse can be accurately identified and tracked throughout the experiment. By combining these technical features, the image data of the marked mouse can be efficiently obtained and processed, thus providing a reliable data basis for the subsequent extraction of activity trajectories, health status and recovery status.

[0044] The technical features of image preprocessing include performing histogram equalization on the image data, establishing an image segmentation model based on the Gaussian mixture model, segmenting the image data according to the image segmentation, removing the interference background image data after segmentation, and merging based on the remaining image data, and performing illumination normalization on the merged image data based on local adaptive histogram equalization. For example, histogram equalization processing can enhance the contrast of the image, making the details in the image more obvious. The Gaussian mixture model is used for image segmentation, which can effectively distinguish the foreground and background, thereby improving the accuracy of target detection. Local adaptive histogram equalization can further improve the illumination uniformity of the image, making subsequent image processing more stable and reliable.

[0045] The technical features of generating unique identification symbols include using target detection technology to identify the labeled mice and generating a unique identification symbol for each mouse. The target detection technology can be based on deep learning algorithms, such as convolutional neural networks (CNNs), by training a model to identify the mice in the image and generate the corresponding identification symbols. This process can ensure that each mouse can be accurately identified and tracked throughout the experiment.

[0046] The technical features of obtaining the labeled mouse image information in each time frame of the preprocessed image data based on the identification symbol include performing frame-by-frame analysis on the preprocessed image data and extracting the labeled mouse image information in each frame. By analyzing each frame of the image, the position information and behavioral characteristics of the labeled mice at different time points can be obtained. This information can be used for subsequent analysis of the activity trajectories, health status, and recovery status.

[0047] Furthermore, the present application improves the quality of the image data and the accuracy of labeled mouse identification through the combination of image preprocessing and target detection technology. Compared with the prior art, the method of the present application reduces manual intervention, improves the experimental efficiency and the reliability of the data, and provides scientific data support for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury.

[0048] Furthermore, the present application also proposes that when performing image preprocessing on the obtained image data, it includes: performing histogram equalization on the obtained image data; establishing an image segmentation model based on the Gaussian mixture model and segmenting the image data according to the image segmentation; removing the interference background image data after segmentation and merging based on the remaining image data; performing illumination normalization on the merged image data based on local adaptive histogram equalization.

[0049] This technical solution includes performing histogram equalization on the image data to enhance the image contrast; using a Gaussian mixture model for image segmentation to accurately separate the foreground and background; removing the interfering background image data after segmentation, and merging the remaining valid image data; finally, using local adaptive histogram equalization for illumination normalization to solve the problem of uneven illumination. Through these steps, the interfering background in the image can be effectively removed and the illumination can be balanced, thereby improving the quality of the image data and the accuracy of subsequent analysis.

[0050] Histogram equalization is a commonly used image enhancement technique that improves the contrast of an image by adjusting the gray-level distribution of the image. The Gaussian mixture model is a probability model that can be effectively used for image segmentation. By iteratively optimizing the model parameters through the expectation-maximization algorithm, the separation of the foreground and background can be achieved. Local adaptive histogram equalization is an image processing method for the problem of uneven illumination. By performing histogram equalization in local regions, it can better adapt to the illumination changes in the image.

[0051] Specifically, during image preprocessing, first perform histogram equalization on the image data to enhance the overall contrast of the image. Then, based on the Gaussian mixture model, establish an image segmentation model and perform image segmentation by calculating the probability of each pixel belonging to different Gaussian components. After segmentation, remove the interfering background image data and merge the remaining valid image data. Finally, use local adaptive histogram equalization to perform illumination normalization on the merged image data to solve the problem of uneven illumination.

[0052] Through the above technical solution, this application can effectively solve the problems of interfering background and uneven illumination existing in the process of image data processing compared with the prior art, thereby improving the quality of the image data and the accuracy of subsequent analysis. Especially through image segmentation using the Gaussian mixture model and local adaptive histogram equalization processing, the effect of image preprocessing is more significant, further improving the reliability and accuracy of image analysis.

[0053] Furthermore, the present application also proposes that when establishing an image segmentation model based on a Gaussian mixture model and segmenting image data according to the image segmentation, it includes: extracting features from the acquired image data to obtain the color features, spatial information, and texture features of pixel points; representing the image data as a Gaussian mixture distribution, and calculating the probabilities of pixel points Xi belonging to different Gaussian components based on the following formula; where γi k represents the probability that pixel point Xi belongs to the k-th Gaussian component, πk is the mixing weight of the k-th Gaussian component, and is the Gaussian probability density function corresponding to this component; the Expectation-Maximization (EM) algorithm is used to iteratively optimize the GMM parameters, and based on the Maximum A Posteriori (MAP) rule, each pixel point is classified to segment the foreground and background; morphological processing is performed on the segmentation result, including erosion, dilation, connected component analysis, and Gaussian smoothing, to remove noise and optimize the foreground object contour.

[0054] The technical features included in the present application are: extracting features from image data to obtain the color features, spatial information, and texture features of pixel points; representing the image data as a Gaussian mixture distribution and calculating the probabilities of pixel points belonging to different Gaussian components; using the Expectation-Maximization (EM) algorithm to iteratively optimize the Gaussian mixture model (GMM) parameters and classifying each pixel point based on the Maximum A Posteriori (MAP) rule; performing morphological processing on the segmentation result, including erosion, dilation, connected component analysis, and Gaussian smoothing, to remove noise and optimize the foreground object contour. These technical features, through the application of feature extraction and Gaussian mixture model, combined with the Expectation-Maximization algorithm and morphological processing, achieve accurate segmentation of the foreground and background in image data, thereby effectively extracting the activity trajectories, health status, and recovery status of mice, and solving the problem of inaccurate segmentation of the foreground and background in image data.

[0055] When extracting features from the acquired image data, various image processing techniques can be used. For example, color features can be extracted through color histograms or color space conversion; spatial information can be obtained through image gradients or edge detection algorithms; texture features can be extracted through methods such as gray-level co-occurrence matrices or Local Binary Pattern (LBP). When representing the image data as a Gaussian mixture distribution, a Gaussian mixture model (GMM) can be used, which can effectively represent the statistical characteristics of different regions in the image. By using the Expectation-Maximization (EM) algorithm to iteratively optimize the GMM parameters, the accuracy of the model can be improved. Classifying each pixel point based on the Maximum A Posteriori (MAP) rule can achieve accurate segmentation of the foreground and background. When performing morphological processing on the segmentation result, erosion and dilation operations can be used to remove noise and fill holes, connected component analysis can be used to extract connected foreground regions, and Gaussian smoothing can further optimize the contour of the foreground object.

[0056] Through the image segmentation method based on the Gaussian mixture model, combined with the expectation maximization algorithm and morphological processing, the accurate segmentation of the foreground and background in the image data is achieved. Compared with the prior art, the method of this application can extract the activity trajectories, health status, and recovery status of mice more accurately, improve the accuracy and robustness of image segmentation, solve the problem of inaccurate foreground and background segmentation, and has significant technical advantages.

[0057] Furthermore, this application also proposes that when extracting the activity trajectories, health status, and recovery status of each labeled mouse based on the image data, it includes: determining the activity trajectory and activity route of the labeled mouse within a preset time period according to the labeled mouse image information; obtaining the movement direction, speed, and behavior pattern of the labeled mouse within the preset time period based on the optical flow method, running trajectory, and activity route; and determining the health status and recovery status of the labeled mouse according to the movement direction, speed, and behavior model.

[0058] The technical solution for extracting the activity trajectories, health status, and recovery status of each labeled mouse based on the image data includes all the features of the preamble part and the feature part. The preamble part includes extracting the activity trajectories, health status, and recovery status of each labeled mouse based on the image data. The feature part includes determining the activity trajectory and activity route of the labeled mouse within a preset time period according to the labeled mouse image information; obtaining the movement direction, speed, and behavior pattern of the labeled mouse within the preset time period based on the optical flow method, running trajectory, and activity route; and determining the health status and recovery status of the labeled mouse according to the movement direction, speed, and behavior model. These technical features cooperate with each other to solve the problem of how to accurately extract the activity trajectories, health status, and recovery status of the labeled mouse based on the image data.

[0059] Through the above solution, the problem of extracting the activity trajectories, health status, and recovery status of each labeled mouse based on the image data is solved. First, the activity trajectory and activity route of the mouse are determined through the image information; secondly, the movement direction, speed, and behavior pattern of the mouse are obtained by using the optical flow method and running trajectory; finally, the health status and recovery status of the mouse are determined according to these data. This solution effectively improves the accuracy and objectivity of health assessment.

[0060] Specifically, based on the marked mouse image information, the unique mark of the mouse can be identified through image processing technology, so as to track the movement trajectory and movement route of the mouse within a preset time period. The optical flow method is a common computer vision technology that can be used to detect motion information in images. Through the optical flow method, the movement direction and speed of the mouse in different time frames can be calculated. Combining the running trajectory and movement route of the mouse, the behavior patterns of the mouse, such as activity frequency, movement regularity, etc., can be further analyzed. Based on the data of these movement directions, speeds and behavior patterns, a health status model of the mouse can be established, and by comparing the movement characteristics in different states, the health status and recovery status of the mouse can be determined.

[0061] For example, as a preferred implementation, the optical flow method can be used to calculate the motion vectors of the mouse in consecutive image frames, and the Kalman filtering method can be combined to smooth the motion trajectory. By analyzing the change trend of the motion vectors, the motion ability and behavior characteristics of the mouse can be judged. In addition, by calculating the activity range and trajectory coverage area of the mouse, its activity level and trajectory smoothness can be evaluated. Combining these data, a comprehensive health status scoring system can be established, and by comparing the score changes in different time periods, the recovery status of the mouse can be determined.

[0062] It can be seen that the technical solution proposed in this application realizes an objective evaluation of the health status and recovery status of the marked mouse through precise image processing and data analysis methods. Compared with the traditional manual observation and subjective scoring methods, this image data-based evaluation method has higher accuracy and repeatability, can provide scientific data support for experiments, and helps to optimize and develop RA treatment methods.

[0063] Furthermore, this application also proposes that when determining the treatment score of each marked mouse according to the activity trajectory, health status and recovery status, it includes:

[0064] Determine the health status score of the marked mouse according to the activity trajectory of the marked mouse within a preset time period; determine the recovery score according to the relationship between the health status score and the health status score of the adjacent historical period, and determine the treatment score of the marked mouse according to the relationship between the recovery score and the health status score:

[0065] T S =α·H S +β·R S +γ·f(H S ,R S ); where HS is the health status score, RS is the recovery score, f(HS,RS) is the non-linear function between the recovery score and the health status score, α, β and γ are weight coefficients, and the sum of α, β and γ is 1.

[0066] This technical solution first analyzes the activity trajectory of the labeled mice within a preset time period, quantifies their health status, and gives corresponding health status scores. Then, by comparing the relationship between the current health status score and the health status scores of adjacent historical time periods, the recovery score of the mice is determined. Finally, combining the health status score and the recovery score, the treatment score of the labeled mice is determined. This method realizes the objective and quantitative evaluation of the health status and treatment effect of mice.

[0067] Specifically, according to the activity trajectory of the labeled mice within a preset time period, their health status scores are determined. The activity trajectory can be obtained through image data analysis, including parameters such as the movement direction, speed, and behavior pattern of the mice. Through these parameters, the movement ability, activity range, trajectory smoothness, and movement activity of the mice can be evaluated, and then their health status can be quantified.

[0068] Next, according to the relationship between the current health status score and the health status scores of adjacent historical time periods, the recovery score is determined. The recovery score reflects the health changes of the mice at different time periods and can be calculated through the score difference. When the score difference is lower than the preset threshold, the recovery score is lower; otherwise, it is higher.

[0069] Finally, combining the health status score and the recovery score, the treatment score of the labeled mice is determined. The treatment score combines the current health status and recovery of the mice and provides a comprehensive evaluation index. In this way, the treatment effect of the mice can be more accurately reflected, and scientific data support can be provided for the experiment.

[0070] The advantage of this method is that through automated data analysis and processing, it realizes the objective and quantitative evaluation of the health status and treatment effect of mice, reduces manual intervention and subjective errors, and improves the accuracy and reliability of the evaluation. Compared with the existing technologies, this method can more efficiently evaluate the health status and treatment effect of mice, providing important technical support for the research on DBA / 1 mice with rheumatoid arthritis synovial injury.

[0071] Furthermore, when determining the health status score of the labeled mouse based on its activity trajectory within a preset time period, the present application also proposes the following steps: smoothing the mouse's movement trajectory based on the optical flow method and Kalman filtering to calculate the speed change per unit time, and determining the movement ability of the labeled mouse based on the degree of deviation of the speed from the optimal movement speed; measuring the activity range of the mouse based on the trajectory coverage area calculation method, determining the smoothness of the movement trajectory of the labeled mouse based on the curvature change rate, and using an exponential decay function to score the degree of trajectory distortion; obtaining the proportion of the mouse's stationary time, and evaluating the movement activity of the mouse based on the ratio of the stationary time to the total activity time. Based on the movement ability, activity range, trajectory distortion score, and movement activity, the health status score of the labeled mouse is determined based on weighted calculation.

[0072] By using the optical flow method and Kalman filtering to smooth the mouse's movement trajectory, the speed change per unit time of the mouse can be accurately calculated, thereby determining its movement ability. The trajectory coverage area calculation method can measure the activity range of the mouse, while the curvature change rate is used to evaluate the smoothness of the movement trajectory. The exponential decay function scores the degree of trajectory distortion, and the proportion of stationary time is used to evaluate the movement activity of the mouse. By comprehensively considering these indicators and based on weighted calculation, the health status score of the labeled mouse can be determined more accurately. These technical features cooperate with each other to objectively and quantitatively evaluate the health status of DBA / 1 mice with rheumatoid arthritis synovial injury, thus solving the problem of evaluating the health status of mice in an objective and quantitative manner.

[0073] The optical flow method is a method for calculating the movement of pixels in an image sequence, which estimates movement information by analyzing the pixel intensity changes between consecutive frames of images. The Kalman filter is a recursive algorithm that optimizes the estimation of the movement trajectory by combining the current measurement value and the previous state estimate, reducing the influence of noise. The trajectory coverage area calculation method can evaluate the activity range of the mouse by calculating the area of the activity region of the mouse within a preset time period. The curvature change rate is used to evaluate the smoothness of the trajectory, reflecting the tortuous degree of the mouse's movement trajectory. The exponential decay function scores the degree of trajectory distortion, quantifying the degree of trajectory distortion. The proportion of stationary time evaluates the movement activity of the mouse by calculating the ratio of the mouse's stationary time to the total activity time.

[0074] Compared with the prior art, the present application provides a more objective and quantitative evaluation method through a combination of various technical means, which can more accurately reflect the health status of DBA / 1 mice with rheumatoid arthritis synovial injury. By comprehensively evaluating the motor ability, range of motion, trajectory smoothness, and motor activity, the health status of the mice can be more comprehensively understood, and the reliability and repeatability of the experimental results can be improved. This method not only improves the accuracy and objectivity of health evaluation but also provides scientific data support for the research of rheumatoid arthritis treatment methods.

[0075] Furthermore, the present application also proposes to obtain the score difference between the health status score and the health status score of the adjacent historical period, and determine the recovery score according to the relationship between the score difference and the preset first preset score difference and second preset score difference: when the score difference is lower than the first preset score difference, the recovery score is determined to be D1; when the score difference is higher than or equal to the first preset score difference and lower than the second preset score difference, the recovery score is determined to be D2; when the score difference is higher than or equal to the second preset score difference, the recovery score is determined to be D3; wherein, the first preset score difference is less than the second preset score difference, and D1 = 0, D1 < D2 < D3, RS = D1, D2 or D3.

[0076] The technical feature includes obtaining the score difference between the health status score and the health status score of the adjacent historical period, and determining the recovery score according to the relationship between the score difference and the preset first preset score difference and second preset score difference. Specifically, when the score difference is lower than the first preset score difference, the recovery score is D1; when the score difference is higher than or equal to the first preset score difference and lower than the second preset score difference, the recovery score is D2; when the score difference is higher than or equal to the second preset score difference, the recovery score is D3. The first preset score difference is less than the second preset score difference, and D1 = 0, D1 < D2 < D3. Through the above technical solution, the recovery score can be clearly determined according to different ranges of the score difference, thus solving the technical problem of how to determine the recovery score based on the relationship between the health status score and the health status score of the adjacent historical period.

[0077] Further, the scoring difference can be obtained in the following manner: First, obtain the health status score HS1 for the current period and the health status score HS2 for an adjacent historical period; then, calculate the difference between HS1 and HS2 to obtain the scoring difference. The calculation of the scoring difference can be achieved through a simple subtraction operation. For example, the scoring difference = HS1 - HS2. Specifically, the preset first preset scoring difference and second preset scoring difference can be set according to the actual situation. For example, the first preset scoring difference can be set to 5 points, and the second preset scoring difference can be set to 10 points. Thus, when the scoring difference is less than 5 points, the restored score is D1; when the scoring difference is between 5 points and 10 points, the restored score is D2; when the scoring difference is greater than or equal to 10 points, the restored score is D3.

[0078] This application can achieve a quantitative assessment of the recovery status by obtaining the scoring difference between the health status score and the health status score of an adjacent historical period, and determining the restored score based on the relationship between the scoring difference and the preset first preset scoring difference and second preset scoring difference. Compared with the prior art, the advantage of this application is that it can more accurately reflect the changes in the health status, thereby improving the accuracy and reliability of health assessment.

[0079] Further, this application also proposes that when determining the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score, it includes:

[0080] Obtain the treatment score difference between the experimental mice with high treatment scores and the preset treatment score, and determine the recovery level of the recommended treatment method based on the relationship between the treatment score difference and the pre-configured first preset treatment score difference and second preset treatment score difference; when the treatment score difference is less than the first preset treatment score difference, then determine that the recovery level of the recommended treatment method is the low level; when the treatment score difference is greater than or equal to the first preset treatment score difference and the treatment score difference is less than the second preset treatment score difference, then determine that the recovery level of the recommended treatment method is the medium level; when the treatment score difference is greater than or equal to the second preset treatment score difference, then determine that the recovery level of the recommended treatment method is the high level. Among them, the first preset treatment score difference is less than the second preset treatment score difference, and the recovery levels of the recommended treatment method are sorted from high to low as the low level, medium level, and high level.

[0081] This application includes all the features of the preamble part and the characterizing part. The feature of the preamble part is to determine the recovery level of the recommended treatment method according to the relationship between the high-treatment-score experimental mice and the preset treatment score. The features of the characterizing part include obtaining the treatment score difference and determining the recovery level according to the relationship between the difference and two preset score differences. Specifically, when the treatment score difference is lower than the first preset treatment score difference, the recovery level is the low level; when the treatment score difference is between the first and second preset treatment score differences, the recovery level is the medium level; when the treatment score difference is greater than or equal to the second preset treatment score difference, the recovery level is the high level. The first preset treatment score difference is less than the second preset treatment score difference, and the recovery levels from low to high are the low level, the medium level, and the high level in sequence.

[0082] This technical solution determines the recovery level of the recommended treatment method by obtaining the treatment score difference between the high-treatment-score experimental mice and the preset treatment score and according to the relationship between the score difference and the preset value. This method solves the problem of how to quantify and classify the effects of different treatment methods, making the recovery level of the recommended treatment method more scientific and objective.

[0083] Furthermore, obtaining the treatment score difference can be achieved by comparing the treatment score of the high-treatment-score experimental mice with the preset treatment score. Specifically, the treatment score difference can be expressed as the result of subtracting the preset treatment score from the treatment score of the high-treatment-score experimental mice. The preset treatment score can be determined in advance according to the experimental design and the expected treatment effect.

[0084] For example, the first preset treatment score difference and the second preset treatment score difference can be configured according to historical data and experimental results to ensure the rationality and scientificity of the score difference. Thus, by comparing the relationship between the treatment score difference and the preset treatment score difference, the recovery level of the recommended treatment method can be accurately determined.

[0085] As a preferred implementation manner, an automated data processing system can be used to calculate the treatment score difference and determine the recovery level. The system can include a data acquisition module, a data processing module, and a result output module. The data acquisition module is responsible for obtaining the treatment score of the high-treatment-score experimental mice and the preset treatment score; the data processing module is responsible for calculating the treatment score difference and determining the recovery level according to the preset score difference relationship; the result output module is responsible for outputting the recovery level of the recommended treatment method.

[0086] This application can effectively improve the scientificity and reliability of treatment method recommendation through precise score difference calculation and reasonable recovery level division, avoiding the errors caused by subjective judgment. Compared with the prior art, this application provides a more objective and quantitative treatment effect evaluation method, which can provide more valuable reference data for experiments and clinical practice.

[0087] In another preferred embodiment based on the above embodiments, as Figure 2 shown, this embodiment provides a system for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury, adopting a health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury, including an image acquisition module, a central control module, and an output module. The image acquisition module is configured to label each experimental mouse and obtain the image data of each labeled mouse within a preset time period; the image acquisition module is further configured to extract the activity trajectory, health status, and recovery status of each labeled mouse based on the image data, and determine the treatment score of each labeled mouse according to the activity trajectory, health status, and recovery status. The central control module is electrically connected to the image acquisition module, and the central control module is configured to obtain the average treatment score among each labeled mouse and screen out the experimental mice with high treatment scores among each treatment score based on the average treatment score. The output module is electrically connected to the central control module, and the output module is configured to obtain the treatment method and recovery data of the experimental mice with high treatment scores, determine the recommended treatment method for this batch of experimental mice, and determine the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score.

[0088] The image acquisition module extracts the activity trajectory, health status, and recovery status of the mice and determines the treatment score by labeling the experimental mice and obtaining the image data within a preset time period. The central control module screens out the experimental mice with high treatment scores by obtaining the average treatment score among each labeled mouse. The output module determines the recommended treatment method by obtaining the treatment method and recovery data of the experimental mice with high treatment scores, and determines the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score. Through the above technical means, the system can realize efficient and accurate quantitative analysis of the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury, reduce subjective errors, and improve the reliability and repeatability of experimental results.

[0089] The implementation manner of the image acquisition module may include using a high-resolution camera to monitor the experimental mice in real time and adopting an image processing algorithm to analyze the acquired image data. The central control module can use a high-performance computer for data processing and analysis to ensure that the treatment score can be calculated quickly and accurately and the experimental mice with high treatment scores can be screened out. The output module can output the recommended treatment method and recovery level through a display screen or a printer, or transmit the data to a remote server through a network interface for further analysis.

[0090] The system of the present application realizes efficient and accurate quantitative analysis of the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury through automated image acquisition and data analysis. Compared with the prior art, the system of the present application reduces subjective errors and improves the reliability and repeatability of experimental results. In addition, by screening experimental mice with high treatment scores, the system can optimize the recommended treatment methods and further improve the scientificity and effectiveness of the experiment.

[0091] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0092] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0093] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific implementation manners of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A health assessment method for DBA / 1 mice with synovial injury in rheumatoid arthritis, characterized in that Including: Mark each experimental mouse and obtain the image data of each marked mouse within a preset time period; Extract the activity trajectories, health status, and recovery status of each marked mouse based on the image data, and determine the treatment scores of each marked mouse according to the activity trajectories, health status, and recovery status; Obtain the average treatment scores among each marked mouse, and screen out the experimental mice with high treatment scores among the treatment scores based on the average treatment scores; Obtain the treatment methods and recovery data of the experimental mice with high treatment scores, determine the recommended treatment methods for this batch of experimental mice, and determine the recovery level of the recommended treatment methods according to the relationship between the experimental mice with high treatment scores and the preset treatment scores.

2. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury as described in claim 1, characterized in that, When obtaining the image data of each marked mouse within a preset time period, it includes: Perform image preprocessing on the obtained image data; Generate a unique identification symbol for the marked mouse based on the target detection technology, and obtain the image information of the marked mouse in each time frame of the preprocessed image data based on the identification symbol.

3. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury as described in claim 2, characterized in that, When performing image preprocessing on the obtained image data, it includes: Perform histogram equalization processing on the obtained image data; Establish an image segmentation model based on the Gaussian mixture model, and segment the image data according to the image segmentation; Eliminate the interference background image data after segmentation, and merge according to the remaining image data; Perform illumination normalization processing on the merged image data based on local adaptive histogram equalization.

4. The health assessment method for DBA / 1 mice with synovial injury of rheumatoid arthritis according to claim 3, characterized in that, When establishing an image segmentation model based on the Gaussian mixture model and segmenting the image data according to the image segmentation, it includes: Extract features from the obtained image data to obtain the color features, spatial information, and texture features of the pixel points; Represent the image data as a Gaussian mixture distribution, and calculate the probability that the pixel point \(X_i\) belongs to different Gaussian components based on the following formula: where γik represents the probability that pixel point Xi belongs to the k-th Gaussian component, and π k is the mixing weight of the k-th Gaussian component, is the Gaussian probability density function corresponding to this component; The Expectation-Maximization (EM) algorithm of Huya iteratively optimizes the GMM parameters, and classifies each pixel point based on the Maximum A Posteriori (MAP) rule to segment the foreground and background; Perform morphological processing on the segmentation result, including erosion, dilation, connected component analysis, and Gaussian smoothing, to remove noise and optimize the foreground object contour.

5. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury according to claim 4, characterized in that When extracting the activity trajectories, health status, and recovery status of each marked mouse based on the image data, it includes: Determine the activity trajectory and activity route of the marked mouse within the preset time period according to the marked mouse image information; Obtain the movement direction, speed, and behavior pattern of the marked mouse within the preset time period based on the optical flow method, running trajectory, and activity route; Determine the health status and recovery status of the marked mouse according to the movement direction, speed, and behavior model.

6. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury according to claim 5, characterized in that, When determining the treatment scores of each marked mouse according to the activity trajectories, health status, and recovery status, it includes: Determine the health status score of the marked mouse according to the activity trajectory of the marked mouse within the preset time period; Determine the recovery score according to the relationship between the health status score and the health status score of the adjacent historical time period, and determine the treatment score of the marked mouse according to the relationship between the recovery score and the health status score: T S = α·H S + β·R S + γ·f(H S , R S ); Among them, H S is the health status score, R S is the recovery score, f(HS,RS) is the non-linear function between the recovery score and the health status score, α, β and γ are weight coefficients, and the sum of α, β and γ is 1.

7. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury according to claim 6, wherein When determining the health status score of the marked mouse according to the activity trajectory of the marked mouse within the preset time period, it includes: Smoothing the movement trajectories of mice based on optical flow method and Kalman filtering to calculate the speed change within a unit time, and determining the movement ability of the marked mice based on the degree of deviation of the speed from the optimal movement speed; Measuring the activity range of mice based on the trajectory coverage area calculation method, determining the smoothness of the movement trajectories of the marked mice based on the curvature change rate, and using an exponential decay function to perform a trajectory distortion score on the degree of trajectory distortion; Obtaining the proportion of the static time of the mice, and evaluating the movement activity of the mice based on the ratio of the static time to the total activity time; Based on the movement ability, activity range, trajectory distortion score, and movement activity, determining the health status score of the marked mice based on weighted calculation; 8. The health assessment method for DBA / 1 mice with synovial injury of rheumatoid arthritis according to claim 7, wherein When determining the recovery score according to the relationship between the health status score and the health status score of the adjacent historical period, it includes: Obtaining the score difference between the health status score and the health status score of the adjacent historical period, and determining the recovery score according to the relationship between the score difference and the preset first preset score difference and the second preset score difference; When the score difference is lower than the first preset score difference, the recovery score is determined to be D1; When the score difference is higher than or equal to the first preset score difference and lower than the second preset score difference, the recovery score is determined to be D2; When the score difference is higher than or equal to the second preset score difference, the recovery score is determined to be D3; Wherein, the first preset score difference is less than the second preset score difference, and D1 = 0, D1 < D2 < D3; 9. The health assessment method for DBA / 1 mice with rheumatoid arthritis synovial injury according to claim 1, wherein When determining the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment score, it includes: Obtaining the treatment score difference between the experimental mice with high treatment scores and the preset treatment score, and determining the recovery level of the recommended treatment method according to the relationship between the treatment score difference and the pre-configured first preset treatment score difference and the second preset treatment score difference; When the treatment score difference is lower than the first preset treatment score difference, the recovery level of the recommended treatment method is determined to be the low level; When the treatment score difference is greater than or equal to the first preset treatment score difference and lower than the second preset treatment score difference, the recovery level of the recommended treatment method is determined to be the medium level; When the treatment score difference is greater than or equal to the second preset treatment score difference, the recovery level of the recommended treatment method is determined to be the high level; Wherein, the first preset treatment score difference is less than the second preset treatment score difference, and the recovery levels of the recommended treatment method are sorted from high to low as the low level, the medium level, and the high level; 10. A system for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury, which adopts the method for the health assessment of DBA / 1 mice with rheumatoid arthritis synovial injury as described in any one of claims 1-9, characterized in that, It includes: An image acquisition module configured to mark each experimental mouse and obtain the image data of each marked mouse within a preset period; The image acquisition module is further configured to extract the activity trajectories, health status, and recovery status of each marked mouse based on the image data, and determine the treatment score of each marked mouse according to the activity trajectories, health status, and recovery status; The central control module is electrically connected to the image acquisition module, and the central control module is configured to obtain the average treatment score among the marked mice, and screen out the experimental mice with high treatment scores among the treatment scores based on the average treatment score; The output module is electrically connected to the central control module, and the output module is configured to obtain the treatment methods and recovery data of the experimental mice with high treatment scores, determine the recommended treatment method for the batch of experimental mice, and determine the recovery level of the recommended treatment method according to the relationship between the experimental mice with high treatment scores and the preset treatment scores.

Images