A fully automatic abdominal withdrawal reflex scoring method and device

Abstract

The present invention provides a fully automatic abdominal withdrawal reflex scoring method and device, including: collecting real-time image data of experimental animals and preprocessing to obtain raw data images, automatically judging whether the current state of the experimental animals is a frequently active state or a resting state according to the raw data images; if it is a resting state, controlling a micropump to slowly inject a set dose of liquid into a balloon, and collecting real-time image data of the experimental animals in a resting state at the same time; preprocessing the collected image data of the experimental animals in a resting state to obtain a number of raw data images; and automatically scoring the abdominal withdrawal reflex of the experimental animals according to the raw data images. The present invention avoids the problems of traditional manual operation and strong subjectivity of manual scoring by automatically timing, automatically injecting, and automatically scoring multiple experimental animals, thereby improving the efficiency of the experiment and the accuracy of the scoring.

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to a fully automatic abdominal wall withdrawal reflex scoring method and device. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Abdominal withdrawal reflex (AWR) scoring is the main method to evaluate whether experimental animals (rats or mice) have visceral hypersensitivity. Existing AWR scoring generally uses manual scoring methods, including artificial water injection and human visual judgment of the movement state of experimental animals. The scoring process is complicated and time-consuming. Most importantly, manual scoring is highly subjective, which seriously affects the accuracy of the scoring and makes it impossible to perform objective and simple AWR scoring. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a fully automatic abdominal withdrawal reflex scoring method and device, which avoids the problems of traditional manual operation and strong subjectivity of manual scoring through automatic timing, automatic injection and automatic scoring of multiple experimental animals, improves the efficiency of the experiment and improves the accuracy of scoring.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a fully automatic abdominal withdrawal reflex scoring method, comprising:

[0007] Inserting the balloon into the body of the experimental animal and collecting real-time image data of the experimental animal after a set period of time;

[0008] Preprocessing the collected image data to obtain several original data images;

[0009] Automatically determine whether the current state of the experimental animal is a frequently active state or a resting state based on the original data image;

[0010] If the animal is in a resting state, the micropump is controlled to slowly inject a set dose of liquid into the balloon, and real-time imaging data of the experimental animal in a resting state is collected at the same time;

[0011] Preprocessing the collected image data of experimental animals in a resting state to obtain a number of raw data images;

[0012] The abdominal withdrawal reflex of experimental animals was automatically scored based on the raw data images.

[0013] A further technical solution is that if the activity is frequent, an alarm is activated to recheck whether the balloon is inserted correctly.

[0014] A further technical solution is that if the activity is frequent, a re-judgment is made after a set time. If the number of re-judgments exceeds the set number, an alarm is activated to re-check whether the balloon is inserted correctly.

[0015] A further technical solution, the automatic scoring of the abdominal wall withdrawal reflex of the experimental animal based on the original data image, comprises:

[0016] Input the original data image into the trained neural network model to realize the target recognition and feature point labeling of the experimental animals in the image;

[0017] The motion parameters of the experimental animal in the image are determined based on the feature points, and the motion parameters of the experimental animal in the image are compared with the motion parameters of the experimental animal in the initial frame image of the resting state. The motion state of the experimental animal is determined based on whether the variation range of the motion parameters exceeds the set threshold. The abdominal wall withdrawal reflex of the experimental animal is scored based on the motion state combined with the scoring rules.

[0018] In a further technical solution, the training process of the neural network model includes:

[0019] Collect real-time imaging data of experimental animals;

[0020] Preprocessing the collected image data to obtain several original data images;

[0021] Manually annotate the feature points of experimental animals in the images;

[0022] The original data image with feature point annotations is input into the neural network for prediction and decision making, and the neural network is trained to obtain a trained neural network model.

[0023] A further technical solution is that after the automatic scoring of the abdominal withdrawal reflex of the experimental animal is completed, the micropump is controlled to slowly release the liquid in the balloon, and the automatic scoring of the abdominal withdrawal reflex is re-performed under different liquid doses.

[0024] A further technical solution, determining the motion parameters of the experimental animal in the image based on the feature points, includes: connecting the feature points of the experimental animal in the image to determine the four motion parameters of the experimental animal: the elevation angle α, the arch angle β, the distance l between the pelvis and the foot, and the abdominal muscle contraction angle γ.

[0025] A further technical solution is to determine the motion state of the experimental animal according to whether the range of change of the motion parameter exceeds the set threshold, and to complete the scoring of the abdominal wall withdrawal reflex of the experimental animal according to the motion state combined with the scoring rules, including:

[0026] If the change range of the four motion parameters did not exceed the set threshold, the experimental animal was judged as having no reaction and scored 1 point;

[0027] If the variation range of the elevation angle α exceeds the set threshold, the experimental animal is judged to have head movement and scored 2 points;

[0028] If the change range of the abdominal muscle contraction angle γ exceeds the set threshold, the experimental animal is judged to have abdominal muscle contraction and scored 3 points;

[0029] If the variation range of the arch angle β exceeds the set threshold, the experimental animal is judged to have arched body and is scored 4 points;

[0030] If the variation range of the distance l between the pelvis or the feet exceeds the set threshold, the experimental animal is judged to have pelvic lifting and is scored 4 points.

[0031] In a second aspect, the present invention provides a fully automatic abdominal wall withdrawal reflex scoring device, comprising:

[0032] A micro pump module is used to slowly inject or release liquid into the balloon in the experimental animal;

[0033] Image acquisition module, used to collect real-time image data of experimental animals;

[0034] An image processing module is used to pre-process the collected image data to obtain a number of original data images;

[0035] The AI ​​automatic analysis module is used to automatically determine whether the current state of the experimental animal is a frequently active state or a resting state based on the original data image.

[0036] As a further technical solution, the AI ​​automatic analysis module is also used to automatically score the abdominal wall withdrawal reflex of the experimental animals based on the original data images.

[0037] A computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device. The fully automatic abdominal wall withdrawal reflex scoring method is disclosed.

[0038] A terminal device includes a processor and a computer-readable storage medium, wherein the processor is used to implement various instructions; and the computer-readable storage medium is used to store multiple instructions, wherein the instructions are suitable for being loaded by the processor and executing the fully automatic abdominal wall withdrawal reflex scoring method.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention avoids the problems of strong subjectivity of traditional manual operation and manual scoring by automatically timing, automatically injecting and automatically scoring multiple experimental animals, improves the accuracy of scoring, shortens the experimental time, saves manpower and improves the efficiency of the experiment.

[0041] The invention conducts experiments on experimental animals in a light-proof and sound-proof environment, thereby avoiding stimulation of the experimental animals by the external environment and increasing the accuracy of the experiments.

[0042] The present invention automatically collects and saves the experimental animal image data, which is saved and exported as the original data of the experimental animal for manual re-scoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0044] Figure 1 is a flow chart of the fully automatic abdominal withdrawal reflex scoring method of the present invention;

[0045] Figure 2 is a training flow chart of the neural network model of the present invention;

[0046] Figure 3 It is a structural schematic diagram of the fully automatic abdominal wall withdrawal reflex scoring device of the present invention.

[0047] Among them, 1 display module, 2 instrument housing, 3 mouse box, 4 insertion tube module interface, 5 fixing clip, 6 image acquisition module, 7 micro pump module, 8 temperature and humidity adjustment module, 9 controller, 10 timer, 11 alarm module, 12 AI automatic analysis module, 13 image processing module, 14 mouse box housing, 15 mouse box air vent, 16 mouse box baffle, 17 mouse box baffle groove, 18 mouse tail hole, 19 balloon, 20 water pipe, 21 insertion tube knob. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0049] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0050] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0051] Example 1

[0052] like Figure 1 A fully automated abdominal withdrawal reflex scoring method is shown, comprising:

[0053] Inserting the balloon into the body of the experimental animal and collecting real-time image data of the experimental animal after a set period of time;

[0054] Preprocessing the collected image data to obtain several original data images;

[0055] Automatically determine whether the current state of the experimental animal is a frequently active state or a resting state based on the original data image;

[0056] If the activity is frequent, the alarm is activated and the balloon is rechecked to see if it is correctly inserted;

[0057] If the animal is in a resting state, the micropump is controlled to slowly inject a set dose of liquid into the balloon, and real-time imaging data of the experimental animal in a resting state is collected at the same time;

[0058] Preprocessing the collected image data of experimental animals in a resting state to obtain a number of raw data images;

[0059] Automatically score the abdominal withdrawal reflex of experimental animals based on the raw data images;

[0060] After the automatic scoring of the abdominal withdrawal reflex of the experimental animals is completed, the micropump is controlled to slowly release the liquid in the balloon, and the automatic scoring of the abdominal withdrawal reflex under different liquid doses is performed again.

[0061] Preferably, if the activity is frequent, a re-judgment is performed after a set time. If the number of re-judgments exceeds the set number, an alarm is activated to re-check whether the balloon is correctly inserted.

[0062] In this embodiment, mice are used as experimental animals. The movement process of the mice is filmed from the side by a camera, and several groups of experimental videos of the mice's movement are recorded. The frame rate of the video is 60FPS. The collected experimental videos are preprocessed, and the video data is extracted at a frequency of one frame every 5 seconds to obtain several original data images.

[0063] In this embodiment, the specific steps of automatically determining whether the current state of the experimental animal is a frequently active state or a resting state according to the original data image include:

[0064] According to the obtained raw data images, the raw data images are extracted every 30 frames, and the extracted raw data images of mice are input into the trained neural network model to realize the target recognition and feature point annotation of mice in the images, and the skeleton diagram of mice is constructed through the annotated feature points;

[0065] The motion parameters of the mouse in the image are determined based on the feature points, and the changes in the mouse motion parameters in the original data images extracted 30 frames apart are compared. If none of them exceeds the set threshold, the mouse is considered to be in a resting state, otherwise it is considered to be in a frequently active state.

[0066] In this embodiment, if Figure 2 As shown, the abdominal withdrawal reflex of the experimental animal is automatically scored based on the original data image, and the specific steps include:

[0067] First, the collected and preprocessed raw data images of mice are input into the trained neural network model to realize the target recognition and feature point annotation of the mice in the image, and the skeleton diagram of the mice is constructed through the annotated feature points.

[0068] In the above process, the target recognition and feature point marking of the mouse in the image are realized by inputting the collected and preprocessed raw data image of the mouse into the trained neural network model. The training process of the neural network model includes:

[0069] The movement process of mice was filmed from the side using a camera, and several groups of experimental videos of mice movement were recorded. The frame rate of the video was 60FPS. The collected experimental videos were preprocessed, and the video data was framed at a frequency of one frame every 5 seconds to obtain 2000 raw data images.

[0070] For the mouse movements involved in the AWR score, including head movement, abdominal muscle contraction, body arching, and pelvic lifting, the mouse feature points were manually annotated, including the nose, ears, forehead, spine, pelvis, hind legs, and abdomen. In fact, in the actual experiment, the forehead of the mouse is often blocked by the ears, so the position of the annotated forehead is set to the midpoint of the two ears;

[0071] The original data image with feature point annotations is input into the neural network for prediction and decision making, and the neural network is trained to obtain a trained neural network model.

[0072] Taking into account the occlusion problems between individuals and within individuals in the experimental data, such as the mutual occlusion of multiple mice in the camera field of view or the occlusion of the pelvis or ears by the mouse body, traditional posture estimation algorithms, such as the graphic structure framework, are difficult to obtain high accuracy. Therefore, the present invention adopts a posture estimation algorithm based on deep learning, such as DeepPose, DeepLabCut, etc., and completes the training of the neural network model by inputting a large number of original data images marked as above.

[0073] Then, after completing the target recognition and feature point marking of the experimental animal in the image, the motion parameters of the mouse in the image are determined based on the feature points, and the motion parameters of the mouse in the image are compared with the motion parameters of the mouse in the initial frame image of the resting state. The motion state of the experimental animal is determined based on whether the variation range of the motion parameters exceeds the set threshold, and the abdominal wall withdrawal reflex of the mouse is scored based on the motion state combined with the scoring rules.

[0074] In this embodiment, the characteristic points of the experimental animal in the image are connected to determine the movement parameters of the mouse as four movement parameters: the elevation angle α, the arch angle β, the distance l between the pelvis and the foot, and the abdominal muscle contraction angle γ.

[0075] The resting state initial frame image is defined as the image of the mouse at the initial still moment in the resting state. The motion parameters of the mouse in each image data are compared with the motion parameters of the mouse in the resting state initial frame image, and automatic scoring is performed according to the comparison results. Specifically, if the change range of the four motion parameters does not exceed the set threshold, the experimental animal is judged to be unresponsive, and 1 point is recorded; if the change range of the elevation angle α exceeds the set threshold, the experimental animal is judged to have head movement, and 2 points are recorded; if the change range of the abdominal muscle contraction angle γ exceeds the set threshold, the experimental animal is judged to have abdominal muscle contraction, and 3 points are recorded; if the change range of the arch angle β exceeds the set threshold, the experimental animal is judged to have arched body, and 4 points are recorded; if the change range of the distance l between the pelvis or feet exceeds the set threshold, the experimental animal is judged to have pelvic lifting, and 4 points are recorded. Through the above scheme, automatic scoring of the mouse abdominal wall withdrawal reflex is achieved.

[0076] Example 2

[0077] like Figure 3 As shown, a fully automatic abdominal withdrawal reflex scoring device, such as Figure 3 As shown, including:

[0078] A micro pump module 7, used for slowly injecting or releasing liquid into the balloon in the experimental animal;

[0079] An image acquisition module 6 is used to acquire real-time image data of experimental animals;

[0080] The image processing module 13 is used to pre-process the collected image data to obtain a number of original data images;

[0081] The AI ​​automatic analysis module 12 is used to automatically determine whether the current state of the experimental animal is a frequently active state or a resting state based on the original data image;

[0082] Display module 1, used to display the collected real-time image data of the experimental animal and the abdominal wall withdrawal reflex score of the experimental animal;

[0083] An alarm module 11 is used to issue an alarm to remind the experimenter to recheck whether the balloon is inserted correctly;

[0084] The temperature and humidity adjustment module 8 is used to adjust and maintain the temperature and humidity in the device.

[0085] Preferably, the AI ​​automatic analysis module 12 is also used to automatically score the abdominal wall withdrawal reflex of the experimental animal based on the original data image.

[0086] In this embodiment, the device further includes a mouse box 3 for accommodating experimental animals during the experiment; an insertion tube module, specifically including a balloon 19 at the front end, a water tube 20 in the middle, and an insertion tube knob 21 at the end.

[0087] The balloon is inserted into the rectum of the experimental animal, and liquid is injected into the balloon through a water tube to expand the balloon to different degrees, thereby stimulating the experimental animal. In the process of injecting liquid into the balloon through the water tube, the injection flow rate is controlled by the insertion tube knob.

[0088] Furthermore, a fixing clip 5 is provided on the mouse box to fix the experimental animal and prevent the experimental animal from escaping from the device during the experiment.

[0089] Specifically, in the process of using the above-mentioned fully automatic abdominal withdrawal reflex scoring device to automatically score the AWR of mice, first, the temperature and humidity adjustment module of the device automatically adjusts to the appropriate temperature and humidity according to the experimental parameters and the type of experimental animals, and the experimenter checks whether the balloon 19 is intact, that is, the insertion tube knob 21 is connected to the insertion tube module interface 4, and the liquid with appropriate temperature is injected into the water pipe 20 and the balloon 19 through the micro pump, and the balloon 19 is not filled. After stimulating the mouse to empty the rectum, the balloon 19 is inserted into the mouse rectum, the mouse is placed in the mouse box 3, the mouse box opening is closed with the mouse box baffle 16, and the mouse tail and the water pipe 20 passing through the mouse tail hole 18 are fixed with the fixing clip 5 on the mouse box 3.

[0090] Start AWR automatic scoring, timer 10 starts timing, and image acquisition module 6 records mouse activity in real time. After 5 minutes, image processing module 13 transmits the real-time image information of the mouse to AI automatic analysis module 12. AI automatic analysis module analyzes and determines whether the mouse state is frequently active or resting, and transmits the information to controller 9. If it is frequently active, the controller controls alarm module 11 to prompt the experimenter to recheck whether the balloon is correctly inserted; if it is resting, the controller controls micro pump module 7 to slowly inject the set dose of liquid. After 20 seconds, image processing module 13 transmits the real-time image information of the mouse to AI automatic analysis module. AI automatic analysis module automatically scores and saves the video, and transmits the information of the end of scoring to controller 9. Afterwards, controller 9 controls micro pump module 7 to slowly release the liquid in the balloon, and controls the timer to continue timing for 4 minutes. After 4 minutes, the abdominal wall withdrawal reflex of the mouse is automatically scored again with different injection times and different doses of liquid.

[0091] In another embodiment, the image processing module 13 transmits the real-time image information of the mouse to the AI ​​automatic analysis module. When the AI ​​automatic analysis module analyzes the mouse state as a frequently active state or a resting state, if it is a frequently active state, it will be evaluated again after 1 minute. If 5 consecutive evaluations are frequent activities, the controller controls the alarm module to prompt the experimenter to recheck whether the balloon is correctly inserted.

[0092] During the above experiment, the experimenter can observe the real-time images of the mice in each mouse box and the experimental progress through the display module 1. The device of the present invention is provided with multiple compartments. When the mice in a compartment have completed the set program, the progress bar of the compartment flashes green light to prompt. At this time, the experimenter can remove the balloon from the mouse rectum and carry out the next step of processing on the mouse.

[0093] During the above experiment, the experimenter does not have to be present throughout the whole process. The experimental video can be exported later for manual scoring and retained as original data.

[0094] Example 3

[0095] A computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device to provide a fully automatic abdominal wall withdrawal reflex scoring method provided in this embodiment 1.

[0096] Example 3

[0097] A terminal device includes a processor and a computer-readable storage medium, wherein the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, wherein the instructions are suitable for being loaded by the processor and executing a fully automatic abdominal wall withdrawal reflex scoring method provided in this embodiment 1.

[0098] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0099] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0100] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0101] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0102] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0103] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A fully automatic abdominal withdrawal reflex scoring method, characterized in that: include: Insert the balloon into the body of the experimental animal, collect real-time image data of the experimental animal after a set time; pre-process the collected image data to obtain a number of raw data images, specifically: shoot the movement process of the mouse from the side by a camera, record a number of groups of experimental videos of the mouse movement, pre-process the collected experimental videos, extract frames from the video data at a set frame frequency, and obtain a number of raw data images; Automatically determine whether the current state of the experimental animal is a frequently active state or a resting state according to the original data image, specifically: according to the obtained several original data images, extract the original data image every set frame and input it into the trained neural network model, identify the target of the mouse in the image and mark the feature points, and construct the skeleton map of the mouse through the marked feature points; determine the motion parameters of the mouse in the image according to the feature points, compare the changes of the mouse motion parameters in the original data images extracted at set intervals, if they do not exceed the set threshold, it is considered that the mouse is in a resting state, otherwise it is considered that the mouse is in a frequently active state; The training process of the neural network model includes: collecting real-time image data of experimental animals; preprocessing the collected image data to obtain a number of original data images; manually marking the feature points of the experimental animals in the images; inputting the original data images with the feature points marked into the neural network for prediction and decision-making, training the neural network, and obtaining a trained neural network model; using a posture estimation algorithm based on deep learning, the posture estimation algorithm includes DeepPose or DeepLabCut, and completing the training of the neural network model by inputting a large number of original data images marked as above; If the animal is in a resting state, the micropump is controlled to slowly inject a set dose of liquid into the balloon, and real-time imaging data of the experimental animal in a resting state is collected at the same time; Preprocessing the collected image data of the experimental animal in a resting state to obtain a plurality of second original data images; automatically scoring the abdominal wall withdrawal reflex of the experimental animal according to the second original data images; The method of automatically scoring the abdominal wall withdrawal reflex of the experimental animal based on the second original data image includes: inputting the second original data image into a trained neural network model to realize target recognition and feature point annotation of the experimental animal in the image; determining the motion parameters of the experimental animal in the image according to the feature points, comparing the motion parameters of the experimental animal in the image with the motion parameters of the experimental animal in the initial frame image of the resting state, determining the motion state of the experimental animal according to whether the variation range of the motion parameters exceeds a set threshold, and completing the scoring of the abdominal wall withdrawal reflex of the experimental animal based on the motion state combined with the scoring rules; the initial frame image of the resting state is defined as an image of the mouse at the initial still moment in the resting state.

2. A fully automatic abdominal withdrawal reflex scoring method as claimed in claim 1, characterized in that: If the activity is frequent, the alarm is activated and the balloon is rechecked to see if it is inserted correctly.

3. A fully automatic abdominal withdrawal reflex scoring method as claimed in claim 1, characterized in that: The method of determining the motion parameters of the experimental animal in the image based on the feature points includes: connecting the feature points of the experimental animal in the image to determine four motion parameters of the experimental animal: the elevation angle α, the arch angle β, the distance l between the pelvis and the foot, and the abdominal muscle contraction angle γ.

4. A fully automatic abdominal withdrawal reflex scoring method as claimed in claim 3, characterized in that: The step of determining the motion state of the experimental animal according to whether the range of motion parameter variation exceeds a set threshold, and completing the scoring of the abdominal wall withdrawal reflex of the experimental animal according to the motion state combined with the scoring rules, includes: If the change range of the four motion parameters did not exceed the set threshold, the experimental animal was judged as having no reaction and scored 1 point; If the variation range of the elevation angle α exceeds the set threshold, the experimental animal is judged to have head movement and scored 2 points; If the change range of the abdominal muscle contraction angle γ exceeds the set threshold, the experimental animal is judged to have abdominal muscle contraction and scored 3 points; If the variation range of the arch angle β exceeds the set threshold, the experimental animal is judged to have an arched body and is scored 4 points; If the variation range of the distance l between the pelvis or the feet exceeds the set threshold, the experimental animal is judged to have pelvic lifting and is scored 4 points.

5. A fully automatic abdominal withdrawal reflex scoring device, used to implement a fully automatic abdominal withdrawal reflex scoring method as claimed in any one of claims 1 to 4, characterized in that: include: A micro pump module is used to slowly inject or release liquid into the balloon in the experimental animal; The image acquisition module is used to collect real-time image data of experimental animals, specifically: the movement process of mice is photographed by the camera from the side, and experimental videos of several groups of mice movement are recorded; The image processing module is used to pre-process the collected image data to obtain a number of original data images, specifically: extracting frames from the video data at a set frame frequency to obtain a number of original data images; AI automatic analysis module, used to automatically determine whether the current state of the experimental animal is a frequently active state or a resting state based on the original data image; The AI ​​automatic analysis module is also used to automatically score the abdominal wall withdrawal reflex of the experimental animal based on the second original data image.

6. A computer-readable storage medium, characterized in that: A plurality of instructions are stored therein, and the instructions are suitable for being loaded by a processor of a terminal device and implementing the steps in a fully automatic abdominal wall withdrawal reflex scoring method as described in any one of claims 1-4.

7. A terminal device, characterized in that: The method comprises a processor and a computer-readable storage medium, wherein the processor is used to implement various instructions; and the computer-readable storage medium is used to store a plurality of instructions, wherein the instructions are suitable for being loaded by the processor and executing the steps in the fully automatic abdominal wall withdrawal reflex scoring method according to any one of claims 1 to 4.

Images