Animal abnormal behavior phenotype rapid evaluation method and device

Abstract

The invention discloses an animal abnormal behavior phenotype rapid evaluation method and device, and is applied to the technical field of animal behavior analysis, and the method comprises the steps: obtaining video data of animals of different weeks old in spontaneous behavior experiments; tracking animal body key points in the video data to obtain animal behavior characteristics; performing classification extraction on the basis of the behavior characteristics to obtain target behavior parameters, and calculating the target behavior parameters to obtain indexes for evaluating animal behavior states; analyzing the difference change of the indexes, and evaluating and classifying the behavior state of the animal; by analyzing the dynamic balance relationship between the exploration behavior and the engraving behavior, the accurate classification of the animal abnormal behavior state is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of animal behavior analysis, and particularly to a method and device for rapidly evaluating abnormal animal behavior phenotypes. Background Art

[0002] As a subtype of pervasive developmental disorder, autism research is of great significance in the field of neuroscience. In autism research, researchers often use a series of complex and time-consuming behavioral tests to evaluate whether autism model animals exhibit autism-like phenotypes. These tests cover multiple aspects such as social behavior, stereotyped behavior, and marble-burying behavior, requiring switching between different experimental devices, with cumbersome operation procedures and a lack of unified standards among different research teams.

[0003] Secondly, the current evaluation methods are not fast and accurate enough. Behavioral parameters mainly rely on manual observation and recording, with strong subjectivity and difficulty in achieving high-throughput analysis. This results in the inability to quickly detect whether an autism model animal has an autism-like phenotype after construction, and also fails to meet the requirements for high-throughput model animals in drug screening.

[0004] In addition, the symptoms of autism are heterogeneous, and existing methods usually rely on a single behavioral threshold for judgment, which is easily affected by factors such as experimental environment and individual differences, leading to insufficient accuracy and stability of the detection results.

[0005] To overcome these deficiencies, the present application proposes a method and device for rapidly evaluating abnormal animal behavior phenotypes. Summary of the Invention

[0006] The purpose of the present application is to provide a method and device for rapidly evaluating abnormal animal behavior phenotypes, aiming to solve the problems of complexity, time consumption, and insufficient accuracy of existing evaluation methods.

[0007] To achieve the above purpose, the present application provides the following technical solutions:

[0008] In the first aspect, the present application provides a method for rapidly evaluating abnormal animal behavior phenotypes, including:

[0009] Obtaining video data of animals in spontaneous behavior experiments at different weeks of age;

[0010] Tracking the body key points of animals in the video data to obtain the behavioral characteristics of the animals;

[0011] Classifying and extracting target behavioral parameters based on the behavioral characteristics, and calculating the target behavioral parameters to obtain an index for evaluating the animal's behavioral state;

[0012] Analyzing the differential changes of the index to evaluate and classify the animal's behavioral state.

[0013] Further, in the step of obtaining video data of animals of different weeks old in the spontaneous behavior experiment, the following steps are specifically included:

[0014] Manually mark the body sites of the animals in the video data of the first week old, and train the animal body skeleton model respectively according to the first marking result;

[0015] Perform mixed marking of the body sites on the video data of the second week old, and perform mixed training on an animal body skeleton model according to the second marking result.

[0016] Further, in the step of tracking the key points of the animal body in the video data to obtain the behavior characteristics of the animal, the following steps are specifically included:

[0017] Adopt behavior analysis software to track the key points of the animal body in the video data through a deep learning model to obtain the behavior characteristics of the animal;

[0018] The behavior characteristics include but are not limited to: posture, speed, action type, action duration, and occurrence frequency.

[0019] Further, in the step of classifying and extracting the target behavior parameters based on the behavior characteristics and calculating the target behavior parameters to obtain an index for evaluating the animal behavior state, the following steps are specifically included:

[0020] Perform action classification on the behavior characteristics to obtain a classification result; the action classification includes but is not limited to: running, trotting, turning left, turning right, walking, stepping, sniffing in place, standing still, standing hunched, standing, climbing the wall, jumping, grooming, staying still;

[0021] Extract the target behavior parameters based on the behavior parameters obtained by classification, and the target behavior parameters are the durations of the sniffing action and the grooming action;

[0022] Calculate the ratio of the durations of the sniffing action and the grooming action, and take the logarithm of the calculation result as an index for evaluating the animal behavior state.

[0023] Further, perform clustering verification by linear discriminant analysis according to the index.

[0024] In a second aspect, the present application provides a device for rapidly evaluating the abnormal behavior phenotype of animals, including:

[0025] An acquisition module: acquiring video data of animals of different weeks old in the spontaneous behavior experiment;

[0026] Analysis module: Track the key points of the animal body in the video data to obtain the behavioral characteristics of the animal; perform classification extraction based on the behavioral characteristics to obtain target behavioral parameters, and calculate the target behavioral parameters to obtain an index for evaluating the animal's behavioral state;

[0027] Evaluation and classification module: Analyze the differential changes of the index to evaluate and classify the behavioral state of the animal.

[0028] In a third aspect, the present application provides a computer device, which includes a processor and a memory coupled to the processor. Among them, the memory stores program instructions for implementing a method for quickly evaluating the abnormal behavior phenotype of animals; the processor is used to execute the program instructions stored in the memory to implement a quick evaluation of the abnormal behavior phenotype of animals.

[0029] In a fourth aspect, the present application provides a storage medium storing program instructions executable by a processor, and the program instructions are used to execute a method for quickly evaluating the abnormal behavior phenotype of animals.

[0030] The present application provides a method and device for quickly evaluating the abnormal behavior phenotype of animals, which have the following beneficial effects:

[0031] In the present application, video data of animals in spontaneous behavior experiments at different weeks of age is obtained, and artificial marking and deep learning training are performed on the animal body points to track multiple body sites of the animal in the original video; subsequently, an unsupervised classification is performed based on the behavioral characteristics of the animal by a data-driven method, similar classifications are merged into one action, and the behavioral characteristic parameters of the animal are analyzed and obtained; on this basis, the durations of the sniffing action and the grooming action are extracted, the ratio of the two is calculated and the logarithm is taken to achieve accurate classification of the abnormal behavior state of the animal. The present application not only improves the accuracy and efficiency of evaluating the abnormal behavior state of animals, but also provides clues for early intervention. By comparing the behavioral changes of animals with different genotypes and genders at key developmental stages, early characteristics are revealed, providing a strong basis for pharmacodynamic evaluation and drug screening; in addition, it also has wide applicability and good generality. Brief Description of the Drawings

[0032] Figure 1 It is a schematic flowchart of a method for quickly evaluating the abnormal behavior phenotype of animals in Embodiment 1 of the present application;

[0033] Figure 2 It is a schematic diagram of the ratio of the total duration of the sniffing-grooming behavioral actions in Embodiment 1 of the present application;

[0034] Figure 3 It is a schematic diagram of verifying the classification result by linear discriminant analysis in Embodiment 1 of the present application;

[0035] Figure 4 This is a schematic structural diagram of a rapid assessment device for animal abnormal behavior phenotypes in Embodiment 2 of the present application;

[0036] Figure 5 This is a schematic structural diagram of a computer device in Embodiment 3 of the present application;

[0037] Figure 6 This is a schematic structural diagram of a storage medium in Embodiment 4 of the present application. Detailed implementation manners

[0038] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0039] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0040] Embodiment 1

[0041] Please refer to Figure 1 , which is a schematic flowchart of a rapid assessment method for animal abnormal behavior phenotypes in Embodiment 1 of the present application; the steps include:

[0042] S1: Obtain video data of animals of different weeks of age in spontaneous behavior experiments.

[0043] In this embodiment, it can be achieved by using multiple high-definition resolution cameras or high-definition resolution cameras with an infrared shooting mode to respectively shoot the spontaneous behaviors of the same species from multiple angles. At this time, different animals have different emotions or behaviors, and by replacing the acquisition object, the acquisition of different states can be realized. The duration of the collected video data can be set as needed. For example, the data acquisition time for a single animal can be set between 15 min and 60 min, and multiple-angle acquisitions can be performed. In actual data acquisition, the number of perspectives can also be increased or decreased as needed.

[0044] For different animals, based on morphological analysis, when each species faces different emotions or different disease states, due to differences in physiological structures, there are subtle differences in the characteristics of the external environment they perceive, and their behavioral tendencies will show obvious species limitations. Therefore, only obtaining video data of the same species in different emotional or disease states can ensure the consistency of the data. Specifically, on the premise of conforming to animal ethics, animal experimental models in different states are constructed, experimental animals in different emotions are constructed, and relevant video data is collected from them.

[0045] Manually mark the animal body sites of the video data in the first week of age, and train the animal body skeleton model respectively according to the first marking results; perform mixed marking of the body sites on the video data in the second week of age, and perform mixed training on an animal body skeleton model according to the second marking results. For example: collect 60-minute spontaneous behavior experiment videos of animals according to the standardized process, and manually mark 1000 - 3000 frame pictures of 16 body points of the animals (nose, left ear, right ear, neck, left front limb, right front limb, left hind limb, right hind limb, left front paw, right front paw, left hind paw, right hind paw, back, root of the tail, middle of the tail, tip of the tail). Train the marked images through DeepLabCut deep learning to obtain the corresponding animal body point marking model. Analyze animals of different weeks of age respectively, perform manual marking of animal body sites on the videos of 3, 4, and 5 weeks of age, and train the animal body skeleton model respectively; for the videos of 6, 7, 8, and 12 weeks of age, mix and mark the body sites to establish an animal body skeleton model.

[0046] S2: Track the key body points of the animals in the video data to obtain the behavioral characteristics of the animals.

[0047] In this embodiment, through fine ethological analysis programs, such as Behavior Atlas (a new intelligent animal behavior precise analysis system), Moseq (action sequencing algorithm / a behavioral analysis algorithm that can recognize three-dimensional mouse body language (referred to as "syllables")), LEAP (behavior analysis model), DeepLabCut (a pose learning tool for pose estimation and motion analysis), EthoVision (an animal movement trajectory tracking system), etc., methods for tracking multiple key body sites of animals are used to automatically track multiple key body sites on the animals. The specific process of the above fine ethological analysis program is as follows: First, calculate the two-dimensional spatial coordinate information of the key body sites of the animal body according to the position information calibrated by the camera. Subsequently, integrate the spatial coordinate information of the key body sites of multiple images and calculate the coordinate information of the key points in three-dimensional space.

[0048] Tracking the key points of the animal body in video data through a deep learning model to obtain the behavioral characteristics of the animal; the behavioral characteristics include but are not limited to: posture, speed, action type, action duration, and occurrence frequency.

[0049] S3: Classifying and extracting the target behavioral parameters based on the behavioral characteristics, and calculating the target behavioral parameters to obtain an index for evaluating the animal's behavioral state.

[0050] In this embodiment, an unsupervised action classification is performed according to the behavioral characteristics by a data-driven method to obtain a classification result. The action classification includes but is not limited to: running, trotting, turning left, turning right, walking, stepping, sniffing in place, standing still, standing hunched, standing, climbing the wall while holding, jumping, grooming, staying still;

[0051] On this basis, the behavioral parameters obtained based on the classification are acquired, and the target behavioral parameters are extracted from the behavioral parameters. The target behavioral parameters are the durations of the sniffing action and the grooming action. These parameters are the basis for subsequent analysis and can reflect the behavioral patterns of animals in a natural state.

[0052] Please refer to Figure 2 , which is a schematic diagram of the ratio of the total duration of the sniffing-grooming behavioral actions in Embodiment 1 of the present application. Calculate the ratio of the durations of the sniffing action and the grooming action, and take the logarithm log10 of the calculation result as an index for evaluating the animal's behavioral state.

[0053] Please refer to Figure 3 , which is a schematic diagram for verifying the classification result by linear discriminant analysis in Embodiment 1 of the present application, and verify Figure 2 the accuracy of the index classification. According to the ratio of the durations of the sniffing action and the grooming action, linear discriminant analysis is further used for clustering verification.

[0054] S4: Analyze the differential changes of the index to evaluate and classify the behavioral state of the animal.

[0055] In this embodiment, the change trend of the ratio of the durations of the sniffing action and the grooming action at different developmental stages is further analyzed to identify potential disease characteristics. By comparing animals of different genotypes and genders, analyzing their behavioral changes at key developmental stages, revealing the early characteristics of diseases, and providing a reference basis for early diagnosis and intervention. If this index has significant changes at certain specific weeks of age, such as 3 weeks of age or 5 weeks of age, it indicates a critical window period for disease development, providing a direction for subsequent in-depth research and clinical applications. At the same time, this method is applicable to mice with Shank3b gene mutations and can be extended to the evaluation of other animals or clinical patients.

[0056] In addition, this application designs systematic verification experiments for multiple genotypes, such as wild type (WT), heterozygous type (HE), homozygous type (KO), transsexuality, and throughout the entire development cycle. The experiment is divided into 6 groups, with each group containing 15 mice. Spontaneous behavior experiments are conducted once at the ages of 3, 4, 5, 6, 7, 8, and 12 weeks. Spontaneous behavior is monitored for 60 minutes at each age stage using a standardized square open field device (50×50×30 cm), and videos are collected through cameras at the four corners. Through the analysis and calculation of behavioral data, it is verified that the logarithm of the ratio of sniffing and grooming actions can accurately distinguish autistic model mice of different genotypes. Homozygous and heterozygous autistic model mice at each age stage have significant statistical differences compared with the control wild type mice, such as Figure 2 . At 3 weeks and 5 weeks of age, this value shows a drastic change compared with other age stages and drops to a negative value, indicating that it is a critical window period for disease development. Although this ratio fluctuates within the same group of mice, it is generally stably maintained within a certain range. This is highly consistent with the results obtained by applying linear discriminant analysis. The above experimental results show that this method has high stability and good generalization ability.

[0057] It should be noted that the application object of this application is not limited to mice, but can be extended to more animals and even applied to the evaluation of clinical patients. Its core logic is to classify abnormal animal behaviors by analyzing the dynamic balance relationship between the animal's instinctive exploration and the stereotyped behaviors caused by disease phenotypes. This logical framework is universal and applicable to a variety of experimental animal models. Further, the method proposed in this application can not only serve as an important clue for early disease prediction and intervention, but also provide a strong basis for early drug screening and intervention in scientific research. By early identifying behavioral characteristics, researchers can more accurately evaluate the effects of potential therapeutic drugs, thereby accelerating the drug R & D process. This method can also be used to monitor disease progression and evaluate the effectiveness of intervention measures, providing data support for clinical treatment.

[0058] In summary, in Example 1 of this application, video data of animals in spontaneous behavior experiments at different ages are first obtained, and artificial markers and deep learning training are performed on the animal body points to track multiple body sites of the animal in the original video. Subsequently, an unsupervised classification is performed based on the behavioral characteristics of the animal through a data-driven method, similar classifications are merged into one action, and the behavioral characteristic parameters of the animal are analyzed and obtained. On this basis, the durations of sniffing and grooming actions are extracted, the ratio of the two is calculated and the logarithm is taken to achieve accurate evaluation of abnormal animal behaviors. This method not only improves the accuracy and efficiency of evaluating abnormal animal behaviors, but also provides clues for early intervention and treatment. By comparing the behavioral changes of animals of different genotypes and genders at critical developmental stages, early characteristics are revealed, providing a strong basis for pharmacodynamic evaluation and drug screening.

[0059] Example 2

[0060] Please refer to Figure 4 , which is a schematic structural diagram of an animal behavior analysis device according to Embodiment 2 of the present application; the specific content includes:

[0061] Acquisition module: Acquire video data of animals of different weeks of age in the spontaneous behavior experiment;

[0062] Analysis module: Track the key points of the animal body in the video data to obtain the behavior characteristics of the animal; classify and extract the target behavior parameters based on the behavior characteristics, and calculate the target behavior parameters to obtain an index for evaluating the animal behavior state;

[0063] Evaluation and classification module: Analyze the differential changes of the index, and evaluate and classify the behavior state of the animal.

[0064] Example 3

[0065] Please refer to Figure 5 , which is a schematic structural diagram of a computer device according to Embodiment 3 of the present application. The computer device 50 includes a processor 51 and a memory 52 coupled to the processor 51.

[0066] The memory 52 stores program instructions for implementing the above-mentioned animal behavior analysis method.

[0067] The processor 51 is used to execute the program instructions stored in the memory 52 to implement an animal behavior analysis.

[0068] Among them, the processor 51 can also be called a CPU (Central Processing Unit, central processing unit).

[0069] The processor 51 may be an integrated circuit chip with signal processing capabilities. The processor 51 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0070] Example 4

[0071] Please refer to Figure 6, which is a schematic structural diagram of the storage medium according to Embodiment 4 of the present application. The storage medium of the embodiment of the present application stores a program file 61 that can implement all the above methods. Among them, the program file 61 can be stored in the above storage medium in the form of a software product, including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, or devices such as computers, servers, mobile phones, and tablets.

[0072] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, device, article or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, device, article or method including that element.

[0073] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present application by the same token.

[0074] Although the embodiments of the present application have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present application. The scope of the present application is defined by the appended claims and their equivalents.

[0075] Certainly, the present invention can also have other various embodiments. Based on this embodiment, other embodiments obtained by those of ordinary skill in the art without any creative work belong to the scope protected by the present invention.

Claims

1. A method for rapid assessment of abnormal behavior phenotypes of animals, characterized in that: include: Obtain video data of animals of different ages during spontaneous behavior experiments; Track the key points of the animal's body in the video data to obtain the animal's behavioral characteristics; Classifying and extracting the behavioral characteristics to obtain target behavioral parameters, and calculating the target behavioral parameters to obtain indicators for evaluating the behavioral state of the animal; The differential changes of the indicators are analyzed to evaluate and classify the behavioral status of the animals.

2. The method for rapid assessment of abnormal behavior phenotypes of animals according to claim 1, characterized in that: The step of obtaining video data of animals of different ages in a spontaneous behavior experiment specifically includes the following steps: Manually mark the body parts of the animals in the video data of the first week of age, and train the animal body skeleton models respectively according to the first marking results; The body parts of the video data of the second week of age are mixedly labeled, and an animal body skeleton model is mixedly trained based on the second labeling result.

3. The method for rapid assessment of abnormal behavior phenotypes of animals according to claim 1, characterized in that: The step of tracking the key points of the animal's body in the video data to obtain the animal's behavioral characteristics specifically includes the following steps: Using behavioral analysis software to track key points of the animal's body in the video data through a deep learning model to obtain the animal's behavioral characteristics; The behavior characteristics include but are not limited to: posture, speed, action type, action duration and frequency of occurrence.

4. The method for rapid assessment of abnormal behavior phenotypes of animals according to claim 1, characterized in that: The step of extracting and classifying the behavior characteristics to obtain target behavior parameters, and calculating the target behavior parameters to obtain indicators for evaluating the behavior state of the animal specifically includes the following steps: Performing action classification on the behavior characteristics to obtain classification results; the action classification includes but is not limited to: running, trotting, turning left, turning right, walking, stepping, sniffing in place, standing in place, standing with a hunched back, standing, climbing with the help of a wall, jumping, grooming, and not moving; Extracting target behavior parameters based on the behavior parameters obtained through classification, wherein the target behavior parameters are the duration of the sniffing action and the grooming action; The ratio of the duration of sniffing action to grooming action was calculated, and the logarithm of the result was taken as an indicator for evaluating the behavioral state of the animal.

5. The method for rapid assessment of abnormal behavior phenotypes of animals according to claim 4, characterized in that: According to the indicators, linear discriminant analysis was used for clustering verification.

6. A device for rapid assessment of abnormal behavior phenotypes of animals, characterized in that: include: Acquisition module: Acquisition of video data of animals of different ages in spontaneous behavior experiments; Analysis module: Track key points of the animal's body in the video data to obtain the animal's behavioral characteristics; Classifying and extracting the behavioral characteristics to obtain target behavioral parameters, and calculating the target behavioral parameters to obtain indicators for evaluating the behavioral state of the animal; Evaluation and classification module: Analyze the differential changes of the indicators to evaluate and classify the behavioral status of the animals.

7. A computer device, characterized in that: The computer device includes a processor and a memory coupled to the processor, wherein the memory stores program instructions for implementing a method for rapid assessment of an abnormal behavior phenotype of an animal as described in any one of claims 1 to 5; and the processor is used to execute the program instructions stored in the memory to implement a rapid assessment of an abnormal behavior phenotype of an animal.

8. A storage medium, characterized in that: The method stores program instructions executable by a processor, wherein the program instructions are used to execute a method for rapidly evaluating an abnormal behavior phenotype of an animal as described in any one of claims 1 to 5.

Images