Apparatus and method for testing social hierarchies and social behaviors of mice

Abstract

The present application relates to the technical field of ethology. Disclosed are an apparatus and method for testing the social hierarchies and social behaviors of mice. A plurality of mice are placed in an experimental box, and a semi-natural environment is provided. At a group level, on the basis of radio frequency identification chips fixed to the bodies of the mice and a radio frequency identification position tracking system arranged at the bottom of the experimental box, real-time identification and tracking is performed on activities or movements of each mouse in a group in the semi-natural environment. Videos of the mice are captured by means of an infrared camera system, and computer vision and a machine learning algorithm are used to automatically identify and analyze movement phenotypes and interactive behaviors of each mouse. Long-term tracking and analysis is performed in the semi-natural environment, so as to observe changes in the social behaviors of a mouse group. Long-term observation can provide more comprehensive data, thereby helping researchers better quantify the social behaviors of the mice in the group, explore the social memories of the mice, understand the process of the establishment of the social hierarchies of the mice, and automatically analyze the social hierarchies of the mice in the group.

Description

Apparatus and method for detecting social rank and social behavior of mice TECHNICAL FIELD

[0001] The present application relates to the technical field of animal behavior, in particular to an apparatus and method for detecting social rank and social behavior of mice. BACKGROUND

[0002] Social hierarchy is a multidimensional trait that has profound effects on the emotions and cognition of humans and other social species, and has important influences on the social organization, survival, reproductive success and health of animals in groups. In the research of neuroscience, there are many methods for measuring the social rank of mice, such as pipeline test, competition for food reward, direct observation and automatic long-term tracking of mouse populations in semi-natural state to determine the rank through phenotype analysis.

[0003] In the pipeline test, two mice are placed in a transparent pipeline, both ends of which are open. When the two mice meet, they will push each other to try to push the other out of the pipeline. By observing which mouse can successfully push the other out, the social rank between them can be determined. However, the rank test using this method is not stable when measuring more than four mice in a cage, is easily affected by the state of the mice themselves, and requires a large amount of time for training and testing in the early stage.

[0004] In the competition for food reward test, mice are placed in a limited food resource environment, and they need to compete for food. Mice with higher social status usually get more food, while mice with lower social status may be excluded or get less food. By observing the behavior of mice in the struggle for food, such as pushing and fighting, their position in the social hierarchy can be inferred. However, this method is not effective in evaluating the social rank of multiple mice in the same group.

[0005] The direct observation method determines the rank by observing the fighting conditions of mice in the cage and whether the whiskers are intact. However, it is time-consuming and labor-intensive, and even if fighting is seen, it is not possible to determine which two mice are involved. The probability of a mouse being shaved is very low.

[0006] In the automatic long-term tracking of mouse populations in semi-natural state to determine the rank through phenotype analysis, radio frequency identification (RFID) technology is used in combination with video analysis to integrate tracking software with recently developed visual-based tracking algorithms. The social rank in the same group can be determined based on the chasing / pursued and avoidance / approach behaviors between mice. However, this method can only track the position of animals and cannot analyze the motion characteristics of mice in detail. The indicators used to judge the social rank are too few, and the results are not significant.

[0007] SUMMARY

[0008] The embodiment of the present application provides a device for detecting social rank and social behavior of mice, so as to solve the problems in the prior art, such as how to identify the identities of multiple mice with the same appearance, the same volume and the same hair color, how to define and judge the social behavior of the positioned mice and the behavior related to the social rank, and how to comprehensively use multi-dimensional data to make the experimental results more reliable.

[0009] Correspondingly, the embodiment of the present application also provides a method for detecting social rank and social behavior of mice, which is used to ensure the implementation and application of the above method.

[0010] In order to solve the above technical problems, the embodiment of the present application discloses a device for detecting social rank and social behavior of mice, which comprises an experimental box in which multiple mice are placed, a radio frequency identification position tracking system arranged at the bottom of the experimental box, a signal input and output control module, and a behavior analysis system; a radio frequency identification chip is fixed on the body of each mouse; the radio frequency identification position tracking system is used to receive the mouse identity information sent by the radio frequency identification chip; the signal input and output control module comprises an infrared camera system, and the infrared camera system is used to shoot mouse videos; the behavior analysis system is in communication connection with the radio frequency identification position tracking system and the infrared camera system, and is used to analyze the mouse identity information and the mouse videos by using a preset computer vision and machine learning algorithm, so as to obtain the social rank and the social behavior of each mouse.

[0011] Preferably, the device for detecting social rank and social behavior of mice according to claim 1, wherein the behavior analysis system comprises:

[0012] a software record storage module, which is used to receive and store the mouse identity information and the mouse videos;

[0013] a time matching server software, which is used to synchronize the mouse identity information and the mouse videos;

[0014] a video processing module, which is used to mark the features of the mice in the feature frames extracted from the mouse videos, so as to generate feature data;

[0015] a behavior analysis module, which is used to identify the social behavior and the social rank of the mice according to the feature data.

[0016] Preferably, the time matching server software comprises:

[0017] a video frame extraction unit, which is used to perform frame extraction processing on the mouse videos, so as to obtain multiple image frames;

[0018] a position synchronization unit, which is used to obtain the mouse identity information of each image frame at the same time point, and synchronize the mouse identity information with the mice in the image frames.

[0019] Preferably, the behavior analysis module comprises:

[0020] a behavior recognition unit configured to recognize the social behavior of the mice according to the mouse identity information and the corresponding feature data;

[0021] a rank ordering unit configured to determine the order of eating and drinking of the mice, and determine the social rank of the mice according to the order of eating and drinking of the mice and the social behavior of the mice.

[0022] Preferably, the experimental box comprises a competition arena, a mouse refuge area, a food delivery device and a water delivery device.

[0023] The radio frequency identification position tracking system comprises:

[0024] The competition arena tracking module comprises a radio frequency identification array composed of a plurality of first radio frequency antennas; the radio frequency identification array is arranged on the bottom of the competition arena.

[0025] The mouse refuge area tracking module comprises at least one second radio frequency antenna arranged at the entrance of the mouse refuge area.

[0026] The food delivery device tracking module comprises at least one third radio frequency antenna arranged in the food delivery device.

[0027] The water delivery device tracking module comprises at least one fourth radio frequency antenna arranged in the water delivery device.

[0028] The radio frequency identification position tracking system further comprises a receiver, and the first radio frequency antenna, the second radio frequency antenna, the third radio frequency antenna and the fourth radio frequency antenna are all connected to the corresponding receiver.

[0029] Preferably, the experimental box further comprises a standby experimental area.

[0030] The radio frequency identification position tracking system further comprises a standby experimental area tracking module, and the experimental area tracking module comprises at least one fifth radio frequency antenna arranged in the standby experimental area.

[0031] The application further discloses a method for detecting the social rank and social behavior of mice, which comprises the following steps:

[0032] Receiving the mouse identity information sent by the radio frequency identification chip fixed on the mouse through the radio frequency identification position tracking system, and sending the mouse identity information to the behavior analysis system;

[0033] Taking a mouse video by using an infrared camera system, and sending the mouse video to the behavior analysis system;

[0034] Based on the behavior analysis system, using a preset computer vision and machine learning algorithm to analyze the mouse identity information and the mouse video, and obtaining the social rank and social behavior of each mouse.

[0035] Preferably, based on the behavioral analysis system, the mouse identity information and the mouse video are analyzed using preset computer vision and machine learning algorithms to obtain the social rank and social behavior of each mouse, including:

[0036] receiving and storing mouse identity information and mouse video;

[0037] synchronizing the mouse identity information with the mouse video;

[0038] annotating the features of the mice in each frame of the mouse video to generate feature data;

[0039] identifying the social behavior and social rank of the mice according to the feature data.

[0040] Preferably, synchronizing the mouse identity information with the mouse video includes:

[0041] frame extraction processing is performed on the mouse video to obtain a plurality of image frames;

[0042] acquiring mouse identity information at the same time point for each image frame, and synchronizing the mouse identity information with the mice in the image frame.

[0043] Preferably, identifying the social behavior and social rank of the mice according to the feature data includes:

[0044] identifying the social behavior of the mice according to the mouse identity information and the corresponding feature data;

[0045] determining the order of eating and drinking of the mice, and determining the social rank of the mice according to the order of eating and drinking of the mice and the social behavior.

[0046] In the embodiments of the present application, an automatic behavior tracking technology is used, multiple mice are placed in an experimental box, and a semi-natural environment is provided. At the group level, based on the radio frequency identification chip fixed on the mice and the radio frequency identification position tracking system placed at the bottom of the experimental box, the activities or movements of each mouse in the semi-natural environment are identified and tracked in real time. The mouse video is captured by an infrared camera system, and computer vision and machine learning algorithms are used to automatically identify and analyze the movement phenotype and interactive behavior (such as social behavior, competition and obedience behavior) of each mouse. Moreover, long-term tracking analysis is performed in the semi-natural environment to observe the changes in social behavior of the mouse group. This long-term observation can provide more comprehensive data to help researchers better quantify the social behavior of mice in the group, explore the social memory of mice, and understand the process of establishing the social rank of mice. The social rank of mice in the group is automatically analyzed.

[0047] The additional aspects and advantages of the embodiments of the present application will be set forth in part in the description that follows, and in part will be obvious from the description, or can be learned by practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0048] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the appended drawings, wherein:

[0049] Fig. 1 is a schematic diagram of an apparatus for detecting social rank and social behavior of mice according to an embodiment of the present application;

[0050] Fig. 2 is an interface diagram for labeling behavior body feature points of mice in video frames using DeepLabCut according to an embodiment of the present application;

[0051] Fig. 3 is an interface diagram for performing individual behavior analysis on a region of interest according to an embodiment of the present application;

[0052] Fig. 4 is a flowchart of a method for detecting social rank and social behavior of mice according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] Embodiments of the present application are described in detail below with reference to the attached drawing figures, wherein the embodiments of the application are shown as examples. The embodiments of the application described below are examples for explaining the present application and should not be construed as limiting the present application.

[0054] Those skilled in the art can understand that, unless specifically stated otherwise, the singular forms "a," "an," and "the" as used herein are intended to include plural forms as well. It should be further understood that the terms "comprising," "including," "containing," or "having," and the like, as used herein, mean that there are possibilities of including other features, integers, steps, operations, elements, components, and / or combinations thereof, in addition to those elements mentioned. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements can also be present. In addition, the use of "connected" or "coupled" herein also includes wireless connection or wireless coupling. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0055] Those skilled in the art will understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0056] To solve at least one of the technical problems existing in the prior art, the device and method for detecting social rank and social behavior of mice are provided.

[0057] The technical solutions of the present application and how the technical solutions solve the above technical problems will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes may not be described again in some examples. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0058] The embodiments of the present application provide a possible implementation, as shown in FIG. 1, which provides a device schematic diagram for detecting social rank and social behavior of mice.

[0059] As shown in FIG. 1, the device includes an experimental box in which multiple mice are placed, a radio frequency identification position tracking system (i.e., the radio frequency identification (RFID) position tracking system array in FIG. 1) arranged at the bottom of the experimental box, a signal input and output control module, and a behavior analysis system; the mice have a radio frequency identification chip fixed thereon; the radio frequency identification position tracking system is used to receive the mouse identity information sent by the radio frequency identification chip; the signal input and output control module includes an infrared camera system, which is used to shoot mouse videos; the behavior analysis system is in communication connection with the radio frequency identification position tracking system and the infrared camera system, and is used to analyze the mouse identity information and the mouse videos using a preset computer vision and machine learning algorithm to obtain the social rank and social behavior of each mouse.

[0060] It is a major challenge to quantitatively and objectively measure the behavior parameters within a group, as it requires monitoring the positions of multiple mice simultaneously and taking into account many complex factors. Automatic tracking and social behavior analysis of each mouse in the group can overcome the limitations of manual observation methods. The experimental device in the embodiments of the present application can automatically track the positions of multiple mice with similar appearances even if the hair color is uniform in a semi-natural environment. Combining video and radio frequency identification tracking data, detailed behavior profiles of individuals and groups can be obtained.

[0061] Successful social behavior analysis requires the identification of key social interactions in real-world environments to capture the social and environmental conditions that occur in natural habitats. However, more and more methods are being developed to study animal social behavior in laboratory environments. Reliability is usually achieved by sacrificing environmental complexity. This is mainly because it is a highly challenging task to study the social behavior of groups over a long period of time in semi-natural environments. It requires tracking multiple identified individuals with high spatiotemporal accuracy, so that a series of complex behavior patterns can be accurately identified.

[0062] For the problem of identity recognition, the radio frequency identification (RFID) technology is adopted in the embodiments of the present application. It is a non-contact data transmission technology, including a transponder (attached to the animal or implanted under the skin of the animal) and a receiver. The miniaturized passive integrated transponder (PIT) tag utilizes the electromagnetic field energy of the receiver, providing reliable long-term identification not only in wild animal research, but also in livestock, laboratory animals, zoo animals or pets. Combined with an automatic monitoring system, the PIT tag further allows innovative research on various aspects of animal behavior with little impact on the physiology or behavior of the research animals. By integrating video and RFID tracking datasets, the device can simultaneously and continuously track the identity and spatial location of many animals without the need for visual markers, and these datasets are time-synchronized and then fused by machine-based algorithms. This hybrid tracking method simultaneously provides X, Y coordinate positions, velocities and directions for many uniquely identified social interaction individuals, with an average spatial accuracy of ±0.5 cm and a temporal accuracy of 30-60 ms. Tracking technology for each mouse behavior is a useful tool for defining complex social interactions between pairs of animals, as well as analyzing the formation and stability of the dominant-subordinate hierarchy within the group.

[0063] In the embodiments of the present application, a radio frequency identification chip fixed on the mouse is used as a transponder, and a radio frequency identification position tracking system is used as a receiver (such as the RFID receiver shown in FIG. 1) to facilitate the identity recognition and position tracking of multiple mice in the experimental box.

[0064] For how to accurately locate and track mice and classify social behavior, computer vision and machine learning algorithms are used in the embodiments of the present application, which can be implemented using existing open source toolkits such as DeepLabCut and SimBA. In the embodiments of the present application, the behavior analysis system can be run on any electronic device, and optionally, the behavior analysis system can be run on a computer. The behavior analysis system is provided with computer vision and machine learning algorithms, and the radio frequency identification position tracking system and the infrared camera system upload and store mouse identity information and mouse video to the computer. The behavior analysis system on the computer processes the mouse position information and mouse identity information, automatically analyzes the social interaction behavior of the mouse, and outputs the results.

[0065] In the embodiments of the present application, an automatic behavior tracking technology is used, and multiple mice are placed in an experimental box and provided with a semi-natural environment. At the group level, the activities or movements of each mouse in the semi-natural environment are identified and tracked in real time based on a radio frequency identification chip fixed on the mouse and a radio frequency identification position tracking system placed at the bottom of the experimental box. A video of the mice is captured by an infrared camera system, and computer vision and machine learning algorithms are used to automatically identify and analyze the movement phenotype and interactive behavior (such as social, competitive, and submissive behavior) of each mouse. Moreover, long-term tracking analysis in the semi-natural environment is performed to observe the changes in the social behavior of the mouse group. This long-term observation can provide more comprehensive data to help researchers better quantify the social behavior of the mice in the group, explore the social memory of the mice, and understand the process of establishing the social hierarchy of the mice. The social hierarchy of the mice in the group is automatically analyzed.

[0066] In an optional embodiment, as shown in FIG. 1, the experimental box includes a competition arena, a mouse escape area, a food delivery device, and a water delivery device.

[0067] The radio frequency identification position tracking system includes:

[0068] The competition arena tracking module includes a radio frequency identification array composed of multiple first radio frequency antennas; the radio frequency identification array is covered at the bottom of the competition arena.

[0069] The mouse escape area tracking module includes at least one second radio frequency antenna, and the second radio frequency antenna is arranged at the entrance of the mouse escape area.

[0070] The food delivery device tracking module includes at least one third radio frequency antenna, and the second radio frequency antenna is arranged in the food delivery device.

[0071] The water delivery device tracking module includes at least one fourth radio frequency antenna, and the fourth radio frequency antenna is arranged in the water delivery device.

[0072] The radio frequency identification position tracking system further includes a receiver, and the first radio frequency antenna, the second radio frequency antenna, the third radio frequency antenna, and the fourth radio frequency antenna are all connected to the corresponding receivers.

[0073] The experimental box in the embodiment of the present application can be an experimental box body, and a competition arena is formed in the experimental box body; as shown in FIG. 1, a food delivery device and a water delivery device are arranged outside the experimental box body, and the food delivery device and the water delivery device are in communication with the competition arena; a mouse escape area can be arranged outside the competition arena and has an opening facing the inside of the competition arena. The above-mentioned experimental box can provide a semi-natural environment for mice and consider the influence of various factors. Optionally, the experimental box body is made of frosted opaque white acrylic material, and has a length / width of 40-80 cm and a height of 20-50 cm. The mouse escape area is made of an acrylic plate, and has a length / width of 10-50 cm and a height of 10-50 cm.

[0074] In the embodiment of the present application, the radio frequency identification position tracking system identifies and tracks the identity and position of the mice in the competition arena through a competition arena tracking module. A radio frequency identification array is arranged on the bottom of the competition arena, and the radio frequency identification array is composed of a plurality of first radio frequency antennas. Each mouse is fixed with a radio frequency identification chip, and the radio frequency identification chip has a unique identification. When each mouse walks to a position, the radio frequency identification chip installed on the mouse exchanges information with the first radio frequency antenna on the bottom of the competition arena, so that the position of each mouse can be reported. In combination with a mouse video, the identity of the mouse can be identified, and the position of each mouse can be tracked.

[0075] The radio frequency identification position tracking system tracks the mouse entering the mouse escape area through a mouse escape area tracking module. When the mouse enters the mouse escape area, the radio frequency identification chip installed on the mouse exchanges information with the first radio frequency antenna at the entrance of the mouse escape area, so that the mouse can be registered.

[0076] The radio frequency identification position tracking system tracks the mouse entering the food and water through a food delivery device tracking module and a water delivery device tracking module. After the mice in the experimental box are deprived of water or food (24 h), the third radio frequency antenna on the food delivery device and the fourth radio frequency antenna on the water delivery device can be used to determine the order of eating and drinking of the mice, so as to understand the competition ability and social rank of the mice and compare them with the competition results analyzed by the software, thereby ensuring the accuracy of the experimental results.

[0077] Optionally, the radio frequency identification array is composed of 16 circular radio frequency antennas, each radio frequency antenna coil has a diameter of 40-70 mm, and is connected to a corresponding array card reader. The diameter between adjacent coils is 120-200 mm. The radio frequency antenna coil in the mouse escape area, the food delivery device and the water delivery device has a diameter of 41-43 mm. The radio frequency identification chip can be implanted subcutaneously in the mouse, and has a size of (1-2 mm) x (5-15 mm).

[0078] In the embodiments of the present application, the radio frequency identification position tracking system operates at 134.2 kHz, and the radio frequency identification array is equipped with a 10A, 5V DC distributed power supply system.

[0079] In the embodiments of the present application, the device for detecting social rank and social behavior of mice further comprises a central control unit, and all the receivers are connected to the central control unit which communicates with a computer. In the embodiments of the present application, the receiver sends the mouse identity information to the central control unit, and then stores it on the computer. The input and output control module is arranged on the top of the experimental box, and automatically records the fine behavior, movement trajectory, speed and time proportion of the mouse in the experimental device through the video recording function of the infrared camera system (such as a CCD camera), obtains the mouse video, and sends the video to the central control unit and stores it on the computer, thereby providing data for evaluating the normal state of the mouse. Alternatively, the infrared camera system comprises an infrared camera, and the model of the infrared camera can be Basler, acA1300-60gmNIR, 30 frames per second, 720x480 pixels.

[0080] The signal input and output control module further comprises a lighting system, a controller and a memory. The lighting system provides illumination in the experimental box. The food delivery device and the water delivery device each comprise a switch control unit, and the controller is in communication connection with the switch control unit, for controlling the delivery time and delivery amount of food and water. Specifically, the signal input end of the switch control unit is connected with the output end of the controller.

[0081] In the embodiments of the present application, the radio frequency identification position tracking system and the signal input and output control module are used for automatic data collection and analysis. This automatic data collection can improve the accuracy and reliability of the data, and reduce the workload of researchers. Moreover, the device in the embodiments of the present application is closer to the natural environment, and the experiment in the semi-natural environment can provide an environment closer to the natural behavior of mice, so that the research results have better ecological validity. Compared with the drill pipe experiment and the three-box social experiment, the semi-natural environment can better simulate the social behavior of mice in the natural environment.

[0082] In an alternative embodiment, the behavior analysis system comprises:

[0083] a software record storage module for receiving and storing mouse identity information and mouse video;

[0084] a time matching server software for synchronizing the mouse identity information with the mouse video;

[0085] a video processing module for labeling the features of the mouse in the feature frames extracted from the mouse video, and generating feature data;

[0086] a behavior analysis module for identifying the social behavior and social rank of the mouse according to the feature data.

[0087] In the embodiments of the present application, the spatial position data obtained by RFID for each mouse is recorded by using a software recording storage module. The module aims to extract the information transmitted to the RFID decoder immediately each time the mouse crosses / stands above one of the radio frequency (RFID) antennas. The module locally stores the mouse identity information on the RFID recording computer, which is time-synchronized with the mouse video, and then fuses it by using the time matching server software (FlexibleSoft) installed on the video and RFID recording computer. Through the integrated data set of mouse video and mouse identity information, high-precision spatiotemporal tracking data is obtained.

[0088] The video processing module in the embodiments of the present application uses the existing DeepLabCut open source package, which is an effective method of 2D and 3D label-free pose estimation based on deep neural network transfer learning, and can achieve excellent results (i.e., can match human labeling accuracy) with minimal training data (usually 50-200 frames). In the embodiments of the present application, the versatility of the framework is demonstrated by tracking various body parts in a wide range of behavioral sets of multiple species. The software package is open source, fast and powerful, and can be used to calculate 3D pose estimation or multi-animal. It can identify 4 mice in the same cage, the head and neck, legs, paws, tail and other feature points of each mouse, and finally export these feature point position information into a csv file for further processing.

[0089] In the embodiments of the present application, the video processing module automatically frames the mouse video by using DeepLabCut, as shown in FIG. 2, and the position synchronization unit labels the feature points on the mouse body for the extracted video frames, including but not limited to the head, neck, legs and feet. Through neural network transfer learning, the feature points of the mouse on the entire video are automatically tracked, and a csv file is generated.

[0090] The behavior analysis module in the embodiments of the present application uses the existing SimBA open source package. SimBA is an open source package with a graphical interface and workflow (Simple Behavior Analysis, SimBA) that uses pose estimation to create supervised machine learning predictive classifiers of rodent social behavior. There are several excellent computational frameworks that enable high-throughput and consistent tracking of freely moving unmarked animals. Here, an embodiment of the present application introduces and distributes a workflow that enables users to combine pose estimation methods, behavior annotation, and generate supervised machine learning behavior prediction classifiers. By developing this workflow to analyze complex social behavior, it is also considered that users can generate prediction classifiers for other forms of behavior with minimal effort and without specialized computing background. More importantly, this open source software can synchronize RFID tracking data. A real-time tracking system is built.

[0091] In the embodiments of the present application, the hierarchical behavior of mice is defined from all social behaviors, including chasing, lateral threat, genital sniffing, following, fighting, and conquest, and the behavior in the video is defined on SimBA.

[0092] In an optional embodiment, the time matching server software includes:

[0093] A video frame extraction unit is configured to extract frames from the mouse video to obtain a plurality of image frames.

[0094] A position synchronization unit is configured to obtain mouse identity information at the same time point for each image frame, and synchronize the mouse identity information with the mouse in the image frame.

[0095] In an optional embodiment, the behavior analysis module includes:

[0096] A behavior recognition unit is configured to recognize the social behavior of the mouse according to the mouse identity information and the corresponding feature data.

[0097] A hierarchical sorting unit is configured to determine the order of eating and drinking of the mouse, and determine the social rank of the mouse according to the order of eating and drinking of the mouse and the social behavior.

[0098] In the embodiments of the present application, SimBA is used to manually define and annotate the social behavior of mice. This tool can learn the movement patterns of the limbs of this behavior, and a part of the data is used for model training. After the model training is completed, the social interaction behavior of the mouse can be automatically analyzed, and the output can be obtained. In the embodiments of the present application, the mice in the group are ranked according to the chasing and being chased, fighting conditions, and conquest behavior, and the social rank of the mouse is obtained by comprehensively considering the order of eating and drinking of the mouse. By analyzing the timing behavior, we can understand how the social rank of the mouse is formed.

[0099] In this embodiment, simply putting the mouse into the experimental box allows for a completely objective measurement of the mouse's social rank, saving time and effort. It also combines multiple methods for determining the mouse's social rank, making the experimental results more reliable.

[0100] The device described in this application can perform long-term tracking and analysis in a semi-natural environment to observe changes in the social behavior of mouse populations. This long-term observation can provide more comprehensive data, helping researchers better understand the formation and changes in social hierarchies. Moreover, by employing multiple experimental paradigms to ensure stable and reliable hierarchical results, and through long-term automatic recording, it is also possible to study the formation process of hierarchies in mouse populations and how this information is encoded.

[0101] In an optional embodiment, the experimental chamber also includes a spare experimental area;

[0102] The radio frequency identification location tracking system also includes a backup experimental area tracking module, which includes at least one fifth radio frequency antenna located in the backup experimental area.

[0103] The behavior analysis system in this embodiment, as shown in Figure 3, uses the SimBA classifier to allow users to create analysis templates containing regions of interest, and can freely select, define, and modify analysis behaviors.

[0104] For example, to investigate a mouse's cognitive abilities, a new object can be placed in the experimental chamber for object recognition testing. Using SimBA to select a backup experimental area allows for the recording of time and frequency of the animal's movement within that area. The mouse's cognitive abilities can be assessed based on its daily activity time and number of entries into that area. Similarly, a balance bar can be placed to automatically detect the time spent and assess the mouse's motor skills. In short, different experimental modules can be added for different experimental tasks to automatically analyze all mouse behaviors.

[0105] The ability to flexibly define behaviors provides complete control over the scope of measurement and allows for addressing multiple questions about social and non-social behaviors arising from a single behavioral analysis. This not only aligns with the 3R principle (replace, refine, reduce) but also saves resources and time compared to meticulously conducting multiple short-term experiments under artificial environmental conditions.

[0106] Based on the same principle as the method provided in the embodiments of this application, the embodiments of this application also provide a method for detecting social hierarchy and social behavior in mice, as shown in Figure 4. The method includes:

[0107] Step 401: Receive mouse identification information from the RFID chip attached to the mouse via the RFID location tracking system, and send the mouse identification information to the behavior analysis system.

[0108] At step 402, a video of the mice is captured by an infrared camera system, and the video of the mice is sent to a behavior analysis system.

[0109] At step 403, based on the behavior analysis system, a preset computer vision and machine learning algorithm is used to analyze the mouse identity information and the video of the mice, to obtain the social rank and social behavior of each mouse.

[0110] In the embodiments of the present application, an automatic behavior tracking technology is used, multiple mice are placed in an experimental box, and a semi-natural environment is provided. At the group level, based on a radio frequency identification chip fixed on the mouse and a radio frequency identification position tracking system placed at the bottom of the experimental box, the activities or movements of each mouse in the semi-natural environment are identified and tracked in real time. Through an infrared camera system, a video of the mice is captured, and a computer vision and machine learning algorithm is used to automatically identify and analyze the movement phenotype and interactive behavior (such as social, competitive and submissive behavior) of each mouse. Moreover, long-term tracking analysis in the semi-natural environment is performed to observe the changes in the social behavior of the mouse group. This long-term observation can provide more comprehensive data to help researchers better quantify the social behavior of the mice in the group, explore the social memory of the mice, and understand the process of establishing the social rank of the mice. The social rank of the mice in the group is automatically analyzed.

[0111] In an optional embodiment, based on the behavior analysis system, a preset computer vision and machine learning algorithm is used to analyze the mouse identity information and the video of the mice, to obtain the social rank and social behavior of each mouse, including:

[0112] Receiving and storing the mouse identity information and the video of the mice;

[0113] Synchronizing the mouse identity information with the video of the mice;

[0114] Labeling the features of the mice in each frame of the video of the mice to generate feature data;

[0115] Identifying the social behavior and social rank of the mice according to the feature data.

[0116] In an optional embodiment, synchronizing the mouse identity information with the video of the mice includes:

[0117] Frame extraction processing is performed on the video of the mice to obtain multiple image frames;

[0118] Obtaining the mouse identity information at the same time point for each image frame, and synchronizing the mouse identity information with the mice in the image frame.

[0119] In an optional embodiment, identifying the social behavior and social rank of the mice according to the feature data includes:

[0120] According to the mouse identity information and the corresponding feature data, social behavior of the mouse is recognized;

[0121] The feeding and water drinking sequence of the mouse is determined, and the social rank of the mouse is determined according to the feeding and water drinking sequence and the social behavior of the mouse.

[0122] The method for detecting the social rank and the social behavior of the mouse provided in the embodiments of the present application can realize the functions of the device embodiments of FIGS. 1 to 3, and thus the detailed description is omitted here.

[0123] The method for detecting the social rank and the social behavior of the mouse provided in the embodiments of the present application can realize the functions of the device for detecting the social rank and the social behavior of the mouse provided in the embodiments of the present application, and the implementation principles are similar. The steps in the method for detecting the social rank and the social behavior of the mouse in the embodiments of the present application are corresponding to the actions performed by the modules and units in the device for detecting the social rank and the social behavior of the mouse. The detailed function description of the steps in the method for detecting the social rank and the social behavior of the mouse can be referred to the description of the corresponding device for detecting the social rank and the social behavior of the mouse provided in the foregoing embodiments, and thus the detailed description is omitted here.

[0124] The above description is merely preferred embodiments of the present application and a description of the principles of the applied technology. It should be understood by those skilled in the art that the disclosed range of the present application is not limited to the technical solutions formed by the specific combinations of the technical features described above, and also covers other technical solutions formed by any combinations of the technical features or equivalent features without departing from the disclosed concept. For example, the technical solutions formed by the mutual replacement of the above-described features and the technical features disclosed in the present application (but not limited to) having similar functions.

Claims

1. An apparatus for detecting social rank and social behavior of mice, characterized by, The device comprises an experimental box in which multiple mice are placed, a radio frequency identification position tracking system arranged at the bottom of the experimental box, a signal input and output control module, and a behavior analysis system; The mice are fixed with radio frequency identification chips; The radio frequency identification position tracking system is used for receiving mouse identity information sent by the radio frequency identification chip; The signal input and output control module comprises an infrared camera system, which is used for shooting mouse videos; The behavior analysis system is in communication connection with the radio frequency identification position tracking system and the infrared camera system, and is used for analyzing the mouse identity information and the mouse videos by using preset computer vision and machine learning algorithms to obtain the social rank and social behavior of each mouse.

2. The apparatus for detecting social rank and social behavior of mice according to claim 1, wherein The behavior analysis system comprises: A software record storage module is used for receiving and storing the mouse identity information and the mouse videos; A time matching server software is used for synchronizing the mouse identity information with the mouse videos; A video processing module is used for labeling the features of mice in the feature frames extracted from the mouse videos to generate feature data; A behavior analysis module is used for identifying the social behavior and the social rank of mice according to the feature data.

3. The apparatus for detecting social rank and social behavior of mice according to claim 2, wherein The time matching server software comprises: A video frame extraction unit is used for frame extraction processing of the mouse videos to obtain multiple image frames; A position synchronization unit is used for obtaining the mouse identity information at the same time point of each image frame, and synchronizing the mouse identity information with the mice in the image frames.

4. The apparatus for detecting social rank and social behavior of mice according to claim 2, wherein The behavior analysis module comprises: A behavior recognition unit is used for identifying the social behavior of mice according to the mouse identity information and the corresponding feature data; A rank sorting unit is used for determining the feeding and water feeding sequence of mice, and determining the social rank of mice according to the feeding and water feeding sequence of mice and the social behavior.

5. The apparatus for detecting social hierarchy and social behavior in mice according to claim 1, characterized in that, The experimental box comprises a competition arena, a mouse escape area, a food delivery device, and a water delivery device; The radio frequency identification position tracking system comprises: A competition arena tracking module comprises a radio frequency identification array composed of multiple first radio frequency antennas; the radio frequency identification array is arranged on the bottom of the competition arena; A mouse escape area tracking module comprises at least one second radio frequency antenna arranged at the entrance of the mouse escape area; A food delivery device tracking module comprises at least one third radio frequency antenna arranged in the food delivery device; A water delivery device tracking module comprises at least one fourth radio frequency antenna arranged in the water delivery device; The radio frequency identification position tracking system further comprises a receiver, and the first radio frequency antenna, the second radio frequency antenna, the third radio frequency antenna, and the fourth radio frequency antenna are connected to the corresponding receiver.

6. The apparatus for detecting social rank and social behavior of mice according to claim 5, wherein The experimental box further comprises a backup experimental area; The radio frequency identification position tracking system further comprises a backup experimental area tracking module, and the experimental area tracking module comprises at least one fifth radio frequency antenna arranged in the backup experimental area.

7. A method of detecting social rank and social behavior of mice based on the device of any one of claims 1-6, characterized in that, The method comprises: Receiving mouse identity information sent by a radio frequency identification chip fixed on the mouse through a radio frequency identification position tracking system, and sending the mouse identity information to a behavior analysis system; Shooting mouse video by using an infrared camera system, and sending the mouse video to the behavior analysis system; Based on the behavior analysis system, using a preset computer vision and machine learning algorithm to analyze the mouse identity information and the mouse video, obtaining the social rank and social behavior of each mouse.

8. The method of detecting social rank and social behavior in mice according to claim 7, wherein, The method based on the behavior analysis system, using a preset computer vision and machine learning algorithm to analyze the mouse identity information and the mouse video, obtaining the social rank and social behavior of each mouse, includes: Receiving and storing the mouse identity information and the mouse video; Synchronizing the mouse identity information with the mouse video; Labeling the features of the mouse in each frame of the mouse video to generate feature data; According to the feature data, identifying the social behavior and the social rank of the mouse.

9. The method of detecting social rank and social behavior in mice according to claim 8, wherein, The method of synchronizing the mouse identity information with the mouse video includes: Frame extraction processing the mouse video to obtain multiple image frames; Obtaining the mouse identity information at the same time point of each image frame, and synchronizing the mouse identity information with the mouse in the image frame.

10. The method for detecting social rank and social behavior of mice according to claim 8, wherein, The method of identifying the social behavior and the social rank of the mouse according to the feature data includes: According to the mouse identity information and the corresponding feature data, identifying the social behavior of the mouse; Determining the feeding and water feeding sequence of the mouse, and determining the social rank of the mouse according to the feeding and water feeding sequence of the mouse and the social behavior.

Images