A new object recognition experimental island intelligent monitoring device

Abstract

This invention discloses a novel intelligent monitoring device for an experimental island used for object recognition, belonging to the field of experimental animal behavioral monitoring technology. It includes an environmental control module housed within the experimental island enclosure to regulate temperature, humidity, light intensity, and odor levels, maintaining environmental stability. A behavior acquisition module, also housed within the island, collects data on the experimental animals' exploratory behavior, physiological correlations, and object interactions during the experiment. A data processing module, communicatively connected to the behavior acquisition module, receives these data, performs fusion analysis based on a preset algorithm, calculates a recognition index, and generates a multi-dimensional experimental report. This invention, from environmental control and stimulus presentation to data acquisition and analysis, minimizes human error and subjective bias, ensuring the reproducibility and cross-laboratory comparability of experimental results.

Description

Technical Field

[0001] This invention belongs to the field of experimental monitoring technology for experimental animal behavior, and specifically relates to a novel intelligent monitoring device for an experimental island for object recognition. Background Technology

[0002] The novel object recognition (SAE) test is a classic behavioral experimental method for assessing animal learning and memory abilities. Its core principle is based on an animal's preference for exploring new objects. By statistically analyzing the exploration time of animals for both new and old objects and calculating a recognition index, the test reflects their cognitive function status. This test is widely used in neurobiology, pharmacology, psychology, and other fields. In particular, in SAE research, it is a key experimental tool for detecting hippocampal neuronal damage and cognitive impairment in mice.

[0003] Traditional novel object recognition experiments suffer from several technical shortcomings: First, the experimental environment is easily affected by factors such as temperature, humidity, light, and residual odors, leading to poor repeatability of experimental results. Second, behavioral data collection relies on manual observation, and the experimenter's criteria for judging "exploratory behavior" vary subjectively, making it impossible to capture subtle movements and physiological changes in mice, resulting in low data accuracy. Third, the data collection dimension is singular, only counting exploration time, lacking simultaneous capture of movement state, interaction intensity, and physiological correlation data, making it difficult to comprehensively reflect the animal's cognitive process. Fourth, the experimental process requires full-time human supervision, which is cumbersome, inefficient, and cannot meet the needs of large-scale, long-term experiments.

[0004] While some existing automated monitoring devices have achieved image acquisition and time statistics, they still suffer from problems such as insufficient positioning accuracy, inaccurate action recognition, and limited data dimensions, making it difficult to meet the high-precision and multi-dimensional requirements for cognitive function assessment in SAE research. Summary of the Invention

[0005] To solve the above problems, the present invention adopts the following technical solution: A novel intelligent monitoring device for an experimental island for object recognition includes: The experimental island enclosure is designed to provide a standardized experimental space for new object recognition. An environmental control module is installed inside the experimental island chamber to control the temperature, humidity, light intensity, and odor environment inside the experimental island chamber, thereby maintaining the stability of the experimental environment. A behavior acquisition module is installed inside the experimental island box and is used to collect exploratory behavior data, physiological correlation data and object interaction data of the experimental animals during the experiment. The data processing module is communicatively connected to the behavior acquisition module. It is used to receive the exploration behavior data, physiological correlation data and object interaction data, perform fusion analysis on the data based on a preset algorithm, calculate the recognition index, and generate a multi-dimensional experimental report. The remote control module is communicatively connected to the environmental control module, the behavior acquisition module, and the data processing module, and is used to remotely monitor the experimental process, adjust experimental parameters, and receive experimental data and experimental reports output by the data processing module.

[0006] Furthermore, the interior of the experimental island enclosure is also equipped with an object placement area, which can selectively place two identical old objects or one old object and one new object.

[0007] Furthermore, an object fixing seat is provided in the object placement area, the object fixing seat including a base, a height adjustment rod and a clamping head for fixing different experimental objects; The base is detachably connected to the bottom of the experimental island box; One end of the height adjustment rod is connected to the base, and the other end is connected to the clamping head.

[0008] Furthermore, the environmental control module includes: A temperature and humidity control unit is installed inside the experimental island chamber to maintain a constant ambient temperature and humidity through closed-loop feedback. A programmable lighting unit is installed on the top wall of the experimental island enclosure to adjust the light intensity and color temperature; Odor removal and standardization unit, which is installed on the top wall of the experimental island box, is used to automatically spray volatile solvents and ventilate during experimental intervals to eliminate residual odors.

[0009] Furthermore, the behavior acquisition module includes: The image acquisition unit is used to acquire video data containing experimental animals and experimental objects; The position tracking and motion state detection unit calculates the kinematic parameters of the experimental animal and identifies its behavioral posture based on the video data. An object interaction sensing unit is located in the object placement area and is used to detect contact interaction signals between the experimental animal and the experimental object. A biosignal synchronization unit is installed on the experimental animal to collect physiological signals from the experimental animal. The exploratory behavior assessment unit is used to comprehensively determine whether an animal has conducted an effective exploration of the experimental object.

[0010] Furthermore, the exploration behavior data includes object exploration time, exploration path, dwell area, movement speed, turning frequency, and posture changes; The physiological correlation data includes heart rate variability, and the object interaction data includes contact duration, contact force, and number of contact times.

[0011] Furthermore, the data processing module includes: A receiving unit, connected to the behavior acquisition module, is used to receive data information from the behavior acquisition module and perform high-precision time alignment on the data information stream of the behavior acquisition module; A feature extraction unit, connected to the receiving unit, is used to extract multi-dimensional features from synchronized data information, including motion trajectory features, object exploration features, behavior classification features, physiological signal features, and object interaction features. A fusion analysis unit, connected to the feature extraction unit, is used to perform correlation analysis on the multi-dimensional features and calculate the recognition index; The report generation unit is connected to the fusion analysis unit and the remote control module, respectively, and is used to automatically generate a multi-dimensional experimental report including behavioral trajectory map, exploration activity heat map, multi-signal time series spectrum map and feature comparison chart.

[0012] Furthermore, the fusion analysis unit includes: The model training subunit constructs and updates a multi-dimensional feature neural network model, which takes the multi-dimensional features as input and animal cognitive state assessment or abnormal behavior recognition as output. The feature analysis subunit is connected to the feature extraction unit, the model training subunit, and the report generation unit. It is used to process the fused feature vectors of real-time acquisition or historical experiments by calling the multi-dimensional feature neural network model trained by the model training subunit, and output cognitive function scores, cognitive state classifications, or early pathological risk probabilities.

[0013] Beneficial effects: 1. Full-process standardization and automation: From environmental control and stimulus presentation to data collection, analysis and reporting, human error and subjective bias are minimized, ensuring the reproducibility and cross-laboratory comparability of experimental results.

[0014] 2. Multidimensional and objective data collection: The behavioral analysis is expanded from a single exploration time to a three-dimensional dimension of movement path, fine interaction and physiological state, directly quantifying indicators such as interaction intensity, resulting in richer and more objective data.

[0015] 3. Data fusion and intelligent analysis: By introducing multimodal data synchronization, it is possible to reveal the complex relationship between behavioral characteristics and physiological state, and provide a deeper level of cognitive function assessment, which surpasses traditional statistical analysis. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of the experimental island box of the present invention; Explanation of reference numerals in the attached diagram: 1. Experimental island enclosure; 2. High-speed macro camera; 3. Programmable illumination unit; 4. Temperature and humidity control unit; 5. Odor removal and standardization unit; 6. Object holder; 7. Data processing module; 8. Object interaction sensing unit. Detailed Implementation

[0017] This section will describe in detail specific embodiments of the present invention. Preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and overall technical solution of the present invention, but they should not be construed as limiting the scope of protection of the present invention.

[0018] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0019] In the description of this invention, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0020] In the description of this invention, unless otherwise explicitly defined, terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0021] Example 1

[0022] refer to Figure 1 A novel intelligent monitoring device for an experimental island for object recognition includes: Experimental island enclosure 1 is used to provide a standardized experimental space for new object recognition. An environmental control module is installed inside the experimental island box 1 to control the temperature, humidity, light intensity, and odor environment inside the experimental island box 1, thereby maintaining the stability of the experimental environment. The behavior acquisition module is set up inside the experimental island box 1 and is used to collect data on the experimental animals' exploratory behavior, physiological correlation data and object interaction data during the experiment. Data processing module 7 is connected to behavior acquisition module and is used to receive exploration behavior data, physiological correlation data and object interaction data. Based on preset algorithm, it performs fusion analysis and processing on the data, calculates recognition index and generates multi-dimensional experimental report. The remote control module is connected to the environmental control module, behavior acquisition module, and data processing module 7 respectively. It is used to remotely monitor the experimental process, adjust experimental parameters, and receive experimental data and reports output by the data processing module.

[0023] Preferably, the experimental island box 1 is also provided with an object placement area, which can selectively place two identical old objects or one old object and one new object.

[0024] Preferably, an object fixing seat 6 is provided in the object placement area. The object fixing seat 6 includes a base, a height adjustment rod, and a clamping head for fixing different experimental objects. The base is detachably connected to the bottom of the experimental island box 1; One end of the height adjustment rod is connected to the base, and the other end is connected to the clamping head.

[0025] In this embodiment, the experimental island box 1 is a cubic structure with an open top, the inner wall is made of matte white polytetrafluoroethylene, and the side wall is equipped with hinged doors.

[0026] Preferably, the environmental control module includes: Temperature and humidity control unit 4 is installed inside the experimental island box 1 and is used to maintain a constant ambient temperature and humidity through closed-loop feedback. Programmable lighting unit 3 is installed on the top wall of the experimental island box 1 and is used to adjust the light intensity and color temperature. Odor removal and standardization unit 5 is installed on the top wall of the experimental island box 1. It is used to automatically spray volatile solvents and ventilate during experimental breaks to eliminate residual odors.

[0027] In this embodiment, the temperature and humidity control unit 4 maintains a constant ambient temperature and humidity through closed-loop feedback between the semiconductor cooling chip / heating wire and the humidity sensor.

[0028] The programmable lighting unit 3 uses an LED surface light source, which is located on the top of the experimental island box 1. The light intensity, color temperature and brightness cycle can be adjusted programmably.

[0029] Odor Removal and Standardization Unit 5 includes an air intake filtration system, an internal circulation fan, and a controllable spray device located in the corner. It can automatically spray 75% alcohol mist and exhaust air during experimental breaks to eliminate animal-residual odors and achieve odor environment standardization.

[0030] Preferably, the behavior acquisition module includes: The image acquisition unit is used to acquire video data containing experimental animals and experimental objects; The position tracking and motion state detection unit calculates the kinematic parameters of experimental animals and identifies their behavioral postures based on video data; The object interaction sensing unit 8 is located in the object placement area and is used to detect the contact interaction signals between the experimental animal and the experimental object. A biosignal synchronization unit is installed on laboratory animals to collect their physiological signals. The exploratory behavior assessment unit is used to comprehensively determine whether an animal has conducted an effective exploration of the experimental object.

[0031] In this embodiment, the image acquisition unit includes at least one top-mounted high-definition global camera and one side-mounted high-speed macro camera 2, which are used for panoramic tracking and capturing details of animal-object interaction, respectively.

[0032] The position tracking and motion state detection unit uses visual data from the image acquisition unit to calculate the animal's center coordinates, movement speed, acceleration, turning angular velocity, and body orientation in real time through a deep learning model, and identifies behaviors such as running, pacing, sniffing, standing, and grooming.

[0033] The biosignal synchronization unit includes a miniature wireless physiological signal acquisition device and a signal demodulation module. The miniature wireless physiological signal acquisition device adopts a lightweight design and is fixed to the chest of a mouse with a flexible strap. It integrates a heart rate sensor and a respiratory sensor to collect heart rate, heart rate variability, and respiratory rate data of the mouse during exploration. The signal demodulation module is connected to the physiological signal acquisition device via wireless communication and performs filtering, amplification, and demodulation processing on the raw physiological signals to ensure the stability and accuracy of data transmission and to achieve timestamp synchronization between behavioral data and physiological data.

[0034] The object interaction sensing unit 8 includes a miniature pressure sensor, a vibration sensor, and a proximity sensor disposed inside the experimental object; the miniature pressure sensor is used to detect the pressure when the mouse comes into contact with the object; the vibration sensor is used to capture the vibration signal generated by the mouse touching the object; the proximity sensor adopts the capacitive sensing principle to determine the proximity state between the mouse and the object; wherein, the data sampling rate of the object interaction sensing unit is synchronized with the image acquisition unit to ensure the temporal consistency between interactive behavior and visual data.

[0035] The behavior determination unit is integrated into the edge computing unit of the data processing module 7. It uses a multimodal temporal fusion neural network to perform fusion analysis on real-time data from the image acquisition unit, the position tracking and motion state detection unit, the object interaction perception unit 8, and the biological signal synchronization unit.

[0036] The inquiry behavior assessment unit includes: The head orientation analysis subunit extracts head posture features based on animal head images acquired by the lateral high-speed macro camera 2, and combines them with an eye localization algorithm to determine whether the gaze direction is towards the experimental object. The motion trajectory and dwelling pattern analysis subunit receives motion trajectory data output by the position tracking unit and calculates the animal's dwell time near the object, the rate of change of movement speed, and the trajectory circumference index. The posture recognition and behavior classification subunit calls the behavior posture recognition model to determine whether the approach is accompanied by typical exploration postures such as sniffing, touching, or standing. The physiological signal co-analysis subunit synchronously analyzes the heart rate variability and respiratory rate data transmitted by the biological signal synchronization unit to identify transient physiological arousal patterns that often accompany exploratory behaviors. The timing and persistence determination subunit sets the minimum effective investigation duration threshold to 0.5 seconds. Combining the consistency of the above multimodal features in the time series, it outputs a binary determination result (effective investigation / non-investigation) and generates an investigation event sequence with timestamps and confidence scores.

[0037] Preferably, the exploration behavior data includes object exploration time, exploration path, dwell area, movement speed, turning frequency, and posture changes; Physiological correlation data includes heart rate variability, and object interaction data includes contact duration, contact force, and number of contact times.

[0038] Preferably, the data processing module includes: The receiving unit is connected to the behavior acquisition module and is used to receive data information from the behavior acquisition module and perform high-precision time alignment on the data information stream of the behavior acquisition module. The feature extraction unit is connected to the receiving unit and is used to extract multi-dimensional features from the synchronized data information, including motion trajectory features, object exploration features, behavior classification features, physiological signal features and object interaction features. The fusion analysis unit, connected to the feature extraction unit, is used to perform correlation analysis on multi-dimensional features and calculate the recognition index. The report generation unit is connected to the fusion analysis unit and the remote control module, respectively, and is used to automatically generate multi-dimensional experimental reports that include behavioral trajectory maps, exploration activity heatmaps, multi-signal time series spectra, and feature comparison charts.

[0039] Preferably, the fusion analysis unit includes: The model training subunit constructs and updates the multi-dimensional feature neural network model. The multi-dimensional feature neural network model takes multi-dimensional features as input and animal cognitive state assessment or abnormal behavior recognition as output. The feature analysis subunit is connected to the feature extraction subunit, the model training subunit, and the report generation subunit. It is used to process the fused feature vectors of real-time acquisition or historical experiments by calling the multi-dimensional feature neural network model trained by the model training subunit, and output cognitive function scores, cognitive state classifications, or early pathological risk probabilities.

[0040] Example 2

[0041] This embodiment presents a novel object recognition experimental method for the device described in Embodiment 1, including the following steps: S1: The environmental control module adjusts and stabilizes the internal environmental parameters of the experimental island to a set value. S2: Place two identical objects in the object placement area, place the experimental animal in the object, and begin the first experimental phase; S3: Simultaneously collect multimodal behavioral and physiological data of the animals during the first experimental phase through the behavior acquisition module; S4: Replace one of the identical objects with a new object, and activate the odor removal and standardization unit to reset the environment; S5: Place the same animal back in and begin the second experimental phase, and collect data from the second experimental phase synchronously through the behavior acquisition module; S6: The data processing module processes and analyzes the data from the two stages to generate an experimental report; wherein, the analysis includes: extracting multi-dimensional features and inputting the multi-dimensional features into a pre-trained neural network model to obtain the cognitive assessment results after fusion analysis.

[0042] Example 3

[0043] This embodiment is a further modification based on embodiment 2.

[0044] The multi-dimensional features include a sequence of valid exploration events and their confidence scores output by the exploration behavior determination subunit. This sequence is used to replace traditional exploration time statistics, improving the accuracy and signal-to-noise ratio of the identification index.

[0045] Preferably, the new object recognition experimental method also includes a multi-dimensional feature neural network model training method: S01. Obtain historical experimental datasets. The datasets contain feature vectors extracted from multiple animals under multimodal data, as well as cognitive state labels or objective physiological indicator labels corresponding to the feature vectors. S02. Supervised training and validation of the initial neural network model are performed using historical experimental datasets to obtain a trained neural network model, which is then deployed in the fusion analysis submodule of the data processing module.

[0046] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the technical scope of the present invention. Therefore, any minor modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A novel intelligent monitoring device for an experimental island for object recognition, characterized in that, include: The experimental island enclosure is designed to provide a standardized experimental space for new object recognition. An environmental control module is installed inside the experimental island chamber to control the temperature, humidity, light intensity, and odor environment inside the experimental island chamber, thereby maintaining the stability of the experimental environment. A behavior acquisition module is installed inside the experimental island box and is used to collect exploratory behavior data, physiological correlation data and object interaction data of the experimental animals during the experiment. The data processing module is communicatively connected to the behavior acquisition module. It is used to receive the exploration behavior data, physiological correlation data and object interaction data, perform fusion analysis on the data based on a preset algorithm, calculate the recognition index, and generate a multi-dimensional experimental report. The remote control module is communicatively connected to the environmental control module, the behavior acquisition module, and the data processing module, and is used to remotely monitor the experimental process, adjust experimental parameters, and receive experimental data and experimental reports output by the data processing module.

2. The novel object recognition experimental island intelligent monitoring device according to claim 1, characterized in that, The experimental island enclosure also has an object placement area, which can selectively place two identical old objects or one old object and one new object.

3. The novel object recognition experimental island intelligent monitoring device according to claim 2, characterized in that, The object placement area is provided with an object fixing seat, which includes a base, a height adjustment rod, and a clamping head for fixing different experimental objects. The base is detachably connected to the bottom of the experimental island box; One end of the height adjustment rod is connected to the base, and the other end is connected to the clamping head.

4. The novel object recognition experimental island intelligent monitoring device according to claim 1, characterized in that, The environmental control module includes: A temperature and humidity control unit is installed inside the experimental island chamber to maintain a constant ambient temperature and humidity through closed-loop feedback. A programmable lighting unit is installed on the top wall of the experimental island enclosure to adjust the light intensity and color temperature; Odor removal and standardization unit, which is installed on the top wall of the experimental island box, is used to automatically spray volatile solvents and ventilate during experimental intervals to eliminate residual odors.

5. The novel intelligent monitoring device for an experimental island for object recognition according to claim 3, characterized in that, The behavior acquisition module includes: The image acquisition unit is used to acquire video data containing experimental animals and experimental objects; The position tracking and motion state detection unit calculates the kinematic parameters of the experimental animal and identifies its behavioral posture based on the video data. An object interaction sensing unit is located in the object placement area and is used to detect contact interaction signals between the experimental animal and the experimental object. A biosignal synchronization unit is installed on the experimental animal to collect physiological signals from the experimental animal. The exploratory behavior assessment unit is used to comprehensively determine whether an animal has conducted an effective exploration of the experimental object.

6. The novel object recognition experimental island intelligent monitoring device according to claim 5, characterized in that, The exploration behavior data includes object exploration time, exploration path, dwell area, movement speed, turning frequency, and posture changes; The physiological correlation data includes heart rate variability, and the object interaction data includes contact duration, contact force, and number of contact times.

7. The novel object recognition experimental island intelligent monitoring device according to claim 1, characterized in that, The data processing module includes: A receiving unit, connected to the behavior acquisition module, is used to receive data information from the behavior acquisition module and perform high-precision time alignment on the data information stream of the behavior acquisition module; A feature extraction unit, connected to the receiving unit, is used to extract multi-dimensional features from synchronized data information, including motion trajectory features, object exploration features, behavior classification features, physiological signal features, and object interaction features. A fusion analysis unit, connected to the feature extraction unit, is used to perform correlation analysis on the multi-dimensional features and calculate the recognition index; The report generation unit is connected to the fusion analysis unit and the remote control module, respectively, and is used to automatically generate a multi-dimensional experimental report including behavioral trajectory map, exploration activity heat map, multi-signal time series spectrum map and feature comparison chart.

8. The novel object recognition experimental island intelligent monitoring device according to claim 7, characterized in that, The fusion analysis unit includes: The model training subunit constructs and updates a multi-dimensional feature neural network model, which takes the multi-dimensional features as input and animal cognitive state assessment or abnormal behavior recognition as output. The feature analysis subunit is connected to the feature extraction unit, the model training subunit, and the report generation unit. It is used to process the fused feature vectors of real-time acquisition or historical experiments by calling the multi-dimensional feature neural network model trained by the model training subunit, and output cognitive function scores, cognitive state classifications, or early pathological risk probabilities.

Images