Animal experiment flexible intelligent sensing monitoring system and monitoring method

Abstract

The invention discloses an animal experiment flexible intelligent sensing monitoring system, which comprises a data acquisition module, a data conversion and transmission module and a data processing module, and is characterized in that the data acquisition module acquires capacitance value changes related to animal activities and outputs analog signals to the data conversion and transmission module; the data conversion and transmission module converts the received signals into digital signals and transmits the digital signals to the data processing module, the data processing module processes the received digital signals, the trained deep learning model is utilized to identify animal tracks and behavior categories, analysis is carried out corresponding to animal brain wave signals at the moment, and the animal brain wave signals at the moment are acquired. And the understanding of animal behaviors is enhanced. According to the flexible intelligent sensing monitoring system and monitoring method for the animal experiment, the animal pressure data can be accurately detected, the action track can be analyzed, the accuracy is high, the stability is good, the abrasion resistance is achieved, and the whole system is accurate in data, stable in transmission, high in portability, low in cost and not limited by the territory environment.

Description

Technical Field

[0001] The present invention relates to the field of brain-computer interface technology, and in particular to an animal experiment flexible intelligent sensor monitoring system and a monitoring method. Background Art

[0002] With the development of brain-computer interface (BCI) technology, more and more research has begun to focus on how to explore the interaction mechanism between the human brain and external devices through animal experiments. In this field, it is crucial to obtain accurate biological signals and environmental data, especially when analyzing the relationship between animal behavior and neural signals. Traditional monitoring methods are often limited to a single physiological signal and lack a comprehensive understanding of the behavior of animals in a specific environment. Therefore, a new monitoring solution is urgently needed.

[0003] Flexible sensors have gradually become a hot research direction in biological signal monitoring in recent years due to their superior adaptability and comfort. Compared with traditional hard sensors, flexible sensors can better fit the surface of organisms and will not interfere with the normal behavior of animals. In addition, ion-type flexible pressure-capacitive sensors still maintain high stability under changes in temperature and humidity, and are not affected by sweating and excretion of organisms, which makes them perform well in complex environments and effectively capture the pressure changes and behavioral trajectories of animals.

[0004] There are currently two implementation options:

[0005] 1. Sensor solution

[0006] Piezoelectric sensors use the properties of piezoelectric materials to construct a piezoelectric sensor array for monitoring the activities of small animals in a specific area. The sensor can sense the weight changes of animals in real time, and then infer their movement trajectory and behavior patterns. However, piezoelectric sensors are sensitive to humidity changes, which limits their application in different experimental environments. Although piezoresistive sensors can provide higher sensitivity, their material properties limit their application in high temperature or humid environments. These sensors are easily damaged during long-term use, affecting the stability and reliability of the data.

[0007] 2. Visual Solution

[0008] Among the existing technical solutions, cameras are also widely used for animal behavior monitoring. Through image processing technology, cameras can record animal activities in real time. However, cameras have some obvious limitations. First, the viewing angle of the camera makes it impossible to fully capture the movement trajectory of the animal, especially in complex environments, where obstructions will cause data loss. Secondly, the camera cannot directly provide pressure data exerted by animals, which is crucial for studying the relationship between animal behavior and physiological signals. Thirdly, the huge amount of image data leads to high storage and processing requirements, which increases the complexity of the system. Building an effective image analysis model requires high computing power, and the complexity of the model may affect real-time performance. Summary of the invention

[0009] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is that the visual solutions existing in the prior art are more sensitive to the environment (angle, light), especially in complex environments, obstructions will cause data loss. At the same time, the amount of image data is huge, resulting in high storage and processing requirements, increasing the complexity of the system. Building an effective image analysis model requires high computing power, and the complexity of the model may affect the real-time performance; the sensor solution has the disadvantages of being sensitive to temperature and humidity changes, not wear-resistant, and high cost. Piezoelectric sensors and piezoresistive sensors are sensitive to humidity changes, which limits their application in animal experimental environments. At the same time, these sensors are easily damaged during long-term use, affecting the stability of the data and the reliability of the system. The present invention provides an animal experiment flexible intelligent sensing monitoring system and monitoring method, which can accurately detect animal pressure data and analyze movement trajectories. The flexible pressure-capacitive sensor system has the advantages of high accuracy, good stability, and wear resistance. The entire system has accurate data, stable transmission, strong portability, low cost, and is not restricted by the geographical environment.

[0010] To achieve the above-mentioned objectives, the present invention provides a flexible intelligent sensing monitoring system for animal experiments, including a data acquisition module, a data conversion and transmission module, and a data processing module. The data acquisition module collects changes in capacitance values ​​related to animal activities and outputs analog signals to the data conversion and transmission module. The data conversion and transmission module converts the received signals into digital signals and transmits them to the data processing module. The data processing module processes the received digital signals and uses a trained deep learning model to identify animal trajectories and behavior categories, and analyzes the corresponding brain wave signals of the animals at that moment to enhance the understanding of animal behavior.

[0011] Furthermore, the data acquisition module is a flexible I-V capacitance sensor array, including m sensors arranged in an n×n matrix.

[0012] Furthermore, the size of the sensor is 2 cm×2 cm.

[0013] Further, the distance between sensors is set to 5-10 mm.

[0014] Furthermore, the data conversion and transmission module includes a data conversion unit and a transmission unit. The data conversion unit includes a 4-channel high-precision capacitance sensor detection chip and a microcontroller. The capacitance sensor detection chip converts the measured capacitance value change into a digital signal and transmits it to the microcontroller through I2C, and then the microcontroller sends the data to the transmission unit through UART.

[0015] Furthermore, the data conversion and transmission module includes a low-power Bluetooth microcontroller, which transmits data to the data processing module via the low-power Bluetooth.

[0016] Furthermore, the transmission protocol of the data conversion and transmission module supports BT 5.0 / 5.1 standards, and the transmission rate supports 125Kbps / 500Kbps / 1Mbps / 2Mbps.

[0017] Another embodiment of the present invention provides an animal experiment flexible intelligent sensing monitoring method, comprising the following steps:

[0018] Before starting the experiment, make sure all devices are connected correctly;

[0019] After ensuring that the devices are connected correctly, start them and check their working status to ensure that they can receive and process signals normally;

[0020] Place the experimental animal on the sensor array and ensure that the experimental animal can move freely. At this time, the system begins to collect the pressure data applied by the experimental animal on the sensor in real time and outputs analog signals related to these pressure changes;

[0021] Then the capacitive sensor detection chip converts the received analog signal into a digital signal and transmits it to the microcontroller through the I2C protocol. After receiving the digital signal, the microcontroller sends it to the low-power Bluetooth microcontroller through the UART protocol, and finally transmits it to the data processing module;

[0022] The data processing module processes the received digital signals and uses the trained deep learning model to identify the animal's trajectory and behavior category, and analyzes the animal's brain wave signal at that moment to enhance the understanding of the animal's behavior; and visualizes the results;

[0023] The system saves the analysis results and visualization data to the database for subsequent review and analysis;

[0024] After the final experiment is completed, turn off all equipment.

[0025] Furthermore, the results are visualized, specifically using squares to represent the flexible I-V capacitive sensor array, different levels of pressure values ​​are represented by different colors, and an n×n matrix is ​​used to calculate the relative position of each sensor.

[0026] Furthermore, the capacitive sensor detection chip converts the received analog signal into a digital signal and transmits it to the microcontroller through the I2C protocol. After receiving the digital signal, the microcontroller sends it to the low-power Bluetooth microcontroller through the UART protocol, and finally transmits it to the data processing module, which specifically includes:

[0027] The self-calibration circuit inside the high-precision capacitive sensor detection chip eliminates the fixed distance error at the sensor end and supports two-way signal input. The differential input mode of the two-way signal offsets the parasitic capacitance of the external environment and the drift of temperature. Finally, the analog is quantized into a numerical value through a high-performance ADC, and then the DSP is used to convert the raw data converted from the ADC into a signal C' that filters out the environmental noise.

[0028] Technical Effects

[0029] The present invention discloses a flexible intelligent sensing monitoring system and method for animal experiments. The system adopts a mat on which a flexible pressure-capacitive sensor array is arranged. The system can densely cover the animal activity area, monitor the pressure changes applied by the animal in real time, ensure the comprehensive capture of the animal behavior, accurately record its movement trajectory, and solve the problem of small detection range of single-point sensors. The use of flexible sensors can effectively reduce interference with animal behavior and improve the naturalness of the experiment. At the same time, the ion-electric flexible sensor adopted has good environmental adaptability and is not affected by temperature and humidity, ensuring stable performance in a changing experimental environment.

[0030] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the framework of a flexible intelligent sensor monitoring system for animal experiments according to a preferred embodiment of the present invention;

[0032] Figure 2 It is a schematic diagram of the framework of a flexible intelligent sensor monitoring system for animal experiments according to a preferred embodiment of the present invention;

[0033] Figure 3 It is a capacitance value-pressure schematic diagram of a flexible intelligent sensor monitoring system for animal experiments according to a preferred embodiment of the present invention;

[0034] Figure 4It is an abstract diagram of a flexible intelligent sensor monitoring system for animal experiments according to a preferred embodiment of the present invention;

[0035] Figure 5 It is a schematic diagram of a deep learning model of a flexible intelligent sensing monitoring method for animal experiments in a preferred embodiment of the present invention;

[0036] Figure 6 It is a data flow chart of a flexible intelligent sensing monitoring method for animal experiments according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] In the following description, specific details such as specific internal procedures and techniques are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0039] like Figure 1-2 As shown, an embodiment of the present invention provides a flexible intelligent sensing monitoring system for animal experiments, including a data acquisition module, a data conversion and transmission module, and a data processing module. The data acquisition module acquires changes in capacitance values ​​related to animal activities, and outputs analog signals to the data conversion and transmission module. The data conversion and transmission module converts the received signals into digital signals and transmits them to the data processing module. The data processing module processes the received digital signals and uses a trained deep learning model to identify animal trajectories and behavior categories, and analyzes the corresponding brain wave signals of the animals at that moment to enhance the understanding of animal behavior.

[0040] Among them, the data acquisition module is a flexible off-voltage capacitance sensor array, including m sensors (force sensitive point units) arranged in an n×n matrix. This array design allows the system to arrange sensors at a high density in a limited space to fully capture the pressure information applied by animals during activities. Through these pressure sensors, the system can obtain changes in capacitance values ​​related to animal activities and output analog signals related to these changes. This process ensures the real-time and accuracy of the data, so that subsequent data processing can obtain a reliable information basis. In this embodiment, the experimental area is a square area of ​​about 40cm×40cm, each sensor is 2cm×2cm in size, the distance between each sensor is set to 5mm, and it is arranged in 16×16 to ensure the uniform distribution of captured pressure.

[0041] The data conversion and transmission module includes a data conversion unit and a transmission unit. The data conversion unit includes a 4-channel high-precision capacitance sensor detection chip (m / 4) and a microcontroller. The capacitance sensor detection chip converts the measured capacitance value change into a digital signal and transmits it to the microcontroller through I2C. Then the microcontroller sends the data to the transmission unit through UART. The data conversion and transmission module is composed of a low-power Bluetooth microcontroller, which transmits data to the processing device through low-power Bluetooth. The protocol supports BT 5.0 / 5.1 standards, and the transmission rate supports 125Kbps / 500Kbps / 1Mbps / 2Mbps, so that it can flexibly adjust the transmission rate according to different needs. This low-power design greatly reduces the energy consumption of the system, is suitable for long-term animal experiments, and ensures the stable transmission of continuous data. In order to more intuitively display the analysis results, the system visualizes the animal pressure data: a square is used to represent the flexible I-V capacitance sensor array, different levels of pressure values ​​are represented by different colors, and an n×n matrix is ​​used to calculate the relative position of each sensor.

[0042] The present invention further provides an animal experiment flexible intelligent sensing monitoring method, comprising the following steps:

[0043] Before starting the experiment, first confirm whether all equipment connections are correct; including the connection of the data acquisition module, data conversion and transmission module, and data processing module. In particular, the flexible I-V capacitance sensor array must be installed in the appropriate position and in good contact with the experimental animals to ensure the effectiveness of subsequent data acquisition. In this embodiment, the data acquisition module includes a flexible I-V capacitance sensor array arranged in a 16×16 matrix, that is, 256 sensors (force sensitive point units). The data conversion and transmission module consists of 64 capacitive sensor detection chips, a microcontroller and a low-power Bluetooth microcontroller. The data processing module consists of a PC and a model.

[0044] After ensuring that the equipment is connected correctly, start the equipment and check the working status of the equipment to ensure that they can receive and process signals normally, that is, the green LED light of the data conversion and transmission module flashes to indicate that the equipment has started normally. This process is crucial for obtaining real-time data and ensuring the accuracy of subsequent analysis.

[0045] Place the experimental animal on the sensor array to ensure that the experimental animal can move freely. At this time, the system begins to collect the pressure data applied by the experimental animal on the sensor in real time and outputs analog signals related to these pressure changes; specifically, the capacitance value Ci measured by each sensor will change with the pressure change, forming an initial capacitance data matrix. The matrix is ​​marked as C = [C11, C12, ···, Cnn] (n = 16), where Cij represents the capacitance value of the sensor in the i-th row and j-th column.

[0046] Then the capacitive sensor detection chip converts the received analog signal into a digital signal and transmits it to the microcontroller through the I2C protocol. After receiving the digital signal, the microcontroller sends it to the low-power Bluetooth microcontroller through the UART protocol, and finally transmits it to the data processing module; the Bluetooth microcontroller adopts the BT 5.0 / 5.1 standard for wireless data transmission to ensure the stability and efficiency of the data. At the same time, its low-power design is also suitable for long-term animal experiments to avoid data collection due to power supply problems. WiFi transmission can also be used here. In an embodiment of the present invention, the self-calibration circuit inside the high-precision capacitive sensor detection chip can eliminate the fixed distance error at the sensor end. At the same time, it can also support 2-way signal input, and the differential input mode of the 2-way signal can offset the drift of the parasitic capacitance and temperature of the external environment. Finally, the analog can be quantized into a numerical value through a high-performance ADC, and then the DSP is used to convert the raw data converted from the ADC into a signal C' that filters out environmental noise.

[0047] The data processing module processes the received digital signal, and uses the trained deep learning model to identify the animal's trajectory and behavior category, and analyzes the animal's brain wave signal at that moment to enhance the understanding of the animal's behavior; and visualizes the results. In this implementation, PC / Raspberry Pi, etc. are used as data processing modules. With the help of pre-trained deep learning models, the system can identify the movement trajectory and behavior category of animals (such as mice). At the same time, the pressure data is analyzed correspondingly with the animal's brain wave signal to enhance the understanding of the animal's behavior. After the analysis is completed, the system displays the results in a visual form. A square graphic is used to represent the sensor array, and different levels of pressure values ​​are marked with different colors. This visual display method is not only intuitive and easy to understand, but also convenient for researchers to quickly identify and analyze the pressure changes of animals. Specifically, the data processing module receives the pressure data after noise filtering, and uses the difference X between the pressure data received for the second time and the pressure data received for the first time as input data into the artificial neural network model, X = (C" k+1 -C″ k ) where k is the kth group of data received, 20 groups per second, and C″ 0 =(0,0,···,0). After the model is calculated, the current mouse behavior category is output. The mouse behavior categories are: normal, squatting, and stretching. The model is generally a network structure consisting of an input layer, a hidden layer, and an output layer. The input layer processes the input data; each layer of the hidden layer contains a large number of neurons and calculates the corresponding features; the output layer maps the output of the hidden layer to the output category. The artificial neural network calculation process includes initializing network weights and neural network thresholds, forward propagation, and back propagation. The neural network initializes the network weights and thresholds using a randomization method; forward propagation calculates the input and output of the hidden layer neurons and the output layer neurons layer by layer; back propagation corrects the weights and thresholds according to the loss function. The loss function is the truth or falsity of the predicted category and the real category, expressed as:

[0048]

[0049] Among them, L represents the calculation function; Y represents the true value of the sample; F(X) represents the model prediction value.

[0050] In an embodiment of the present invention, a large number of labeled sample data of different mice are first collected, and each sample data reflects the correspondence between the distribution of mouse stress data and the behavior category label. The sample data is constructed into a training data set and a test data set. For example, 80% of the samples are used as training data and 20% of the data are used as test data. The model is trained using multiple rounds of training, and the final accuracy of the model reaches the requirement of being able to accurately identify mouse behavior. The trained model is deployed to a PC / Raspberry Pi to infer the mouse behavior category in real time. The deep learning model can adopt a variety of types, for example, a convolutional network or a deep belief network DBN. The present invention does not limit the network type and specific structure.

[0051] In order to reduce the error caused by possible interference, the data processing module normalizes and denoises C′ to obtain C ″ ′. The data is classified and processed using a convolutional neural network model based on deep learning to identify animal behaviors. Input layer: receives the X matrix. Hidden layer: extracts the spatial features of pressure distribution for identifying different behavior patterns. Output layer: outputs the probability of each behavior and uses the softmax function to determine the classification result. Trajectory calculation: identifies the movement path of the animal by tracking the changes in pressure intensity over a period of time.

[0052] The system saves the analysis results and visualization data to the database for subsequent review and analysis to prevent data loss and ensure the long-term availability and reliability of experimental results.

[0053] After the final experiment is completed, turn off all equipment.

[0054] A specific embodiment is given below to illustrate the flexible intelligent sensing monitoring method for animal experiments of the present invention.

[0055] The flexible sensor array is fixed to the bottom of the mouse experimental box to ensure that the entire experimental area is covered by sensors. The size of this area is 40cm×40cm, which is consistent with the size of the sensor array, ensuring that the mouse can generate pressure data at every position in the activity area.

[0056] Experimental mouse adaptation period: Place the experimental mice in the experimental box and allow them to move freely for 10 minutes to adapt to the new environment and reduce unnecessary stress reactions.

[0057] Start data collection of the flexible sensor array. When the experimental mouse moves in the experimental box, the sensor will record the pressure changes it exerts. The sensor data is recorded in times, and the data is collected 20 times per second.

[0058] Each frame of data is initially processed by a capacitive sensing chip and a microcontroller and transmitted to a data processing device in the form of a digital signal via Bluetooth. The data processing device performs processing including denoising and normalization, and then uses a deep learning-based convolutional neural network model to classify the data to identify animal behavior and trajectory.

[0059] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A flexible intelligent sensor monitoring system for animal experiments, characterized in that: It includes a data acquisition module, a data conversion and transmission module, and a data processing module. The data acquisition module acquires changes in capacitance values ​​related to animal activities and outputs analog signals to the data conversion and transmission module. The data conversion and transmission module converts the received signals into digital signals and transmits them to the data processing module. The data processing module processes the received digital signals and uses a trained deep learning model to identify animal trajectories and behavior categories, and analyzes the corresponding brain wave signals of the animals at that moment to enhance the understanding of animal behavior.

2. The flexible intelligent sensing monitoring system for animal experiments according to claim 1, characterized in that: The data acquisition module is a flexible I-V capacitance sensor array, including m sensors arranged in an n×n matrix.

3. The flexible intelligent sensor monitoring system for animal experiments as claimed in claim 2, characterized in that: The size of the sensor is 2 cm×2 cm.

4. The flexible intelligent sensing monitoring system for animal experiments as claimed in claim 3, characterized in that: The distance between the sensors is set to 5-10 mm.

5. The flexible intelligent sensing monitoring system for animal experiments according to claim 1, characterized in that: The data conversion and transmission module includes a data conversion unit and a transmission unit. The data conversion unit includes a 4-channel high-precision capacitance sensor detection chip and a microcontroller. The capacitance sensor detection chip converts the measured capacitance value change into a digital signal and transmits it to the microcontroller through I2C, and then the microcontroller sends the data to the transmission unit through UART.

6. The flexible intelligent sensor monitoring system for animal experiments according to claim 1, characterized in that: The data conversion and transmission module includes a low-power Bluetooth microcontroller, which transmits data to the data processing module via low-power Bluetooth.

7. The animal experiment flexible intelligent sensor monitoring system according to claim 6, characterized in that: The transmission protocol of the data conversion and transmission module supports BT 5.0 / 5.1 standards, and the transmission rate supports 125Kbps / 500Kbps / 1Mbps / 2Mbps.

8. A monitoring method using an animal experiment flexible intelligent sensing monitoring system as claimed in any one of claims 1 to 7, characterized in that: The following steps are involved: Before starting the experiment, make sure all devices are connected correctly; After ensuring that the devices are connected correctly, start them and check their working status to ensure that they can receive and process signals normally; Place the experimental animal on the sensor array and ensure that the experimental animal can move freely. At this time, the system begins to collect the pressure data applied by the experimental animal on the sensor in real time and outputs analog signals related to these pressure changes; Then the capacitive sensor detection chip converts the received analog signal into a digital signal and transmits it to the microcontroller through the I2C protocol. After receiving the digital signal, the microcontroller sends it to the low-power Bluetooth microcontroller through the UART protocol, and finally transmits it to the data processing module; The data processing module processes the received digital signals and uses the trained deep learning model to identify the animal's trajectory and behavior category, and analyzes the corresponding brain wave signals of the animal at that moment to enhance the understanding of the animal's behavior; and visualize the results; The system saves the analysis results and visualization data to the database for subsequent review and analysis; After the final experiment is completed, turn off all equipment.

9. The flexible intelligent sensing monitoring method for animal experiments according to claim 8, characterized in that: The results are displayed visually, specifically using squares to represent the flexible I-V capacitive sensor array, different levels of pressure values ​​are represented by different colors, and an n×n matrix is ​​used to calculate the relative position of each sensor.

10. The flexible intelligent sensing monitoring method for animal experiments according to claim 8, characterized in that: Then the capacitive sensor detection chip converts the received analog signal into a digital signal and transmits it to the microcontroller through the I2C protocol. After receiving the digital signal, the microcontroller sends it to the low-power Bluetooth microcontroller through the UART protocol, and finally transmits it to the data processing module, which specifically includes: The self-calibration circuit inside the high-precision capacitive sensor detection chip eliminates the fixed distance error at the sensor end and supports two-way signal input. The differential input mode of the two-way signal offsets the parasitic capacitance of the external environment and the drift of temperature. Finally, the analog is quantized into a numerical value through a high-performance ADC, and then the DSP is used to convert the raw data converted from the ADC into a signal C' that filters out the environmental noise.

Images