Cumulative pressure stress effect evaluation system and method

Abstract

Disclosed in the present invention are a cumulative pressure stress effect evaluation system and method. The system comprises: an animal modeling module, configured to construct various pressure stress level models by means of applying different durations of pressure to a target model animal; a spontaneous behavior capture module, configured to capture spontaneous behaviors of the target model animal corresponding to different pressure stress levels to obtain corresponding action recognition data, and construct a first data set according to a corresponding relationship between the action recognition data and the pressure stress levels; a behavior phenotype analysis module, configured to use linear discriminant analysis to perform dimension reduction on the first data set, so as to obtain a second data set in a discrimination space; and a pressure degree evaluation module, configured to use the second data set to train a machine learning model, so as to obtain an optimized cumulative pressure stress effect evaluation model. The present invention can be used for cumulative pressure stress effect evaluation, is suitable for precise medicine, and can achieve behavior phenotype difference recognition induced by the same pressure stress source and different pressure degrees.

Description

A system and method for evaluating cumulative effects of stress Technical Field

[0001] The present invention relates to the field of medical health technology, and more specifically, to a system and method for evaluating the cumulative effects of stress. Background Art

[0002] Stress represents a series of physiological responses to external stimuli, including neural, endocrine, and immune responses. Excessive stress has been linked to a variety of diseases, including cardiovascular, metabolic, psychological, neurological, and psychiatric disorders. Therefore, quantifying the effects of stress, assessing stress levels, and providing timely warnings before stress causes pathological harm to the body are of profound significance.

[0003] While the pathological mechanisms of stress-induced depression are not fully understood, it is known that an overactive hypothalamic-pituitary-adrenal (HPA) axis plays a key role in the pathogenesis of depression. The HPA axis is a crucial component of the human endocrine system, regulating stress responses, metabolism, immunity, and other functions. It plays a crucial role in the production of the stress hormone cortisol. In response to stressful situations such as danger or stress, the hypothalamus releases thyrotropin-releasing hormone (TRH), which stimulates the anterior pituitary gland to release adrenocorticotropic hormone (CRH), which in turn stimulates the adrenal cortex to secrete hormones such as cortisol to help the body cope with stress. It is important to note that the human body's response to stress and cortisol is complex. While moderate stress is beneficial, excessive cortisol secretion can have adverse effects. While high levels of cortisol help the body cope with stress, they can also induce behavioral changes such as depression and anxiety, and can also cause pathological damage to neurons. To mitigate these harmful effects, it is essential to be able to promptly assess stress levels and prescribe appropriate treatments. On the other hand, drug development for stress-related mental illnesses such as depression, anxiety, and post-traumatic stress disorder is difficult. One of the key issues is that the modeling evaluation indicators of animal models are difficult to quantify. In particular, traditional behavioral methods have difficulty distinguishing the different cumulative effects of the same stressor. At the same time, there are subjective factors, individual differences and other problems. Therefore, the development of an accurate and reliable stress evaluation system is an issue that needs to be urgently addressed. Technical issues

[0004] Gait studies of patients with depression have shown that the spontaneous behavior of those with stress-related mental illnesses differs from that of normal individuals. For example, during walking, there are reduced vertical head movements, smaller limb amplitudes, and slower gait speeds. This also suggests the possibility of using spontaneous behavior to assess physical and mental states. Regarding behavioral detection and analysis, patent application CN202010886434.1 provides a three-dimensional capture device, method, system, and application for animal behavioral recording, enabling the automatic capture of spontaneous behavior in unmarked animals. Patent application CN202210527842.7 discloses an animal behavior reconstruction system, method, device, and storage medium capable of three-dimensional reconstruction of animal behavior using two or more cameras. Patent application CN202010493350.1 discloses a behavior recognition device and method that achieves unsupervised animal behavior decomposition by extracting features, decomposing information, and identifying behavior. However, these methods primarily focus on capturing, storing, and identifying animal behavior, and do not address the construction of animal stress models or the analysis of stress-related behavioral phenotypes.

[0005] In terms of drug evaluation, patent application CN202211554237.5 discloses a method for evaluating the efficacy of psychiatric medications, a system for evaluating the efficacy of treatment devices, and a method for evaluating the efficacy of psychiatric medications based on behavior. However, it does not distinguish between the cumulative effects of the same stressor, even when the differences between groups are small. Technical Solutions

[0006] The purpose of the present invention is to overcome the above-mentioned defects of the prior art and provide a system and method for evaluating the cumulative effect of pressure stress.

[0007] According to a first aspect of the present invention, a system for evaluating the cumulative effects of stress and pressure is provided. The system comprises: an animal modeling module, a spontaneous behavior capture module, a behavioral phenotype analysis module, and a stress level evaluation module, wherein:

[0008] The animal modeling module is configured to construct various stress level models by applying stress of different durations to the target model animals;

[0009] The spontaneous behavior capture module is configured to capture the spontaneous behaviors of the target model animal corresponding to different stress levels, obtain corresponding action recognition data, and construct a first data set based on the correspondence between the action recognition data and the stress levels;

[0010] The behavioral phenotyping module is configured to perform dimensionality reduction on the first data set using linear discriminant analysis to obtain a second data set in a discriminant space;

[0011] The stress level evaluation module is configured to use the second data set to train a machine learning model to obtain an optimized stress cumulative effect evaluation model.

[0012] According to a second aspect of the present invention, a method for evaluating the cumulative effects of stress and pressure is provided. The method comprises the following steps:

[0013] Apply stress of different durations to target model animals to construct models with multiple stress levels;

[0014] capturing the spontaneous behaviors of the target model animal corresponding to different stress levels, obtaining corresponding action recognition data, and constructing a first data set based on the correspondence between the action recognition data and the stress levels;

[0015] Using linear discriminant analysis to reduce the dimensionality of the first data set to obtain a second data set in the discriminant space;

[0016] The second data set is used to train the machine learning model to obtain an optimized stress cumulative effect evaluation model. Beneficial effects

[0017] Compared with the existing technology, the advantage of the present invention is that the system designed for identifying and evaluating the degree of stress is based on the behavioral patterns of patients after being under stress, and performs stress modeling evaluation and stress cumulative effect evaluation according to the differences in spontaneous behavioral phenotypes. It can distinguish the cumulative effects of the same stress source with small differences between groups, and realize the identification of behavioral phenotypes of the same stress source and different stress levels, which can be applied to fields such as precision medicine and mental stress level judgment.

[0018] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0020] FIG1 is a schematic diagram of a system for evaluating cumulative stress effects according to an embodiment of the present invention;

[0021] FIG2 is a schematic diagram of an animal pressure modeling method according to one embodiment of the present invention;

[0022] FIG3 is a schematic diagram of a visualization of dimensionality reduction of action recognition data according to an embodiment of the present invention;

[0023] FIG4 is a schematic diagram of using motion recognition data prediction for stress assessment according to an embodiment of the present invention;

[0024] FIG5 is a flow chart of a method for evaluating cumulative effects of pressure stress according to an embodiment of the present invention. Best Mode for Carrying Out the Invention

[0025] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention.

[0026] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0027] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0028] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0029] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0030] As shown in FIG1 , the provided stress cumulative effect evaluation system generally includes an animal modeling module, a spontaneous behavior capture module, a behavioral phenotype analysis module, and a stress level evaluation module.

[0031] The Animal Modeling Module is used to construct various stress models by applying stress of varying duration to target animals. Animal models can be mice, rats, or other animal species. Modeling methods can include tail suspension stress, social defeat stress, predator stress, chronic restraint stress, unpredictable stress, or other stress types.

[0032] For example, using the tail suspension stress test on mice, different lengths of tail suspension can be used to induce stress in the animals, thereby achieving a model with different degrees of cumulative stress effects. For example, seven levels of stress can be generated by adjusting the length of tail suspension.

[0033] Figure 2 is a schematic diagram of an animal stress model. The tail of a mouse is fixed at a certain height, keeping the tip of its nose approximately 20 cm from the ground, for example, within a range of 15 cm to 25 cm. Different degrees of stress modeling can be achieved by adjusting the time the mouse is suspended.

[0034] The spontaneous behavior capture module is used to capture the spontaneous behavior of the target model animal corresponding to different stress levels, obtain corresponding action recognition data, and construct a data set (or called the first data set) based on the correspondence between the action recognition data and the stress levels.

[0035] For example, immediately after a stress experiment, spontaneous animal behavior capture, skeleton reconstruction, and behavioral segmentation are performed to obtain behavioral segments. These segments are then clustered using a clustering algorithm to identify motion data and determine several motion types. Specifically, the data type of the animal's spontaneous behavioral data can be video data or image data. For example, the motion type acquisition method includes: extracting body feature information (such as the position of the head and limbs) corresponding to the time series of the target animal from video data; performing posture decomposition on the body feature information to obtain posture information (such as the morphological characteristics of the animal's organs and limbs in the video image); performing temporal dynamic clustering on the posture information to obtain motion information (such as walking, hunching, etc.); calculating the target animal's speed information based on the body feature information; and clustering this speed information with the motion information to obtain motion sequence information. Based on this motion sequence information, the proportion of different motions in the captured video duration is calculated, and the target animal's motion type identification result (such as walking, running, grooming, etc.) is output. This design also allows the identification of the trajectory of animal body parts, such as limbs, head, and tail, over time from video or image data.

[0036] The behavioral phenotype analysis module is used to perform dimensionality reduction on the first data set using linear discriminant analysis to obtain a data set in a discriminant space.

[0037] For example, given the relatively high dimensionality of the first dataset, typically 40 or even higher, further dimensionality reduction of the action recognition data is performed using methods such as linear discriminant analysis to obtain a discriminant space that maximizes differentiation across stress levels. The basic principle of linear discriminant analysis is to project examples from a dataset with different class labels into a low-dimensional space, aligning the projected points of similar examples as closely as possible and those of heterogeneous examples as far apart as possible. When classifying new examples, they are projected into the same space, and their class is determined based on the position of the projected points. This design allows the first dataset to be projected into a lower-dimensional space, maximizing the inter-class variance and minimizing the intra-class variance. For example, Figure 3 is a visualization of dimensionality reduction for action recognition data. Seven different levels of stress were selected for data collection. The control group was subjected to no stress. Models 1 through 6 for different stress levels represent hanging tails of 1, 2, 3, 4, 5, and 6 minutes, respectively. The granularity of the time length and the number of stress levels can be set according to actual needs (for example, longer than 6 minutes), and the granularity of the model using minutes as the stress level proves that the cumulative effect of stress even for such a short time can be accurately identified.

[0038] Analysis of animal motion recognition data revealed that each group has a specific distribution within the discriminant space of linear discriminant analysis. Two-dimensional visualization results showed that there were clear dividing lines between almost all groups. Furthermore, by comparing the spatial location of the data, the effectiveness of stress modeling can be evaluated.

[0039] The stress level evaluation module is used to train the machine learning model using the reduced-dimensionality dataset to obtain an optimized stress cumulative effect evaluation model.

[0040] Specifically, the machine learning model includes a training process and an application process. During model training, the reduced-dimensionality dataset is input into the machine learning model. For example, with minimizing the classification error rate as the optimization goal, the correspondence between the motion recognition data and the stress level is learned, and optimized model parameters such as weights and biases are obtained. This optimized model is then used as a stress-stress cumulative effect evaluation model. During the model application process, the newly collected motion recognition data is input into the stress-stress cumulative effect evaluation model to obtain the stress-stress cumulative effect (i.e., stress-stress level). For example, during the model application process, the reduced-dimensional six-dimensional features of seven groups can be newly collected for the target animal model and input into the stress-stress cumulative effect evaluation model to obtain a predicted stress-stress level label.

[0041] In one embodiment, to better identify stress levels, machine learning algorithms such as support vector machines or k-nearest neighbors are combined to further improve the efficiency and accuracy of stress level identification. During this process, for visualization purposes, the first two dimensions of the reduced data can be extracted and plotted. By analyzing the distribution of data from different groups in the scatter plot, the interfaces between the groups can be identified. Machine learning models can include k-nearest neighbors, support vector machines, decision trees, and neural networks. Metrics such as confusion matrices and F1 scores can be used to accurately assess the model's effectiveness in identifying the cumulative effects of stress.

[0042] Figure 4 shows the application of spontaneous behavior prediction to stress assessment. Taking the support vector machine as an example, the confusion matrix of the support vector machine classification shows that the seven stress groups can be well distinguished in the discriminant space.

[0043] Accordingly, the present invention also provides a method for evaluating the cumulative effects of stress and pressure, which is used to implement one or more aspects of the above-mentioned system. Referring to FIG5 , the method includes: step S110, applying stress of varying durations to a target model animal to construct multiple stress and pressure level models; step S120, capturing the spontaneous behaviors of the target model animal corresponding to different stress and pressure levels, obtaining corresponding motion recognition data, and constructing a first dataset based on the correspondence between the motion recognition data and the stress and pressure levels; step S130, performing dimensionality reduction on the first dataset using linear discriminant analysis to obtain a second dataset in the discriminant space; and step S140, using the second dataset to train a machine learning model to obtain an optimized model for evaluating the cumulative effects of stress and pressure.

[0044] It should be noted that, without violating the spirit and scope of the present invention, those skilled in the art may make appropriate changes or modifications to the above embodiments. For example, a comprehensive evaluation of the behavioral phenotype of animals can be applied to the identification of the same stressor and different stress levels. Without loss of generality, the identification of spontaneous behavior of animals with different stressors is still applicable. For another example, more or fewer stress groups can be used. In addition, the present invention is not limited to the identification of stress levels, but can also be extended to other data, such as the identification of psychotropic drugs, and has been attempted on public data and is feasible.

[0045] To further verify the effectiveness of the present invention, an experimental verification was conducted. Model analysis of mice using tail suspension experiments of different lengths of time showed that the present invention can achieve efficient and accurate identification of the cumulative effects of stress.

[0046] In summary, compared with the prior art, the present invention has the following advantages:

[0047] 1) This invention can be used to evaluate the cumulative effects of stress and achieve identification of behavioral phenotypes of the same stressor but different stress levels, and is suitable for precision medicine, mental stress level assessment, and other aspects.

[0048] 2) This paper explores the relationship between the changing pattern of action type proportions under the same stress source and different stress levels and the machine learning model, and uses machine learning to evaluate the cumulative effect of stress, thereby improving the accuracy of the evaluation results.

[0049] 3) This paper designs an animal behavioral phenotyping and visualization method based on dimensionality reduction. By using linear dimensionality reduction to obtain a discriminant subspace, the distribution of data from different groups within the discriminant space is identified. This allows for differentiation of the cumulative effects of the same stressor, even when differences between groups are minimal, in drug evaluation.

[0050] 4) The present invention designs a method for analyzing spontaneous behavior in stress-induced animal models, which includes dimensionality reduction of spontaneous behavior data and classification of the reduced data using machine learning methods, thereby improving the classification efficiency and recognition accuracy of stress levels.

[0051] The present invention may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0052] A computer-readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure within a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0053] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0054] The computer program instructions for performing the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, Python, MATLAB, and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present invention.

[0055] Various aspects of the present invention are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0056] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0057] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0058] The flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of an instruction, and the module, program segment or part of the instruction contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0059] While various embodiments of the present invention have been described above, the foregoing description is intended to be illustrative, non-exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. A system for evaluating the cumulative effect of stress, comprising: Animal modeling module, spontaneous behavior capture module, behavioral phenotype analysis module and stress level assessment module, including: The animal modeling module is configured to construct models of various stress levels by applying stress of different durations to the target model animals; The spontaneous behavior capture module is configured to capture the spontaneous behaviors of the target model animal corresponding to different stress levels, obtain corresponding action recognition data, and construct a first data set according to the corresponding relationship between the action recognition data and the stress level; The behavioral phenotyping module is configured to reduce the dimension of the first data set using linear discriminant analysis to obtain a second data set in a discriminant space; The stress level evaluation module is configured to use the second data set to train the machine learning model to obtain an optimized stress cumulative effect evaluation model.

2. The system according to claim 1, characterized in that The target model animal is a mouse, and the nose tip of the mouse is controlled to be at a predetermined distance from the ground, and the multiple pressure stress level models are constructed by adjusting the tail suspension time of the mouse.

3. The system according to claim 2, characterized in that Each of the multiple stress level models is defined according to the length of the tail suspension time of the mouse, and the time granularity between each stress level is set to the minute level.

4. The system according to claim 2, characterized in that The distance between the nose tip of the mouse and the ground was set to be in the range of 15 cm to 25 cm.

5. The system according to claim 1, characterized in that The behavioral phenotype analysis module is also used to extract the first two dimensions of the reduced dimension data to draw a scatter plot after reducing the dimension of the first data set, and measure the quality of the second data set by analyzing the distribution positions of different stress level models in the scatter plot.

6. The system according to claim 1, characterized in that The performance of the stress cumulative effect evaluation model is evaluated using the confusion matrix and F1 score indicators of the support vector machine.

7. The system according to claim 1, characterized in that The machine learning model is a support vector machine or a K-nearest neighbor model.

8. A method for evaluating the cumulative effect of stress, comprising the following steps: Apply stress of different durations to target model animals to construct models with multiple stress levels; Capturing the spontaneous behaviors of the target model animal corresponding to different stress levels, obtaining corresponding action recognition data, and constructing a first data set according to the corresponding relationship between the action recognition data and the stress levels; Using linear discriminant analysis to reduce the dimension of the first data set to obtain a second data set in a discriminant space; The second data set is used to train the machine learning model to obtain an optimized stress cumulative effect evaluation model.

9. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the steps of the method according to claim 8 are implemented.

10. A computer device comprising a memory and a processor, wherein a computer program capable of running on the processor is stored in the memory, characterized in that: When the processor executes the computer program, the steps of the method according to claim 8 are implemented.

Images