A method for identifying homeostatic and transitional states of a human body

Abstract

The application provides a method for identifying a physiological steady state and a transition state of a human body, which comprises the following steps: obtaining modeling features by collecting physiological signals of a subject for at least 24 hours, constructing a historical state matrix H0 by using the modeling features; constructing an individualized physiological state representation matrix {D (Sl)}; reconstructing observation data X obs to obtain estimated data X est , and matching a state sequence; calculating a residual sequence R between the observation data X obs and the estimated data X est ; if the residual sequence R satisfies a low-variance zero-mean Gaussian distribution, it is judged that the physiological system is in a steady state; if the R variance shows a gradual increasing trend, the state is a transition state; if the R mean deviates from zero but is relatively stable, and the variance does not change much, it indicates that a physiological state that is not observed when the H0 is constructed appears, and the individualized physiological state representation matrix {D (Sl)} needs to be updated; if the R first shows a gradual increasing trend, and then shows a relatively stable state after a period of time, it is judged that a new steady state of the physiological system appears.

Description

Technical Field

[0001] This application relates to the field of physiological state monitoring and analysis, and in particular to methods for identifying the physiological homeostasis and transitional states of the human body. Background Technology

[0002] In fields such as medicine, sports science, psychology, and biofeedback, understanding and differentiating human physiological states is crucial for diagnosis, treatment, and training. In recent years, the concept of "homeostasis" has been increasingly applied to the field of physiological health. The theory of "homeostasis" is a key supporting theory in physiology, referring to how the body maintains a relatively stable physiological system through negative feedback mechanisms under normal physiological conditions. When the body experiences a stress response, it can adaptively adjust its target set point to adapt to a new homeostasis. This process involves a "transition state"—a state where the body transitions from one homeostasis to another. A successful transition enhances the body's ability to adapt to the environment and increases homeostasis levels; conversely, a poor transition lowers homeostasis levels, leading to bodily damage.

[0003] The concept of "homeostasis" in the human physiological system was proposed long ago, providing a theoretical basis for the study of physiological system state identification. However, its abstract nature makes quantitative analysis difficult in practical applications. Most existing physiological state monitoring technologies focus on assessing an individual's health status through static or intermittent data collection. For example, they extract time-domain, frequency-domain, and standard template correlation indicators based on physiological signals such as electrocardiogram, respiration, and activity levels, and then analyze and evaluate the human state under specific conditions using signal processing or machine learning methods. These methods typically lack a deep understanding of the continuity and dynamic changes in an individual's physiological state.

[0004] The patent application with application number 202110245934.1 discloses a physiological state assessment method and a physiological state assessment device. Although it can convert electrocardiogram data into digital integrated data for assessment, it may not fully consider the dynamic changes of individual homeostasis and transition state.

[0005] Patent application number 202310297907.8 discloses a method and system for monitoring and evaluating the physiological state of pregnant women during the perinatal period. This system can monitor the physiological state of pregnant women in real time based on perinatal characteristic attribute parameters and obtain maternal risk assessment information through model evaluation, but it may not be applicable to the extensive assessment of other physiological states.

[0006] Patent application number 202310579449.7 discloses a method and system for processing vital sign data based on smart wearable devices. Although it can assess the level of physiological state, it may have limitations in the accuracy of individualized modeling and state identification. Summary of the Invention

[0007] In view of the above problems, this application aims to propose a method for identifying the physiological homeostasis and transitional state of the human body, which can accurately identify and quantify the homeostasis and transitional state of an individual and adapt to the assessment needs under different physiological conditions.

[0008] The method for identifying the physiological homeostasis and transitional state of the human body as described in this application includes: Modeling features are obtained by collecting physiological signals from subjects for at least 24 hours, and historical state matrix H0 is constructed using these features. Using temporal subspace segmentation techniques and external auxiliary information, an individualized physiological state representation matrix {D} is constructed. (Sl)}; For the observed data X obs Reconstruction based on a structured sparse coding framework is performed to obtain the estimated data X. est and match the state sequence; Calculate the observed data X obs Compared with estimated data X est The residual sequence R between them; If the residual sequence R follows a low-variance, zero-mean Gaussian distribution, the physiological system is considered to be in its original steady state. If the variance of R shows a gradually increasing trend, the state is in a transitional state. If the mean of R deviates from zero but remains relatively stable, and the variance does not change significantly, it indicates the presence of a physiological state that was not observed during the construction of H0, requiring an update to the individualized physiological state representation matrix {D}. (Sl) If R first shows a gradual increasing trend, and then shows a relatively stable state after a period of time, it is judged that a new homeostasis has appeared in the physiological system.

[0009] Preferably, the physiological signals include electrocardiogram signals, respiratory signals, and blood oxygen saturation signals.

[0010] Preferably, the modeling features include basic vital sign parameters derived from physiological signals and derived parameters.

[0011] Preferably, the external auxiliary information includes: human posture and activity status information, and sleep stage information.

[0012] Preferably, the modeling features include: heart rate, heart rate variability, respiratory rate, respiratory rate variability, pulse rate, blood oxygen saturation, cardiopulmonary coupling, body posture, and activity intensity.

[0013] Preferably, the modeling features have undergone noise and outlier removal, missing value imputation, and aggregation analysis.

[0014] Preferably, the subjects' activity states are divided into static, low-intensity, and high-intensity activity states based on different activity intensities; the historical state matrix H0 is divided into subspaces representing different activity states based on the subjects' activity state information.

[0015] Preferably, sleep stage information is obtained through sleep stage detection, including: light sleep, deep sleep and REM sleep; the historical state matrix H0 is divided into subspaces for different sleep stages according to the subject's sleep stage information.

[0016] Preferably, for the observation data X obs The reconstruction is achieved by solving the optimization problem.

[0017] The method for distinguishing between the transitional and steady states of the human body in this application preprocesses physiological signals such as electrocardiogram, respiration, and body position / movement collected by wearable devices, then extracts features, and performs individualized modeling using a multivariate human state estimation method. Through matrix construction, data reconstruction, and residual analysis, the method distinguishes between the transitional and steady states of the human body, providing technical support for individualized physiological homeostasis monitoring. This method is expected to further enhance the application value of physiological data from wearable devices and provide quantitative analysis support for individualized disease management, rehabilitation training, and performance improvement. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the method for distinguishing between the human body's transition state and steady state in this application; Figure 2 A physiological sequence state partitioning diagram based on external auxiliary information; Figure 3 Flowchart of technology for identifying transitional and steady-state states in the human body; Figure 4 This is a graph showing the residual changes during the high-altitude hypoxia experiment. Detailed Implementation

[0019] The implementation environment of this application includes, but is not limited to, medical institutions, sports training centers, home or personal health monitoring scenarios. Required equipment includes wearable devices, non-invasive physiological signal sensors, and data processing and analysis software.

[0020] It should be noted that the content of each module in this application refers to the corresponding functional modules implemented by the method of distinguishing between the human body's homeostatic and transient states when running on a computing device (such as a smartphone, tablet, computer, or server).

[0021] The method for distinguishing between the homeostatic and transient states of the human body, as described in this application, will now be explained in detail with reference to the accompanying drawings.

[0022] 1. Collect physiological signals: Select appropriate wearable devices or non-invasive sensors, such as electrocardiogram (ECG) sensors, breathing belts, pulse oximeters, etc., to continuously collect physiological signals for at least 24 hours, and use signal quality assessment algorithms to eliminate unusable data segments. Physiological data acquisition and quality assessment include... Figure 1 As shown.

[0023] 2. Modeling feature calculation and selection: A set of features is extracted every t minutes from the raw physiological signals, including parameters such as heart rate, heart rate variability, respiratory rate, respiratory rate variability, pulse rate, blood oxygen saturation, cardiopulmonary coupling, body posture, and activity intensity, as modeling feature parameters. These modeling feature parameters are further processed, including noise and outlier removal, missing value imputation, and aggregation analysis, to filter out usable parameters and form a historical state matrix H that can be used for subsequent modeling. 0。

[0024] 3. Divide H0 into states based on external auxiliary information: State classification based on body position / movement information, such as... Figure 2 As shown in Figure a, the activity intensity is quantified using triaxial accelerometer signals and categorized into static, low-intensity, and high-intensity activity states based on their intensity. H0 is then divided into subspaces representing different activity states.

[0025] Based on sleep information, state classification such as Figure 2 As shown in b, sleep stages are detected by integrating physiological signals from various dimensions, and are divided into light sleep, deep sleep, and REM sleep states. H0 is then divided into subspaces representing different sleep states.

[0026] Additionally, depending on the data, H0 can be segmented using temporal subspace partitioning techniques, and after normalization, several subspaces {D} can be obtained. (1) D (2) ,…D (m)}, generate the state matrix {D} (Sl)}. Here, {D (Sl)} = A(H0), where A represents a certain algorithm or rule used to map H0 into several state subspaces.

[0027] 4. Reconstruct new observation data X obs : For the new observation data X obs According to the size of the Gaussian kernel in {D (Sl) Select from X obs The highest-scoring samples form the matching matrix D. obs Simultaneously, the matching state sequence S corresponding to the matching matrix is ​​generated. obs ; The reconstruction result is obtained by solving the following optimization problem, and the specific calculation process is as follows: ; Where C is the coding coefficient matrix, λ>0 is one of the compromise parameters, and D={D (1) D (2) ,…D (m)},or , Where δ>0 is a certain permissible level of interpretation error, or , Here, α>0 is a constraint imposed on the rationality of the explanation, and r(C) is a structured sparse regularization term corresponding to the structure in D. By solving the optimization problem, the observed data is reconstructed to form the estimated data X. est .

[0028] 5. Calculate the residuals: Calculate the residual sequence between the observed and estimated data: R = Xobs - Xest 6. Identification of transition and steady states The flowchart for identifying the transitional and steady states of the human body is as follows: Figure 4 As shown, if the residual sequence R satisfies a low-variance, zero-mean Gaussian distribution, and the matched state sequence S... obs If a system exhibits stationarity and uniformity, it indicates that its observed data X is in a steady state. obs The individualized physiological state representation matrix {D (Sl) If a valid explanation is provided, it can be determined that the physiological system is at its original homeostatic level; conversely, if... (a) If the mean of the residual sequence R is still close to zero, but the variance increases significantly, then it is in a transition state; (b) If the mean of the residual sequence R deviates significantly from zero, but the variance does not change significantly, it means that a physiological state that was not observed when H0 was constructed has appeared. In this case, the individualized physiological state representation matrix {D} needs to be updated. (Sl)}

[0029] Example The following is a specific embodiment to demonstrate the application of the method of the present invention: Example Environment: A volunteer was selected to participate in a simulated high-altitude oxygen chamber experiment, which included approximately 20 hours of baseline physiological data from the plains and approximately 2 hours of simulated high-altitude data. Physiological signals were continuously collected using wearable devices.

[0030] Data processing flow: The plain baseline data from the first two hours was used as the initial data. Preprocessing and feature extraction were performed in 1-minute windows, with features including heart rate, respiratory rate, and activity-related parameters, to construct H0. H0 was then segmented into D based on sleep stages and activity levels. obs .

[0031] The remaining plain baseline data is used as X. obs , for X obs Any single observation vector in the dataset is matched with a corresponding D based on its sleep and activity level ratings. obs The observed vector is then reconstructed to obtain the 10 optimal historical vectors, resulting in the reconstructed value X. est Calculate the residual R.

[0032] Results analysis: After entering the simulated high-altitude environment, such as Figure 4 As shown, residual analysis revealed a significant increase in the variance of R, indicating a transition state.

[0033] After a period of time, the variance of R decreases and stabilizes, by updating {D} (Sl) After that, it is determined to be a new steady state.

[0034] Implementation results: Through the above embodiments, the method of the present invention can effectively distinguish between the homeostatic and transitional states of the human body, providing a new technical means for individualized physiological state monitoring.

[0035] This application's method, through individualized modeling and analysis of continuous physiological data, can distinguish between homeostatic and transitional states in the human body. The individualized modeling method overcomes the limitations of traditional methods in physiological state identification, improving the accuracy and individual adaptability of state identification. The construction of the individualized physiological state representation matrix proposed in this application can solve the state aliasing problem, improving the accuracy of state identification. The reconstruction of new observational data and residual analysis methods used in this application can achieve sensitivity to minute state changes and fitting ability to complex human physiological system states through data-driven approaches. The dynamic update mechanism for state identification proposed in this application can dynamically update the individualized physiological state representation matrix based on residual analysis results to adapt to changes in human physiological states. Application results show that this application's method can effectively distinguish changes in human states, with good test results, and is applicable to a wide range of application scenarios, including but not limited to medical health monitoring, sports training, psychological research, and biofeedback.

[0036] Unless otherwise defined, all technical and / or scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this invention relates. The materials, methods, and embodiments mentioned in this application are illustrative only and not restrictive.

[0037] Although the present invention has been described in conjunction with specific embodiments, those skilled in the art can make appropriate substitutions, modifications and changes within the inventive spirit of this application, and such substitutions, modifications and changes still fall within the protection scope of this application.

Claims

1. A method for identifying the physiological homeostasis and transitional state of the human body, comprising: Modeling features are obtained by collecting physiological signals from subjects for at least 24 hours, and historical state matrix H0 is constructed using these features. Using temporal subspace segmentation techniques and external auxiliary information, an individualized physiological state representation matrix {D} is constructed. (Sl) }; For the observed data X obs Reconstruction based on a structured sparse coding framework is performed to obtain the estimated data X. est and match the state sequence; Calculate the observed data X obs Compared with estimated data X est The residual sequence R between them; If the residual sequence R follows a low-variance, zero-mean Gaussian distribution, the physiological system is considered to be in its original steady state. If the variance of R shows a gradually increasing trend, the state is in a transitional state. If the mean of R deviates from zero but remains relatively stable, and the variance does not change significantly, it indicates the presence of a physiological state that was not observed during the construction of H0, requiring an update to the individualized physiological state representation matrix {D}. (Sl) If R first shows a gradual increasing trend, and then shows a relatively stable state after a period of time, it is judged that a new homeostasis has appeared in the physiological system.

2. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 1, characterized in that: The physiological signals include electrocardiogram signals, respiratory signals, and blood oxygen saturation signals.

3. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 2, characterized in that: The modeling features include basic vital sign parameters derived from physiological signals and derived parameters.

4. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 1, characterized in that: The external auxiliary information includes: human posture and activity status information, and sleep stage information.

5. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 2, characterized in that: The modeling features include: heart rate, heart rate variability, respiratory rate, respiratory rate variability, pulse rate, blood oxygen saturation, cardiopulmonary coupling, body posture, and activity intensity.

6. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 5, characterized in that: The modeling features have undergone noise and outlier removal, missing value imputation, and aggregation analysis.

7. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 4, characterized in that: Based on different activity intensities, the subjects' activity states were divided into rest, low-intensity activity, and high-intensity activity states; based on the subjects' activity state information, the historical state matrix H0 was divided into subspaces representing different activity states.

8. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 4, characterized in that: Sleep stage information is obtained through sleep stage detection, including light sleep, deep sleep, and REM sleep. The historical state matrix H0 is divided into subspaces representing different sleep stages based on the subject's sleep stage information.

9. The method for identifying the physiological homeostasis and transitional state of the human body according to claim 1, characterized in that: For the observed data X obs The reconstruction is achieved by solving the optimization problem.

Images