Method for identifying physiological steady state and transition state of human body
By constructing individualized physiological state representation matrix and data reconstruction technology, the problem of insufficient accuracy of human homeostasis and transition state recognition in the existing technology is solved, and dynamic monitoring and individualized management of human physiological state is realized, improving the The accuracy and adaptability of state identification.
Patent Information
- Application Number
- CN202411315113.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-20
AI Technical Summary
The prior art is difficult to accurately identify and quantify the homeostasis and transitional states of human physiological states, lacks an in-depth understanding of the continuity and dynamic changes of individual physiological states, and the existing methods have limitations in individualized modeling and state identification accuracy.
By collecting physiological signals for at least 24 hours from subjects, constructing historical state matrix H0, using time-sequence subspace segmentation technology and external auxiliary information to construct individualized physiological state representation matrix, performing data reconstruction based on structured sparse coding framework, and computing Residual sequences are used to determine whether the physiological system is in a steady state or transition state, and dynamically update the individualized physiological state representation matrix.
It realizes accurate identification and quantification of human homeostasis and transition states, improves the accuracy and individual adaptability of state identification, is suitable for a wide range of physiological state assessment needs, enhances the application value of physiological data of wearable devices, and is individualized Support for disease management and rehabilitation training.
Smart Images

Figure CN118924263A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of physiological state monitoring and analysis, and in particular to a method for identifying a physiological steady state and a transition state of a human body. Background Art
[0002] In the fields of medicine, sports science, psychology, and biofeedback, understanding and distinguishing the physiological state of the human body is of great significance for diagnosis, treatment, and training. In recent years, the concept of "homeostasis" has gradually been used in the field of physiological health. The "homeostasis" theory is an important supporting theory in physiology, which means that under normal physiological conditions, the body maintains the physiological system in a relatively stable state through a negative feedback mechanism. When the body has a stress response, it can adaptively adjust the target set point and adjust the body to a homeostasis that adapts to the new environment. In this process, a "transition state" will be generated from one homeostasis to another new homeostasis. A good transition will improve the body's ability to adapt to the environment and improve the homeostasis level. Otherwise, it will reduce the homeostasis level and cause damage to the body.
[0003] The concept of "steady state" of the human physiological system was proposed very early, providing a theoretical basis for the study of the state identification of the human physiological system, but its abstractness makes it difficult to achieve quantitative analysis in practical applications. In the existing physiological state monitoring technology, most methods focus on evaluating the health status of individuals through static or intermittent data collection, such as extracting indicators such as time domain, frequency domain, and standard template correlation based on physiological signals such as electrocardiogram, respiration, and activity, and analyzing and evaluating the human state under specific conditions through signal processing or machine learning methods. These methods usually lack an in-depth understanding of the continuity and dynamic changes of individual physiological states.
[0004] The patent 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 evaluation, it may not fully consider the dynamic changes of individual steady state and transition state.
[0005] The patent with application number 202310297907.8 discloses a perinatal maternal physiological state monitoring and evaluation method and system. The system can monitor the maternal physiological state in real time based on perinatal characteristic attribute parameters, and obtain maternal risk assessment information through model evaluation, but may not be suitable for the extensive evaluation of other physiological states.
[0006] Patent application number 202310579449.7 discloses a vital sign data processing method and system based on smart wearable devices. Although it can evaluate 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, the present application aims to propose a method for identifying the physiological steady state and transition state of the human body, which can accurately identify and quantify the steady state and transition state of an individual and adapt to the evaluation needs under different physiological conditions.
[0008] The method of identifying the physiological steady state and transition state of the human body of the present application comprises: Obtain modeling features through at least 24 hours of physiological signals collected from the subjects, and use the modeling features to construct a historical state matrix H0; Using temporal subspace segmentation technology and external auxiliary information, an individualized physiological state representation matrix {D (Sl)}; For the observed data X obs Reconstruct the structured sparse coding framework to obtain the estimated data X est , and match the state sequence; Calculate the observed data X obs With the estimated data X est The residual sequence R between them; If the residual sequence R satisfies the zero-mean Gaussian distribution with low variance, the physiological system is judged to be in the original steady state; if the variance of R shows a trend of gradual increase, the state is a transitional state; if the mean of R deviates from zero but is relatively stable, and the variance does not change much, it indicates that a physiological state that was not observed when H0 was constructed has occurred, and the individualized physiological state representation matrix {D (Sl)}; If R first shows a gradual increase trend and then shows a relatively stable state after a period of time, it is judged that a new steady state 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 and derived parameters from physiological signals.
[0011] Preferably, the external auxiliary information includes: human body posture and activity state 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 been processed by removing noise and outliers, interpolating missing values, and performing aggregation analysis.
[0014] Preferably, the subject's activity state is divided into static, low-intensity activity and high-intensity activity states according to different activity intensities; and the historical state matrix H0 is divided into subspaces in different activity states according to the subject's activity state information.
[0015] Preferably, the sleep staging information is obtained through sleep staging detection, including: light sleep, deep sleep and REM state; the historical state matrix H0 is divided into subspaces in different sleep states according to the sleep staging information of the subject.
[0016] Preferably, for the observed data X obs The reconstruction is achieved by solving the optimization problem.
[0017] The method of distinguishing the transition state and steady state of the human body in the present application pre-processes the physiological signals such as electrocardiogram, respiration, body position / body movement collected by wearable devices, and then extracts features, and performs individualized modeling by the method of multivariate state estimation of the human body. Through matrix construction, data reconstruction and residual analysis and other methods, the human body state is distinguished between transition state and steady state, providing technical support for individualized physiological homeostasis monitoring, which is expected to further improve the application value of physiological data of wearable devices, and provide quantitative analysis support technical means for individualized disease management, rehabilitation training, and performance improvement. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 A flow chart of the method for distinguishing the transition state and steady state of the human body according to the present application; Figure 2 A physiological sequence state partition diagram based on external auxiliary information; Figure 3 It is a technical flow chart for the identification of transition and steady states of the human body; Figure 4 This is the residual change diagram of the plateau hypoxia experiment. DETAILED DESCRIPTION
[0019] The implementation environment of this application includes but is not limited to medical institutions, sports training centers, home or personal health monitoring, etc. The required equipment includes wearable devices, non-invasive physiological signal sensors, and data processing and analysis software.
[0020] It should be noted here that the contents of each module of the present application are the corresponding functional modules implemented when the method of distinguishing between the steady state and transition state of the human body of the present application is run on a computing device (such as a smart phone, tablet, computer or server, etc.).
[0021] Below, the method for distinguishing the steady state and transition state of the human body of the present application is described 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, respiratory belts, oximeters, etc., to continuously collect physiological signals for at least 24 hours, and remove unusable data segments through signal quality assessment algorithms. Figure 1 shown.
[0023] 2. Modeling feature calculation and screening: A set of features is extracted from the original physiological signal every t minutes, including heart rate, heart rate variability, respiratory rate, respiratory rate variability, pulse rate, blood oxygen saturation, cardiopulmonary coupling, body posture, activity intensity and other dimensional parameters as modeling feature parameters. The modeling feature parameters are further processed, including removing noise and outliers, missing value interpolation, aggregation analysis, etc., to screen out the available 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: Based on body position / body movement information, the state is divided into Figure 2 As shown in a in the figure, the activity intensity is quantified by the three-axis accelerometer signal, and the activity intensity is divided into static, low-intensity activity and high-intensity activity states, and H0 is divided into subspaces in different activity states.
[0025] The state classification based on sleep information is as follows: Figure 2 As shown in b, the sleep stage detection is performed by integrating physiological signals of various dimensions, which are divided into light sleep, deep sleep and REM states, and H0 is divided into subspaces in different sleep states.
[0026] In addition, according to the data situation, H0 can also be segmented according to the time series subspace separation technology, and several subspaces {D (1) , D (2) ,…D (m)}, generate the state matrix {D (Sl)}. Here {D (Sl)}=A(H0), A represents an algorithm or rule used to map H0 into several state subspaces.
[0027] 4. Reconstruct new observation data X obs : For new observation data X obs , according to the size of the Gaussian kernel in {D (Sl)}Select and X obs Several samples with the highest scores constitute the matching matrix D obs , and generate the matching state sequence S corresponding to the matching matrix obs ; The reconstruction result is obtained by solving the following optimization problem. The specific calculation process is as follows: ; Where C is the encoding coefficient matrix, λ>0 is one of the trade-off parameters, and D={D (1) , D (2) ,…D (m)},or , where δ>0 is some permissible level of interpretation error, or , Where α>0 is a constraint on the rationality of the interpretation, and r(C) is a structured sparse regularization term corresponding to the structure in D. By solving the optimization problem, the observed data are reconstructed to form the estimated data X est .
[0028] 5. Calculate the residual: Calculate the residual sequence of observed data and estimated data: R=Xobs-Xest 6. Identification of transition and steady states The flow chart of human body transition state and steady state identification is as follows Figure 4 As shown, if the residual sequence R satisfies the zero-mean Gaussian distribution with low variance and matches the state sequence S obs With stability and consistency, it means that the system is in a steady state and its observation data X obs Can be individualized physiological state representation matrix {D (Sl)} effective explanation, it can be judged that the physiological system is at the original steady-state level; conversely, if (a) The mean of the residual sequence R is still close to zero, but the variance increases significantly, which is 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 appears. In this case, the individual physiological state representation matrix {D (Sl)}.
[0029] Examples A specific embodiment is provided below to demonstrate the application of the method of the present invention: Example environment: A volunteer was selected to conduct a simulated plateau oxygen chamber experiment, including about 20 hours of plain baseline physiological data and about 2 hours of simulated plateau data, and his physiological signals were continuously collected using wearable devices.
[0030] Data processing flow: The plain baseline data of the first 2 hours is used as the initial data, and preprocessing and feature extraction are performed with a window of 1 minute. The features include heart rate, respiratory rate and activity-related parameters to construct H0. H0 is segmented by sleep stage and activity to form D obs .
[0031] The remaining plain baseline data is taken as X obs , for X obs Any single observation vector in D is matched to its corresponding sleep and activity rating according to its sleep and activity rating. obs , and reconstruct the observation vector to obtain the optimal 10 historical vectors, and obtain the reconstruction value X est , calculate the residual R.
[0032] Result analysis: After entering the simulated plateau environment, Figure 4 As shown in the figure, the residual analysis found that the R variance increased significantly, which was determined to be a transition state.
[0033] After a period of time, the R variance decreases and stabilizes, and by updating {D (Sl)}, it is determined to be a new steady state.
[0034] Implementation effect: Through the above embodiments, the method of the present invention can effectively distinguish the steady state and transition state of the human body, and provides a new technical means for individualized physiological state monitoring.
[0035] The method of the present application can distinguish the steady state and transition state of the human body through individualized modeling and analysis of continuous physiological data. The individualized modeling method of the present application solves the limitations of traditional methods in physiological state identification and improves the accuracy and individual adaptability of state identification; the construction of the individualized physiological state representation matrix proposed in the present application can solve the state aliasing problem and improve the accuracy of state identification; the reconstruction and residual analysis method of the new observation data adopted in the present application can achieve sensitivity to small state changes and the ability to fit the state of complex human physiological systems through data-driven; the dynamic update mechanism of state identification proposed in the present application can dynamically update the individualized physiological state representation matrix according to the residual analysis results to adapt to changes in human physiological state. The application results of the present application show that the method of the present application can effectively distinguish changes in human state, with good test results, and is suitable for 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 the invention relates. The materials, methods and examples mentioned in this application are illustrative only and not limiting.
[0037] Although the present invention has been described in conjunction with specific embodiments, within the spirit of the invention of this application, those skilled in the art may make appropriate substitutions, modifications and changes, and such substitutions, modifications and changes still fall within the scope of protection of this application.
Claims
1. A method for identifying a physiological steady state and a transition state of a human body, comprising: Obtain modeling features through at least 24 hours of physiological signals collected from the subjects, and use the modeling features to construct a historical state matrix H0; Using temporal subspace segmentation technology and external auxiliary information, an individualized physiological state representation matrix {D (Sl) }; For the observed data X obs Reconstruct the structured sparse coding framework to obtain the estimated data X est , and match the state sequence; Calculate the observed data X obs With the estimated data X est The residual sequence R between them; If the residual sequence R satisfies the zero-mean Gaussian distribution with low variance, the physiological system is judged to be in the original steady state; if the variance of R shows a trend of gradual increase, the state is a transitional state; if the mean of R deviates from zero but is relatively stable, and the variance does not change much, it indicates that a physiological state that was not observed when H0 was constructed has occurred, and the individualized physiological state representation matrix {D (Sl) }; If R first shows a gradual increase trend and then shows a relatively stable state after a period of time, it is judged that a new steady state has appeared in the physiological system.
2. The method for identifying the physiological steady state and transition 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 steady state and transition state of the human body according to claim 2, characterized in that: The modeling features include basic vital sign parameters and derived parameters from physiological signals.
4. The method for identifying the physiological steady state and transition state of the human body according to claim 1, characterized in that: The external auxiliary information includes: human body posture and activity status information, and sleep stage information.
5. The method for identifying the physiological steady state and transition 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 steady state and transition state of the human body according to claim 5, characterized in that: The modeling features have been processed to remove noise and outliers, interpolate missing values, and perform aggregation analysis.
7. The method for identifying the physiological steady state and transition state of the human body according to claim 4, characterized in that: According to the different activity intensities, the activity states of the subjects are divided into static, low-intensity activity and high-intensity activity states; according to the activity state information of the subjects, the historical state matrix H0 is divided into subspaces in different activity states.
8. The method for identifying the physiological steady state and transition state of the human body according to claim 4, characterized in that: The sleep staging information is obtained through sleep staging detection, including light sleep, deep sleep and REM state; the historical state matrix H0 is divided into subspaces in different sleep states according to the sleep staging information of the subject.
9. The method for identifying the physiological steady state and transition 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.
Citation Information
Patent Citations
Multivariate residual-based health index for human health monitoring
CN102917661A
Method for automatically detecting human physiological states and movement bracelet
CN106388771A
Multivariate Residual-Based Health Index for Human Health Monitoring
US20110172504A1
Brain state advisory system using calibrated metrics and optimal time-series decomposition
US20160242690A1
System and method for characterizing and tracking aging, resilience, cognitive decline, and disorders using brain dynamic biomarkers
WO2024092277A1
Cited By
Method for identifying physiological steady state and transition state of human body
WO2026061176A1
