A data-driven measurement system for risk of hypoxic damage at high altitude

Abstract

The application discloses a data-driven plateau hypoxia injury risk measurement system and method, belongs to the technical field of human hypoxia injury risk assessment, and comprises an intermittent hypoxia training module, a wearable physiological signal collection module, a hypoxia cabin test module, a centralized data storage and learning module, an edge data processing and hypoxia injury risk assessment module and a communication and transmission module.The data-driven plateau hypoxia injury risk measurement system and method adopt the above structure, realize the function of low-cost, rapid and dynamic assessment of the hypoxia injury risk of subjects, and can assess the hypoxia injury risk of subjects in a safer, faster and lower-cost manner without actual entry of the subjects into a plateau environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human hypoxia injury risk assessment, in particular to a data-driven plateau hypoxia injury risk measurement system. BACKGROUND

[0002] China has vast high-altitude regions that occupy one-fourth of the country's land area, which is of great importance in military, economic and social strategies and is rich in resources. However, the natural environment in this region is extremely harsh, facing a series of climate problems such as low oxygen and low pressure, dry and cold, strong wind and strong ultraviolet light, etc. The construction and improvement of the Qinghai-Tibet Railway have promoted the vigorous development of tourism in the plateau region, attracting a large number of tourists from low-altitude areas. However, these tourists may suffer a series of dangerous symptoms due to their inability to adapt to the plateau environment, including headaches, nausea, coma and even death. The health problems of low-altitude residents entering the plateau have become increasingly prominent. In addition, although the residents of high-altitude regions have a certain physical adaptability to the hypoxic environment, their physiological state has always been at an undesirable level. Compared with people in normal altitude areas, their blood oxygen saturation and heart rate are too low and too fast, respectively, which easily leads to the occurrence of chronic high-altitude disease. Therefore, it is a crucial task to ensure and improve the health quality of people in high-altitude regions.

[0003] A feasible measure to prevent acute high-altitude disease is to test the risk of high-altitude hypoxia injury before personnel enter the plateau. The currently available methods mainly include two categories. One category uses a self-scoring questionnaire, and the other mainly uses physiological indicators to predict high-altitude disease.

[0004] However, the above prediction and evaluation methods require the subjects to actually enter the plateau or to be tested for a long time in a simulated plateau environment. Such methods have long testing time and high cost. Based on this, the present application proposes a data-driven plateau hypoxia injury risk measurement system. SUMMARY

[0005] The purpose of the present application is to provide a data-driven plateau hypoxia injury risk measurement system that can realize the function of low-cost, rapid and dynamic evaluation of the hypoxia injury risk of subjects.

[0006] To achieve the above purpose, the present application provides a data-driven plateau hypoxia injury risk measurement system, which comprises:

[0007] An intermittent hypoxia training module controls the oxygen content in the air inhaled by the subject;

[0008] A wearable physiological signal acquisition module monitors the physiological indicator data related to the risk of hypoxia injury of the subject in real time;

[0009] The hypoxic chamber test module generates a hypoxic environment and monitors various indicators of the subject's sleep process.

[0010] The centralized data storage and learning module collects physiological indicator data, as well as sleep indicators and Lake Louise scores, to build a relationship between physiological data and hypoxic injury risk during intermittent hypoxic training.

[0011] The edge data processing and hypoxic injury risk assessment module assesses the hypoxic injury risk of the subject.

[0012] The communication and transmission module serves as a bridge for communication between all modules.

[0013] Preferably, the intermittent hypoxic training module includes a hypoxic generator for generating hypoxic gas, a hypoxic storage for storing hypoxic gas, and a gas mixer for mixing hypoxic gas and air.

[0014] Preferably, the wearable physiological signal acquisition module uses wearable devices to measure the subject's blood oxygen saturation, heart rate, and blood pressure.

[0015] Preferably, the hypoxic chamber test module has pressure sensors in the mattress and pillow for monitoring the subject's wake-up frequency, turning frequency, and deep sleep time.

[0016] The present application also provides a data-driven high-altitude hypoxic injury risk measurement method, comprising the following steps:

[0017] S1, generating a hypoxic environment and collecting hypoxic environment data;

[0018] S2, inputting the hypoxic environment into a hypoxic chamber, and the subject enters the hypoxic chamber for testing, and monitoring and recording the subject's continuous physiological data;

[0019] S3, investigating the subject's Lake Louise score and assessing the subject's hypoxic injury degree in combination with the continuous physiological data;

[0020] S4, collecting the hypoxic environment data obtained in S1, as well as the subject's Lake Louise score and hypoxic injury degree in S3, using the centralized data storage and learning module to learn the subject's hypoxic injury risk virtual measurement data through a data-driven method, and obtaining an estimation model;

[0021] S5, transmitting the continuous physiological data collected in S2 and the estimation model obtained in S4 to the edge data processing and hypoxic injury risk assessment module, and assessing the subject's physiological data for hypoxic injury risk through the estimation model.

[0022] Preferably, the process of generating a hypoxic environment in S1 is as follows:

[0023] S11, precisely control oxygen concentration and breathing gas ratio to generate hypoxic gas using a hypoxic generator, and provide appropriate intermittent hypoxic environment;

[0024] S12, store hypoxic gas generated by the hypoxic generator using a hypoxic storage device;

[0025] S13, send the hypoxic gas into a gas mixer, and mix the hypoxic gas and air through the gas mixer to obtain a hypoxic environment.

[0026] Preferably, the process of obtaining continuous physiological data in S2 is as follows:

[0027] S21, input the hypoxic environment obtained in S13 into a hypoxic cabin, and let the subject sleep in the hypoxic cabin;

[0028] S22, use a wearable device to non-invasively measure and record the blood oxygen saturation and heart rate of the subject based on optical sensing technology;

[0029] S23, use a wearable device to non-invasively measure and record the blood pressure of the subject through pulse wave sensing technology;

[0030] S24, monitor and record the sleep state of the subject through detection equipment in the hypoxic cabin.

[0031] Preferably, the process of investigating the Lewis Lake score and the degree of hypoxic injury of the subject in S3 is as follows:

[0032] S31, the subject wakes up and obtains the Lewis Lake score of the subject through a questionnaire;

[0033] S32, evaluate the degree of hypoxic injury of the subject based on the Lewis Lake score and the blood oxygen saturation, heart rate, blood pressure, and sleep state of the subject.

[0034] Preferably, the process of obtaining an estimation model through a centralized data storage and learning module in S4 is as follows:

[0035] S41, reduce the dimensionality of the hypoxic injury risk physiological indicators through principal component analysis, the process being as follows:

[0036] Let the obtained K hypoxic environment measured data be H={η 1 (k),…,η K (k)}, and perform a decentralization operation on the data,

[0037]

[0038] wherein η(k) represents the hypoxic injury risk physiological indicator, represents the measured data after decentralization, denotes The eigenvalues in the covariance matrix on the measured data set are sorted in descending order as is the eigenvector of each eigenvalue, P η is the transformation matrix;

[0039] According to the centralized measured data and the transformation matrix P η , the K low oxygen damage risk state data x i (k) after dimensionality reduction is obtained by the following formula SFA

[0040]

[0041] S42, slow feature analysis is performed by solving the variational optimization method, and the variational optimization process is as follows:

[0042]

[0043] Some constraints are imposed on the data in the variational optimization process, and the constraint content is as follows:

[0044]

[0045] where g SFA (·):γ k (t)→∈ k (t) represents the function variable to be optimized, and the final result obtained is the optimized slow feature signal ∈ k , γ k,i (t)=Γ i,t (k), represents an n Γ dimensional time series signal, and Γ(k) represents the continuous physiological data signal of the subject on the kth day, represents a dimensionality reduction signal, represents the continuous physiological data of the subject;

[0046] The physiological characteristic variable function of the subject during intermittent hypoxic training is constructed,

[0047] y(k):=τ(∈ k );

[0048] where y(k) represents the physiological monitoring signal feature during intermittent hypoxic training, τ(·) is a feature extraction function, and ∈ k represents the optimized slow feature signal;

[0049] S43, the training data set is obtained by physiological signal monitoring of the subject, Lewis Lake score and low oxygen damage degree The nonlinear function h(·) is learned, and a Gaussian process regression method is used to obtain an estimated model

[0050]

[0051] where x' is data in a training data set , represents an expected function, κ(x, x') represents a covariance kernel function, represents a Gaussian process.

[0052] Preferably, the process of using the edge type data processing and the hypoxia injury risk assessment module to assess the hypoxia injury risk state in S5 is as follows:

[0053] S51, after receiving the estimated model obtained in S4 , the hypoxia injury risk state is assessed according to an extended Kalman filter estimator, first state prediction and error covariance prediction are performed:

[0054]

[0055] where represents the prior predictive state of the hypoxia injury risk, and when there is no prior information represents the estimated hypoxia injury risk, represents a state evolution equation obtained from the prior information of the physiological development process of human hypoxia injury, represents the Jacobian matrix of the state prediction function, represents the covariance matrix of the state prediction error, and u(k-1) represents the input hypoxia signal, represents the covariance of , P(k-1) represents the state confidence at time k-1, represents the transpose of

[0056] S52, the prediction result obtained in S51 is brought into Kalman gain calculation,

[0057]

[0058]

[0059] where is a virtual measurement process error covariance matrix, H(k) represents a virtual measurement process Jacobian matrix, H T (k) represents the transpose of H(k), and K(k) represents the ratio of the reliability of the two data sources;

[0060] S53, updating the state and error covariance update using the results in S51 and S52,

[0061]

[0062] wherein represents the updated evaluation of the hypoxic injury risk state, is the final output of the injury risk at this moment, represents the expectation, P(k) represents the covariance at K moment, and is taken as the input in the error covariance calculation of the next moment, I represents the unit matrix.

[0063] Therefore, the data-driven high-altitude hypoxic injury risk measurement system with the above structure has the following advantages:

[0064] 1. Easy to deploy, high portability, the physiological signal acquisition module in the system adopts a wearable device, and the data storage learning and hypoxic injury risk evaluation are isolated, and after training is completed, only the edge data processing module can complete the subject information acquisition and hypoxic injury risk evaluation;

[0065] 2. The data-driven method is used to construct the hypoxic injury risk simulation measurement process, which avoids the complex mechanism modeling process, and better reflects the relationship between the intermittent hypoxic training physiological signal data of the group of subjects and the hypoxic injury risk in the hypoxic environment;

[0066] 3. The hypoxic injury risk of the subject can be evaluated in real time and dynamically, and the hypoxic injury risk of the subject can be evaluated in a safer, faster and lower cost manner without actually entering the high-altitude environment;

[0067] 4. A large amount of information can be stored and utilized, and an individualized physiological database can be created, long-term physiological health monitoring and prediction can be realized, and a basis for active health intervention when necessary is provided.

[0068] The technical solutions of the present application will be further described in detail below with the help of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 is a work flow diagram between each module of the data-driven high-altitude hypoxic injury risk measurement system of the present application;

[0070] Figure 2 is a work process diagram of the data-driven high-altitude hypoxic injury risk measurement system of the present application;

[0071] Figure 3 is an oxygen supply regulation flow diagram of the data-driven high-altitude hypoxic injury risk measurement system of the present application;

[0072] Figure 4A system implementation method flow chart of the data-driven plateau hypoxia damage risk measurement system. DETAILED DESCRIPTION

[0073] EMBODIMENT

[0074] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0075] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0076] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0077] In the description of the present application, it should be noted that the terms "upper", "lower", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the present application is usually placed, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0078] In the description of the present application, it should also be noted that, unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection" should be understood broadly, for example, it can be fixedly connected, or detachably connected, or integrally connected; it can be mechanically connected, or electrically connected; it can be directly connected, or indirectly connected through an intermediate medium, or the internal communication of two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0079] As Figures 1-4As shown, the data-driven plateau hypoxia injury risk measurement system of the present application comprises an intermittent hypoxic training module, a wearable physiological signal acquisition module, a hypoxic chamber test module, a centralized data storage and learning module, an edge data processing and hypoxic injury risk assessment module, and a communication and transmission module.

[0080] The intermittent hypoxic training module comprises a hypoxic generator for generating hypoxic gas, a hypoxic storage for storing hypoxic gas, and a gas mixer for mixing hypoxic gas and air.

[0081] The wearable physiological signal acquisition module uses wearable devices to measure the blood oxygen saturation, heart rate and blood pressure of the subject.

[0082] The hypoxic chamber test module is provided with pressure sensors in the mattress and pillow for monitoring the number of awakenings, the frequency of turning over and the deep sleep time of the subject.

[0083] The present application also provides a data-driven plateau hypoxia injury risk measurement method, comprising the following steps:

[0084] S1, generating a hypoxic environment and collecting hypoxic environment data;

[0085] S11, using a hypoxic generator to accurately control the oxygen concentration and the ratio of breathing gas to generate hypoxic gas and provide a suitable intermittent hypoxic environment;

[0086] S12, using a hypoxic storage to store the hypoxic gas generated by the hypoxic generator;

[0087] S13, sending the hypoxic gas into the gas mixer, and mixing the hypoxic gas and air through the gas mixer at a certain ratio and regularity to obtain a hypoxic environment.

[0088] S2, inputting the hypoxic environment into the hypoxic chamber, and the subject enters the hypoxic chamber for testing, and monitoring and recording the continuous physiological data of the subject;

[0089] S21, inputting the hypoxic environment obtained in S13 into the hypoxic chamber, and the subject sleeps in the hypoxic chamber;

[0090] S22, using wearable devices based on optical sensing technology to non-invasively measure and record the blood oxygen saturation and heart rate of the subject;

[0091] S23, using the wearable device to non-invasively measure and record the blood pressure of the subject through pulse wave sensing technology;

[0092] S24, monitoring and recording the sleep state of the subject through the detection device in the hypoxia chamber.

[0093] S3, investigating the Lewis Lake score of the subject, and evaluating the degree of hypoxic injury of the subject in combination with continuous physiological data;

[0094] S31, the subject wakes up and gets the Lewis Lake score of the subject through a questionnaire;

[0095] S32, in combination with the Lewis Lake score of the subject and the blood oxygen saturation, heart rate, blood pressure and sleep state, the degree of hypoxic injury of the subject is evaluated.

[0096] S4, collecting the hypoxic environment data obtained in S1, and the Lewis Lake score and the degree of hypoxic injury of the subject in S3, using a centralized data storage and learning module to learn the virtual measurement data of the risk of hypoxic injury of the subject through a data-driven method, and obtaining an estimation model, the process being as follows:

[0097] S41, reducing the dimensionality of the physiological indicators of the risk of hypoxic injury through the method of principal component analysis, the process being as follows:

[0098] Let the obtained K sets of measured data of hypoxic environment be H={η 1 (k),…,η K (k)},and the data set is decentered,

[0099]

[0100] wherein η(k) represents the physiological indicators of the risk of hypoxic injury, represents the measured data after decentering, represents the covariance matrix on the measured data set, the eigenvalues in are sorted from large to small as is the eigenvector of each eigenvalue, and P η is the transformation matrix;

[0101] According to the measured data after decentering and the transformation matrix P η , the K sets of reduced dimensionality of the risk of hypoxic injury state data x i (k) are obtained through the following formula,

[0102]

[0103] S42, slow feature analysis is performed by solving the variational optimization, and the solving process of the variational optimization is as follows:

[0104]

[0105] Some constraints are imposed on the data in the process of solving the variational optimization, and the contents of the constraints are as follows:

[0106]

[0107] Where g SFA (·):γ k (t)→∈ k (t) represents the function variable to be optimized, and the final result obtained is the optimized slow feature signal ∈ k , γ k,i (t)=Γ i,t (k), represents an n Γ dimensional time series signal, and Γ(k) represents the continuous physiological data signal of the subject on the kth day, represents a reduced dimension signal, represents the continuous physiological data of the subject;

[0108] The physiological feature variable function of the subject during intermittent hypoxic training is constructed,

[0109] y(k):=τ(∈ k );

[0110] Where y(k) represents the physiological monitoring signal feature during intermittent hypoxic training, τ(·) is a feature extraction function, and ∈ k represents the optimized slow feature signal;

[0111] S43, by monitoring the physiological signals of the subject and the Lake Louise score and the degree of hypoxic injury, a training data set is obtained Through the training data set The nonlinear function h(·) is learned, and a Gaussian process regression method is used to obtain an estimation model

[0112]

[0113] Where x' is the data in the training data set , represents an expectation function, and κ(x,x') represents a covariance kernel function, represents a Gaussian process.

[0114] S5, the continuous physiological data collected in S2 and the estimation model obtained in S4 are transmitted to an edge data processing and hypoxic injury risk assessment module, and the physiological data of the subject is assessed for hypoxic injury risk by the estimation model, and the process is as follows:

[0115] S51, receiving the estimated model obtained in S4 After that, the evaluation of the hypoxic injury risk state is obtained according to the extended Kalman filter estimator, and first, state prediction and error covariance prediction are performed:

[0116]

[0117] wherein represents the prior prediction state of the hypoxic injury risk, and when there is no prior information represents the estimated hypoxic injury risk, represents the state evolution equation obtained from the prior information of the physiological development process of the human hypoxic injury, represents the Jacobian matrix of the state prediction function, represents the covariance matrix of the state prediction error, and u(k-1) represents the input hypoxic signal, represents the covariance, and P(k-1) represents the state confidence at k-1 time, represents the transpose of

[0118] S52, the prediction result obtained in S51 is brought into the Kalman gain calculation,

[0119]

[0120] wherein is the virtual measurement process error covariance matrix, H(k) represents the virtual measurement process Jacobian matrix, H T (k) represents the transpose of H(k), and K(k) represents the ratio of the reliability of the two data sources;

[0121] S53, state updating and error covariance updating are performed using the results in S51 and S52,

[0122]

[0123] wherein represents the updated evaluation of the hypoxic injury risk state, and is the final output injury risk at this time, represents the expectation, P(k) represents the covariance at K time, and is taken as the input in the error covariance calculation at the next time, and I represents the unit matrix.

[0124] Therefore, the present invention provides a data-driven high-altitude hypoxia injury risk measurement system with the above-described structure. This system offers low-cost, rapid, and dynamic assessment of subjects' hypoxia injury risk. Furthermore, the use of wearable devices to isolate data storage and learning from hypoxia injury risk assessment provides advantages such as ease of deployment and high portability. The data-driven approach to constructing a hypoxia injury risk simulation process better reflects the relationship between physiological signal data from intermittent hypoxia training and hypoxia injury risk in hypoxic environments for the group of subjects. Moreover, it allows for a safer, faster, and lower-cost assessment of subjects' hypoxia injury risk without them actually entering a high-altitude environment. Finally, it can store and utilize a large amount of information to create an individualized physiological database, enabling long-term physiological health monitoring and forecasting, and providing a foundation for proactive health intervention when necessary.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A data driven high altitude hypoxia injury risk measurement system characterized by: Comprise: Intermittent hypoxic training module, control oxygen content, including for generating low oxygen gas low oxygen generator, low oxygen gas storage low oxygen storage and mixing low oxygen gas and air gas mixer, low oxygen storage low oxygen gas generated by low oxygen generator, low oxygen gas in low oxygen storage into the gas mixer, by gas mixer low oxygen gas and air by a certain proportion and regular mixing get low oxygen environment; Wearable physiological signal acquisition module, real-time monitoring of physiological indicators related to hypoxic injury risk data; Hypoxic chamber test module, generate low oxygen environment and monitor the indicators of sleep process; Centralized data storage and learning module, collect physiological indicators, and sleep indicators and Lake Louise score, build the relationship between intermittent hypoxic training physiological data and hypoxic injury risk; Centralized data storage and learning module through principal component analysis method for dimension reduction of hypoxic injury risk physiological indicators η(k), the process is as follows: The obtained K low-oxygen environment measured data is H = {η 1 (k),…,η K (k)}, and a decentralized operation is performed on the data set, wherein η(k) represents a hypoxia injury risk physiological index, denotes the measured data after decentralization, denotes the covariance matrix on the measured data set, the eigenvalues in are sorted from large to small as is the eigenvector of each eigenvalue, and P η is a conversion matrix; According to the centralized measured data and the conversion matrix P η , the K low-oxygen damage risk state data x i (k) after dimension reduction is obtained by the following formula After dimension reduction, the centralized data storage and learning module solves the slow feature analysis of the variation optimization method, the process is as follows: Some constraints are put on the data in the process of solving the variation optimization, the constraint content is as follows: where g SFA (·): γ k (t)→∈ k (t) represents the function variable to be optimized, and the final result is the slow feature signal∈ k , γ k,i (t) = Γ i,t (k), represents the n Γ -dimensional time series signal, and Γ(k) represents the continuous physiological data signal on the kth day, represents the reduced dimension signal, represents the continuous physiological data; Constructing the physiological characteristic variable function of intermittent hypoxic training, y(k) := τ(∈ k ); where y(k) represents the physiological monitoring signal features during intermittent hypoxia training, τ(·) is a feature extraction function, ∈ k represents the optimized slow feature signal; Finally, the centralized data storage and learning module learns the nonlinear function h(·) by using a Gaussian process regression method to obtain an estimation model Finally, the centralized data storage and learning module learns the nonlinear function h(·) by using a Gaussian process regression method to obtain an estimation model where x' is data in the training data set represents the expected function, K(x, x') represents the covariance kernel function, represents a Gaussian process;​ Edge data processing and hypoxic injury risk assessment module, assess the risk of hypoxic injury; The edge data processing and hypoxia injury risk assessment module obtains the estimated model constructed by the centralized data storage and learning module Afterwards, first, the estimation of the hypoxia injury risk state is obtained according to the extended Kalman filter estimator, state prediction and error covariance prediction are performed: wherein represents a prior prediction state of the risk of hypoxic injury, in the absence of prior information represents an estimated risk of hypoxic injury, represents a state evolution equation obtained from prior information of the physiological development process of hypoxic injury of the human body, represents a Jacobian matrix of the state prediction function, represents a covariance matrix of the state prediction error, u(k-1) represents an input hypoxic signal, represents a covariance, P(k-1) represents a state confidence at time k-1, represents a transpose; Then the predicted results into the Kalman gain calculation, wherein is the virtual measurement process error covariance matrix, H(k) represents the virtual measurement process Jacobian matrix, H T (k) represents the transpose of H(k), K(k) represents the ratio of the two data source reliabilities; Finally, the state update and error covariance update, wherein represents the updated assessed hypoxic damage risk state, is the final output damage risk at this time instant, represents the expectation, P(k) represents the covariance at time k, and is used as input in the computation of the error covariance at the next time instant, I represents the identity matrix; Communication and transmission module, the bridge between all modules.

2. The data driven high altitude hypoxia injury risk measurement system of claim 1, wherein: Wearable physiological signal acquisition module uses wearable devices to collect blood oxygen saturation, heart rate and blood pressure.

3. A data driven high altitude hypoxia injury risk measurement system according to claim 2, wherein: The mattress and pillow of the hypoxic chamber test module are provided with pressure sensors for monitoring the number of awakenings, turning frequency and deep sleep time.

4. A data driven high altitude hypoxia injury risk measurement system according to claim 3, wherein: The wearable physiological signal acquisition module uses optical sensing technology to non-invasively collect blood oxygen saturation and heart rate, and also uses pulse wave sensing technology to non-invasively collect blood pressure, while recording blood oxygen saturation, heart rate and blood pressure.

5. A data driven high altitude hypoxia injury risk measurement system according to claim 4, wherein: The low oxygen environment generated by the intermittent hypoxic training module is input into the hypoxic chamber test module, which is used to obtain the sleep state.

6. A data driven high altitude hypoxia injury risk measurement system according to claim 5, wherein: The questionnaire is used to obtain the Lake Louise score, and the hypoxia injury degree is obtained in combination with the Lake Louise score, blood oxygen saturation, heart rate, blood pressure and sleep state, and the training data set is constructed according to the physiological signals, the Lake Louise score and the hypoxia injury degree

Images