Portable cardiopulmonary function monitoring and risk early warning system and method in extreme environment
By constructing an individualized interaction influence matrix and risk assessment model, and combining sliding window technology and Bayesian classifiers, the problems of individual differences and environmental complexity in cardiopulmonary function monitoring under extreme environments were solved, achieving efficient risk warning and personalized monitoring.
Patent Information
- Application Number
- CN202511045026.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-29
- Publication Date
- 2025-11-04
AI Technical Summary
Existing technologies struggle to dynamically correlate environmental parameters with changes in cardiopulmonary function in extreme environments, and cannot adapt to complex and ever-changing environmental factors and individual differences. This results in insufficient accuracy and timeliness of risk warnings, making it difficult to effectively prevent emergencies such as acute altitude sickness.
By integrating historical cardiopulmonary data, physical characteristics, and environmental parameters through the data acquisition module, an individualized interaction influence matrix is constructed. Combined with the physiological monitoring module, cardiopulmonary function indicators are collected in real time. Using CEEMD decomposition and PCA dimensionality reduction, a personalized risk assessment model is constructed, and anomaly warning is performed using sliding window technology and a Bayesian classifier.
It enables precise monitoring of dynamic changes in individual cardiopulmonary function under extreme environments, reduces false alarms, improves the accuracy and timeliness of risk warnings, reduces equipment noise interference, and extends equipment operating time.
Smart Images
Figure CN120895243A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of risk assessment technology, specifically to a portable cardiopulmonary function monitoring and risk warning system and method for extreme environments. Background Technology
[0002] In extreme environments, such as high altitudes, deep seas, or high-temperature locations, workers face significant challenges to their cardiopulmonary function, making their health and safety paramount. Cardiopulmonary function monitoring and risk warning technologies play an irreplaceable role in protecting personnel safety and improving operational efficiency. However, existing methods have significant limitations in addressing the dynamic changes in extreme environments and individual differences. Many solutions rely on single environmental parameters or general models, making it difficult to adapt to complex and changing environmental factors and accurately capture dynamic changes in individual cardiopulmonary function. This results in limited accuracy and timeliness of risk warnings in extreme environments, hindering effective prevention of emergencies such as acute altitude sickness, decompression sickness, or heatstroke.
[0003] The core challenge lies in how to integrate multi-dimensional data to construct a predictive model that dynamically correlates environmental parameters with changes in cardiopulmonary function. First, the complexity of extreme environments means that a single environmental parameter cannot fully reflect the decline in cardiopulmonary function. For example, high-altitude hypoxia, deep-sea high pressure, or high-temperature and high-humidity environments can have cumulative effects on the cardiopulmonary system, and current technologies struggle to analyze the interactions between these parameters in real time. Second, due to individual differences, the same environmental parameter has varying degrees of impact on the health of different individuals, and general models cannot provide personalized risk assessments. These two factors are interconnected: the complexity of the interactions between environmental parameters requires models to have dynamic analytical capabilities, while individual differences further increase the difficulty of constructing accurate models.
[0004] Therefore, the key issue of this study is how to construct a risk prediction model that can analyze the correlation between multidimensional environmental parameters and cardiopulmonary function decline in real time based on historical data and individual characteristics, and provide personalized risk level assessment and prevention suggestions. Summary of the Invention
[0005] To address the above technical problems, this invention proposes a portable cardiopulmonary function monitoring and risk warning system for extreme environments, the method of which specifically includes:
[0006] The data acquisition module is used to collect the cardiopulmonary history data of existing workers. The cardiopulmonary history data includes basic cardiopulmonary function indicators, physical characteristic parameters, and environmental adaptation records.
[0007] The data association module is used to collect environmental parameters in extreme environments and calculate the interaction influence matrix based on the environmental parameters.
[0008] The physiological monitoring module is used to collect the cardiopulmonary function indicators of workers in real time and obtain the cardiopulmonary changes based on the workers' real-time cardiopulmonary function indicators.
[0009] The environmental risk early warning module is used to obtain individualized abnormal state identifiers based on the interaction influence matrix and the cardiopulmonary change, and to complete portable cardiopulmonary function monitoring and risk early warning in extreme environments based on the individualized abnormal state identifiers.
[0010] Optionally, the data acquisition module includes a basic parameter acquisition unit, an adaptive record acquisition unit, and a preprocessing unit;
[0011] The basic parameter acquisition unit is used to collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume.
[0012] The adaptive recording acquisition unit is used to record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction;
[0013] The preprocessing unit is used to perform dimensionality reduction and data cleaning on the data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit to remove high-frequency noise, and then perform discrete wavelet filtering on the data after removing high-frequency noise to correct baseline drift and obtain preprocessed data.
[0014] Optionally, the data association module includes an environmental acquisition unit, a fusion analysis unit, and a matrix calculation module;
[0015] The environmental acquisition unit is used to collect temperature, humidity, air pressure, oxygen concentration and wind speed in extreme environments in real time using a multi-dimensional sensor array;
[0016] The fusion analysis unit is used to preprocess the data acquired by the environmental acquisition unit to obtain a complete dataset, and to perform time-series analysis on the complete dataset using the sliding window method;
[0017] The matrix calculation module is used to calculate the interaction effect matrix using the Pearson correlation coefficient based on the time series analysis results.
[0018] Optionally, the preprocessing in the fusion analysis unit specifically includes:
[0019] The 3σ principle and the isolated forest algorithm are used to remove outliers from the data acquired by the environmental acquisition unit to obtain the first processed data.
[0020] The missing values in the first processed data are processed using the cubic spline interpolation method and the Kalman smoothing algorithm to obtain the second processed data.
[0021] The second-processed data was formatted using Z-score, and the complete dataset was calculated using principal component analysis.
[0022] Optionally, the environmental risk early warning module includes a momentum calculation unit, a risk assessment unit, and a risk early warning unit;
[0023] The momentum calculation unit is used to calculate the dynamic changes of the cardiopulmonary system of the operator under extreme conditions based on the interaction influence matrix and the cardiopulmonary change.
[0024] The risk assessment unit is used to construct a risk assessment model based on the existing cardiopulmonary historical data of the workers, and input the dynamic changes of the workers' cardiopulmonary system into the risk assessment model to obtain risk prediction results;
[0025] The risk warning unit is used to provide early warning of cardiopulmonary abnormalities to the current workers based on the risk prediction results.
[0026] This invention also discloses a portable method for monitoring and warning of cardiopulmonary function in extreme environments, the method comprising:
[0027] Collect historical cardiopulmonary data from existing workers, including basic cardiopulmonary function indicators, physical fitness parameters, and environmental adaptation records.
[0028] Collect environmental parameters in extreme environments, and calculate the interaction effect matrix based on the environmental parameters;
[0029] Real-time collection of cardiopulmonary function indicators of workers, and acquisition of cardiopulmonary changes based on real-time cardiopulmonary function indicators of workers;
[0030] Based on the interaction influence matrix and the cardiopulmonary change, an individualized abnormal state identifier is obtained, and based on the individualized abnormal state identifier, portable cardiopulmonary function monitoring and risk warning in extreme environments are completed.
[0031] Optionally, the process of collecting historical cardiopulmonary data from existing workers specifically includes:
[0032] Collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume;
[0033] Record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction;
[0034] The data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit are subjected to dimensionality reduction and data cleaning to remove high-frequency noise. The data after removing high-frequency noise is then subjected to discrete wavelet filtering to correct baseline drift and obtain preprocessed data.
[0035] Optionally, the process of calculating the interaction influence matrix based on the environmental parameters specifically includes:
[0036] A multi-dimensional sensor array is used to collect temperature, humidity, air pressure, oxygen concentration, and wind speed in extreme environments in real time.
[0037] The data acquired by the environmental acquisition unit is preprocessed to obtain a complete dataset, and the sliding window method is used to perform time series analysis on the complete dataset.
[0038] Based on the time series analysis results, the interaction effect matrix was calculated using the Pearson correlation coefficient.
[0039] Optionally, the process of obtaining the complete dataset is as follows:
[0040] The 3σ principle and the isolated forest algorithm are used to remove outliers from the data acquired by the environmental acquisition unit to obtain the first processed data.
[0041] The missing values in the first processed data are processed using the cubic spline interpolation method and the Kalman smoothing algorithm to obtain the second processed data.
[0042] The second-processed data was formatted using Z-score, and the complete dataset was calculated using principal component analysis.
[0043] Optionally, the specific content of portable cardiopulmonary function monitoring and risk warning in extreme environments based on the individualized abnormal state identifier includes:
[0044] Based on the interaction matrix and the cardiopulmonary change, the dynamic changes of the worker's cardiopulmonary system under extreme conditions are calculated.
[0045] A risk assessment model is constructed based on the existing cardiopulmonary history data of the workers. The dynamic changes in the cardiopulmonary function of the workers are input into the risk assessment model to obtain the risk prediction results.
[0046] Based on the risk prediction results, early warning of cardiopulmonary abnormalities is issued to the current workers.
[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0048] This invention integrates historical cardiopulmonary indicators (such as rMSSD, LF / HF), physical characteristics (body temperature, skin temperature), and environmental adaptation records (PMV / PPD) through a data acquisition module. After CEEMD decomposition and PCA dimensionality reduction, an individualized baseline model is constructed. Compared to traditional fixed threshold warnings, this embodiment utilizes time series analysis to dynamically update the baseline, adapting to physiological differences among different workers and sudden environmental changes, reducing false alarms caused by individual differences (such as individuals with high baseline heart rates being misjudged as abnormal).
[0049] The physiological monitoring module of this invention calculates real-time deviation using sliding window technology, triggering a Bayesian classifier to combine historical data to determine abnormal states (such as hypoxic arrhythmia), and then uses cluster analysis to determine the abnormal category (acute altitude sickness / chronic fatigue). Based on this, the risk warning unit generates tiered warnings (such as a yellow warning indicating adjustment of work intensity, and a red warning mandating evacuation), reducing unnecessary work stoppages by more than 30% compared to traditional "one-size-fits-all" warnings.
[0050] The preprocessing unit of this invention employs discrete wavelet filtering and Kalman smoothing, reducing noise while compressing data volume by 90%, enabling portable devices (such as wrist monitors) to operate continuously for over 72 hours. The residual terms of CEEMD decomposition preserve long-term environmental trends, avoid high-frequency noise interference, and ensure monitoring stability in extreme scenarios such as high altitudes and deep seas. Attached Figure Description
[0051] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0052] Figure 1 This is a system structure diagram of a portable cardiopulmonary function monitoring and risk warning system for extreme environments provided in an embodiment of the present invention. Detailed Implementation
[0053] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0054] Example 1
[0055] A portable cardiopulmonary function monitoring and risk warning system for extreme environments, such as Figure 1 As shown, the system includes:
[0056] The data acquisition module is used to collect the cardiopulmonary history data of existing workers. The cardiopulmonary history data includes basic cardiopulmonary function indicators, physical characteristic parameters, and environmental adaptation records.
[0057] The data acquisition module includes a basic parameter acquisition unit, an adaptive record acquisition unit, and a preprocessing unit.
[0058] The basic parameter acquisition unit is used to collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume.
[0059] The adaptive recording acquisition unit is used to record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction;
[0060] Oral and skin temperatures (including forehead, chest, and back of hands) were measured using an infrared interrogation device for staff. Electrocardiograms were measured using a Holter monitor to calculate heart rate variability (rMSSD, root mean square of the difference between adjacent RR intervals) and LF / HF (low-frequency heart rate / high-frequency heart rate). The predicted mean thermal sensation index was calculated based on temperature, humidity, wind speed, mean radiant temperature, metabolic rate, and clothing thermal resistance. The predicted percentage of dissatisfaction was based on heart rate variability, respiratory rhythm, and body temperature.
[0061] The preprocessing unit is used to perform dimensionality reduction and data cleaning on the data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit to remove high-frequency noise, and then perform discrete wavelet filtering on the data after removing high-frequency noise to correct baseline drift and obtain preprocessed data.
[0062] The data association module is used to collect environmental parameters in extreme environments and calculate the interaction influence matrix based on the environmental parameters.
[0063] The data association module includes an environmental acquisition unit, a fusion analysis unit, and a matrix calculation module:
[0064] The environmental acquisition unit is used to collect temperature, humidity, air pressure, oxygen concentration and wind speed in extreme environments in real time using a multi-dimensional sensor array.
[0065] The fusion analysis unit is used to preprocess the data acquired by the environmental acquisition unit to obtain a complete dataset, and to perform time series analysis on the complete dataset using the sliding window method. The preprocessing in the fusion analysis unit specifically includes: using the 3σ principle and the isolated forest algorithm to remove outliers from the data acquired by the environmental acquisition unit to obtain first-processed data; using the cubic spline interpolation method and the Kalman smoothing algorithm to process missing values in the first-processed data to obtain second-processed data; using Z-score to unify the format of the second-processed data; and using principal component analysis to calculate the complete dataset from the format-unified data.
[0066] The matrix calculation module is used to calculate the interaction effect matrix using the Pearson correlation coefficient based on the time series analysis results.
[0067] The complete dataset is decomposed using CEEMD to obtain environmental subsequences. The specific process is as follows:
[0068] Add a pair of positive and negative white noise signals to the complete dataset. A new sequence of signals is obtained, denoted as Right now:
[0069]
[0070] right and EMD decomposition was performed separately to obtain different environmental subsequences and residual terms, specifically including:
[0071] The maxima and minima of the original environmental sequence signal O(t) are determined by cubic spline interpolation. and The upper and lower envelopes are calculated, and then the mean m(t) of the upper and lower envelopes is obtained. The sequence signal is then... The difference between each is m(t) and s(t):
[0072]
[0073] If the obtained s(t) satisfies the two conditions of IMF, namely: ① within the sequence signal interval, the number of zero crossings is equal to or differs from the number of extreme points by 1; ② the average of the envelopes of the local maximum and the local minimum is zero at any time; then s(t) can be defined as the i-th IMF sequence; similarly, the residual component r(t) is the difference between the decomposed O(t) and s(t);
[0074] Each time a different white noise signal is added, the process continues until no new IMF sequences can be extracted from sequence O(t); thus obtaining k sets of IMF sequences. and and a set of residual terms c ki (t) represents the i-th IMF sequence generated by the k-th addition of white noise; ultimately, we obtain:
[0075]
[0076] For the final and The sequences are first averaged and then summed to obtain k IMF sequences and one residual term from the CEEMD decomposition:
[0077]
[0078] By using CEEMD decomposition, the prediction error is significantly reduced.
[0079] High- and low-frequency reconstructions were performed using the Pearson correlation coefficient and the environmental subsequence. High-frequency terms were superimposed, low-frequency terms were superimposed, and the residual terms remained unchanged. The Pearson correlation coefficient was calculated as follows:
[0080]
[0081] Where X is any data point after decomposition; Y is data point that is different from X after decomposition. The average value of X; This represents the average value of Y.
[0082] The physiological monitoring module is used to collect the cardiopulmonary function indicators of workers in real time and obtain the cardiopulmonary changes based on the workers' real-time cardiopulmonary function indicators.
[0083] The environmental risk early warning module is used to obtain individualized abnormal state identifiers based on the interaction influence matrix and cardiopulmonary changes, and to complete portable cardiopulmonary function monitoring and risk early warning in extreme environments based on the individualized abnormal state identifiers.
[0084] The environmental risk early warning module includes a momentum calculation unit, a risk assessment unit, and a risk early warning unit;
[0085] The momentum calculation unit is used to calculate the dynamic changes in the cardiopulmonary system of workers under extreme conditions based on the interaction effect matrix and cardiopulmonary changes.
[0086] The risk assessment unit is used to build a risk assessment model based on the existing cardiopulmonary history data of workers. The dynamic changes in the cardiopulmonary function of workers are input into the risk assessment model to obtain risk prediction results.
[0087] The risk warning unit is used to provide early warning of cardiopulmonary abnormalities to the current workers based on the risk prediction results.
[0088] Personalized cardiopulmonary function data is acquired from cardiopulmonary function monitoring equipment, and dynamic change curves are extracted using time series analysis. Based on these dynamic change curves and a pre-established standardized feature dataset, principal component analysis is used to construct an individualized baseline model, resulting in a personalized baseline. Functional indicators are extracted from the individualized baseline model, and the real-time deviation of the dynamic change curves is calculated using a sliding window technique. If the real-time deviation exceeds a preset dynamic threshold, an anomaly detection mechanism is triggered, generating a preliminary anomaly signal. Based on the preliminary anomaly signal and historical cardiopulmonary function data, a Bayesian classifier is used to determine the individualized abnormal state, obtaining a state identifier. For each state identifier, associated dynamic change curves and functional indicators are acquired, and cluster analysis is used to determine the category of the abnormal state. Based on the category of the abnormal state and the standardized feature dataset, the personalized baseline model is updated, resulting in an optimized baseline model.
[0089] Example 2
[0090] Collect historical cardiopulmonary data from existing workers, including basic cardiopulmonary function indicators, physical characteristics parameters, and environmental adaptation records.
[0091] Collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume;
[0092] Record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction;
[0093] Oral and skin temperatures (including forehead, chest, and back of hands) were measured using an infrared interrogation device for staff. Electrocardiograms were measured using a Holter monitor to calculate heart rate variability (rMSSD, root mean square of the difference between adjacent RR intervals) and LF / HF (low-frequency heart rate / high-frequency heart rate). The predicted mean thermal sensation index was calculated based on temperature, humidity, wind speed, mean radiant temperature, metabolic rate, and clothing thermal resistance. The predicted percentage of dissatisfaction was based on heart rate variability, respiratory rhythm, and body temperature.
[0094] The data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit are subjected to dimensionality reduction and data cleaning to remove high-frequency noise. The data after removing high-frequency noise is then subjected to discrete wavelet filtering to correct baseline drift and obtain preprocessed data.
[0095] Environmental parameters in extreme environments are collected, and an interaction influence matrix is calculated based on these environmental parameters.
[0096] A multi-dimensional sensor array is used to collect temperature, humidity, air pressure, oxygen concentration, and wind speed in extreme environments in real time.
[0097] The data acquired by the environmental acquisition unit is preprocessed to obtain a complete dataset. A sliding window method is then used to perform time-series analysis on the complete dataset. The preprocessing in the fusion analysis unit specifically includes: using the 3σ principle and the isolated forest algorithm to remove outliers from the data acquired by the environmental acquisition unit, obtaining first-processed data; using the cubic spline interpolation method and the Kalman smoothing algorithm to process missing values in the first-processed data, obtaining second-processed data; using Z-score to unify the format of the second-processed data; and using principal component analysis to calculate the complete dataset from the format-unified data.
[0098] Based on the time series analysis results, the interaction effect matrix was calculated using the Pearson correlation coefficient.
[0099] The complete dataset is decomposed using CEEMD to obtain environmental subsequences. The specific process is as follows:
[0100] Add a pair of positive and negative white noise signals to the complete dataset. A new sequence of signals is obtained, denoted as Right now:
[0101]
[0102] right and EMD decomposition was performed separately to obtain different environmental subsequences and residual terms, specifically including:
[0103] The maxima and minima of the original environmental sequence signal O(t) are determined by cubic spline interpolation. and The upper and lower envelopes are calculated, and then the mean m(t) of the upper and lower envelopes is obtained. The sequence signal is then... The difference between each is m(t) and s(t):
[0104]
[0105] If the obtained s(t) satisfies the two conditions of IMF, namely: ① within the sequence signal interval, the number of zero crossings is equal to or differs from the number of extreme points by 1; ② the average of the envelopes of the local maximum and the local minimum is zero at any time; then s(t) can be defined as the i-th IMF sequence; similarly, the residual component r(t) is the difference between the decomposed O(t) and s(t);
[0106] Each time a different white noise signal is added, the process continues until no new IMF sequences can be extracted from sequence O(t); thus obtaining k sets of IMF sequences. and and a set of residual terms c ki (t) represents the i-th IMF sequence generated by the k-th addition of white noise; ultimately, we obtain:
[0107]
[0108] For the final and The sequences are first averaged and then summed to obtain k IMF sequences and one residual term from the CEEMD decomposition:
[0109]
[0110] By using CEEMD decomposition, the prediction error is significantly reduced.
[0111] High- and low-frequency reconstructions were performed using the Pearson correlation coefficient and the environmental subsequence. High-frequency terms were superimposed, low-frequency terms were superimposed, and the residual terms remained unchanged. The Pearson correlation coefficient was calculated as follows:
[0112]
[0113] Where X is any data point after decomposition; Y is data point that is different from X after decomposition. The average value of X; This represents the average value of Y.
[0114] Real-time collection of cardiopulmonary function indicators of workers, and acquisition of cardiopulmonary changes based on real-time cardiopulmonary function indicators of workers.
[0115] Individualized abnormal state identifiers are obtained based on the interaction effect matrix and cardiopulmonary changes, and portable cardiopulmonary function monitoring and risk warning in extreme environments are completed based on these individualized abnormal state identifiers.
[0116] Interaction effect matrix and calculation of cardiopulmonary changes: dynamic changes in the cardiopulmonary function of workers under extreme conditions;
[0117] A risk assessment model is constructed based on the existing cardiopulmonary history data of workers. The dynamic changes in the cardiopulmonary function of workers are input into the risk assessment model to obtain risk prediction results.
[0118] Based on the risk prediction results, early warning of cardiopulmonary abnormalities is issued to the current workers.
[0119] Personalized cardiopulmonary function data is acquired from cardiopulmonary function monitoring equipment, and dynamic change curves are extracted using time series analysis. Based on these dynamic change curves and a pre-established standardized feature dataset, principal component analysis is used to construct an individualized baseline model, resulting in a personalized baseline. Functional indicators are extracted from the individualized baseline model, and the real-time deviation of the dynamic change curves is calculated using a sliding window technique. If the real-time deviation exceeds a preset dynamic threshold, an anomaly detection mechanism is triggered, generating a preliminary anomaly signal. Based on the preliminary anomaly signal and historical cardiopulmonary function data, a Bayesian classifier is used to determine the individualized abnormal state, obtaining a state identifier. For each state identifier, associated dynamic change curves and functional indicators are acquired, and cluster analysis is used to determine the category of the abnormal state. Based on the category of the abnormal state and the standardized feature dataset, the personalized baseline model is updated, resulting in an optimized baseline model.
[0120] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims
1. A portable cardiopulmonary function monitoring and risk warning system for extreme environments, characterized in that, The system includes: The data acquisition module is used to collect the cardiopulmonary history data of existing workers. The cardiopulmonary history data includes basic cardiopulmonary function indicators, physical characteristic parameters, and environmental adaptation records. The data association module is used to collect environmental parameters in extreme environments and calculate the interaction influence matrix based on the environmental parameters. The physiological monitoring module is used to collect the cardiopulmonary function indicators of workers in real time and obtain the cardiopulmonary changes based on the workers' real-time cardiopulmonary function indicators. The environmental risk early warning module is used to obtain individualized abnormal state identifiers based on the interaction influence matrix and the cardiopulmonary change, and to complete portable cardiopulmonary function monitoring and risk early warning in extreme environments based on the individualized abnormal state identifiers.
2. The portable cardiopulmonary function monitoring and risk warning system for extreme environments according to claim 1, characterized in that, The data acquisition module includes a basic parameter acquisition unit, an adaptive record acquisition unit, and a preprocessing unit; The basic parameter acquisition unit is used to collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume. The adaptive recording acquisition unit is used to record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction; The preprocessing unit is used to perform dimensionality reduction and data cleaning on the data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit to remove high-frequency noise, and then perform discrete wavelet filtering on the data after removing high-frequency noise to correct baseline drift and obtain preprocessed data.
3. The portable cardiopulmonary function monitoring and risk warning system for extreme environments according to claim 1, characterized in that, The data association module includes an environmental acquisition unit, a fusion analysis unit, and a matrix calculation module; The environmental acquisition unit is used to collect temperature, humidity, air pressure, oxygen concentration and wind speed in extreme environments in real time using a multi-dimensional sensor array; The fusion analysis unit is used to preprocess the data acquired by the environmental acquisition unit to obtain a complete dataset, and to perform time-series analysis on the complete dataset using the sliding window method; The matrix calculation module is used to calculate the interaction effect matrix using the Pearson correlation coefficient based on the time series analysis results.
4. The portable cardiopulmonary function monitoring and risk warning system for extreme environments according to claim 3, characterized in that, The preprocessing in the fusion analysis unit specifically includes: The 3σ principle and the isolated forest algorithm are used to remove outliers from the data acquired by the environmental acquisition unit to obtain the first processed data. The missing values in the first processed data are processed using the cubic spline interpolation method and the Kalman smoothing algorithm to obtain the second processed data. The second-processed data was formatted using Z-score, and the complete dataset was calculated using principal component analysis.
5. The portable cardiopulmonary function monitoring and risk warning system for extreme environments according to claim 1, characterized in that, The environmental risk early warning module includes a momentum calculation unit, a risk assessment unit, and a risk early warning unit. The momentum calculation unit is used to calculate the dynamic changes of the cardiopulmonary system of the operator under extreme conditions based on the interaction influence matrix and the cardiopulmonary change. The risk assessment unit is used to construct a risk assessment model based on the existing cardiopulmonary historical data of the workers, and input the dynamic changes of the workers' cardiopulmonary system into the risk assessment model to obtain risk prediction results; The risk warning unit is used to provide early warning of cardiopulmonary abnormalities to the current workers based on the risk prediction results.
6. A portable method for monitoring cardiopulmonary function and providing risk warning in extreme environments, wherein the method is applied to the system described in any one of claims 1-5, characterized in that, The methods include: Collect historical cardiopulmonary data from existing workers, including basic cardiopulmonary function indicators, physical fitness parameters, and environmental adaptation records. Collect environmental parameters in extreme environments, and calculate the interaction effect matrix based on the environmental parameters; Real-time collection of cardiopulmonary function indicators of workers, and acquisition of cardiopulmonary changes based on real-time cardiopulmonary function indicators of workers; Based on the interaction influence matrix and the cardiopulmonary change, an individualized abnormal state identifier is obtained, and based on the individualized abnormal state identifier, portable cardiopulmonary function monitoring and risk warning in extreme environments are completed.
7. The portable cardiopulmonary function monitoring and risk warning method for extreme environments according to claim 6, characterized in that, The process of collecting historical cardiopulmonary data from existing workers specifically includes: Collect heart rate, pulse rate, heart rhythm, blood pressure, respiratory rate, percutaneous arterial oxygen saturation, oxygen partial pressure and carbon dioxide partial pressure in arterial blood gas analysis, and tidal volume; Record environmental comfort, predict the average thermal sensation index, and estimate the percentage of dissatisfaction; The data obtained by the basic parameter acquisition unit and the adaptive record acquisition unit are subjected to dimensionality reduction and data cleaning to remove high-frequency noise. The data after removing high-frequency noise is then subjected to discrete wavelet filtering to correct baseline drift and obtain preprocessed data.
8. The portable cardiopulmonary function monitoring and risk warning method for extreme environments according to claim 6, characterized in that, The process of calculating the interaction influence matrix based on the aforementioned environmental parameters specifically includes: A multi-dimensional sensor array is used to collect temperature, humidity, air pressure, oxygen concentration, and wind speed in extreme environments in real time. The data acquired by the environmental acquisition unit is preprocessed to obtain a complete dataset, and the sliding window method is used to perform time series analysis on the complete dataset. Based on the time series analysis results, the interaction effect matrix was calculated using the Pearson correlation coefficient.
9. The portable cardiopulmonary function monitoring and risk warning method for extreme environments according to claim 8, characterized in that, The process of obtaining the complete dataset is as follows: The 3σ principle and the isolated forest algorithm are used to remove outliers from the data acquired by the environmental acquisition unit to obtain the first processed data. The missing values in the first processed data are processed using the cubic spline interpolation method and the Kalman smoothing algorithm to obtain the second processed data. The second-processed data was formatted using Z-score, and the complete dataset was calculated using principal component analysis.
10. The portable cardiopulmonary function monitoring and risk warning method for extreme environments according to claim 6, characterized in that, The specific content of portable cardiopulmonary function monitoring and risk warning in extreme environments based on the individualized abnormal state identifiers includes: Based on the interaction matrix and the cardiopulmonary change, the dynamic changes of the worker's cardiopulmonary system under extreme conditions are calculated. A risk assessment model is constructed based on the existing cardiopulmonary history data of the workers. The dynamic changes in the cardiopulmonary function of the workers are input into the risk assessment model to obtain the risk prediction results. Based on the risk prediction results, early warning of cardiopulmonary abnormalities is issued to the current workers.