Diver safety monitoring method and system based on data fusion analysis

Abstract

The invention relates to the technical field of diving monitoring, solves the problem that the safety of a diver cannot be effectively monitored in the prior art, and provides a diver safety monitoring method and system based on data fusion analysis. The method comprises the following steps: acquiring real-time physical sign data of a diver and real-time environment data of the diver; analyzing the real-time environment data to obtain an environment influence factor used for reflecting the influence degree of the current environment on the physical sign data; according to the environmental impact factor, correcting the real-time sign data to obtain corrected target sign data; performing fusion analysis on the target sign data to obtain a real-time state analysis result of the diver; and performing comparison and safety assessment on the real-time state analysis result and a preset diver health state template corresponding to the real-time environment data to obtain a safety assessment result. According to the invention, the safety of the diver can be effectively monitored.

Description

Technical Field

[0001] The present invention relates to the technical field of diving monitoring, and in particular to a diver safety monitoring method and system based on data fusion analysis. Background Art

[0002] During underwater operations, scientific expeditions, military diving, and deep-sea rescue missions, divers often face complex and changing underwater environments, such as rapid changes in water temperature, sudden increases in water pressure, and deep-sea hypoxia. These environmental changes can easily have a significant impact on divers' physiological state, potentially inducing health risks such as abnormal heart rate, difficulty breathing, and decreased blood oxygen saturation, and even leading to life-threatening emergencies such as fainting and loss of consciousness. Therefore, to ensure the safety of divers and the successful completion of missions, it is urgent to establish a safety monitoring mechanism that can acquire and analyze diver's vital signs and environmental data in real time, and conduct dynamic assessments of divers' health status and provide safety warnings.

[0003] In existing technologies, some safety monitoring solutions only monitor divers' vital signs through single-variable monitoring, such as monitoring heart rate or hypoxemia, and triggering an alarm when these thresholds are exceeded. These methods ignore the combined impact of underwater environmental conditions (such as water temperature, water pressure, and water depth) on vital sign data and fail to consider the differences in adaptability of divers to different diving tasks. This can easily lead to false positives or missed positives, and prevents individualized health assessments. Furthermore, existing methods lack effective data fusion mechanisms and are unable to model and analyze the relationships between multiple vital sign data and complex environmental factors. This results in inaccurate safety judgments or delayed responses in diving scenarios with high task intensity or large environmental fluctuations.

[0004] Therefore, how to consider the impact of the environment on the measurement of vital sign data in a complex underwater environment, and effectively monitor the safety of divers by fusing multi-dimensional vital sign data and accurately issue safety warnings is an urgent problem to be solved. Summary of the Invention

[0005] In view of this, the present invention provides a diver safety monitoring method and system based on data fusion analysis to solve the problem that the existing technology cannot effectively monitor the safety of divers.

[0006] The technical solution adopted in the present invention is: In a first aspect, the present invention provides a diver safety monitoring method based on data fusion analysis, the method comprising: Acquire the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure, and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate, and real-time blood oxygen saturation; Analyzing the real-time environmental data to obtain an environmental impact factor reflecting the degree of impact of the current environment on the vital sign data; Correcting the real-time vital sign data according to the environmental impact factors to obtain corrected target vital sign data; Performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; The real-time status analysis result is compared with a preset diver health status template corresponding to the real-time environmental data and a safety assessment is performed to obtain a safety assessment result.

[0007] Preferably, the analysis of the real-time environmental data to obtain the environmental impact factor reflecting the degree of impact of the current environment on the vital sign data includes: Constructing an environmental template database based on the diving mission type, diving time, and individual diver characteristics. The environmental template database includes ideal environmental parameter range information corresponding to multiple diving mission types. The ideal environmental parameter range information includes a safe water temperature range, a safe water pressure range, and a safe water depth range. Comparing the real-time environmental data with the environmental template database to determine whether the real-time water temperature is within the safe water temperature range, whether the real-time water pressure is within the safe water pressure range, and whether the real-time water depth is within the safe water depth range; When the real-time water temperature is within the safe water temperature range, the real-time water pressure is within the safe water pressure range, and the real-time water depth is within the safe water depth range, a mapping model between environmental data and environmental influencing factors is established based on pre-collected historical environmental data and historical vital sign data, and the real-time environmental data is input into the mapping model to obtain the environmental influencing factors, wherein the environmental influencing factors include a first influencing factor corresponding to heart rate, a second influencing factor corresponding to respiratory rate, and a third influencing factor corresponding to blood oxygen saturation.

[0008] Preferably, the establishing of a mapping model between environmental data and environmental influencing factors based on pre-collected historical environmental data and historical vital sign data includes: Collecting historical diving data samples, the historical diving data samples including historical environmental data of historical water temperature, historical water pressure, and historical water depth, and historical vital sign data corresponding to historical heart rate, historical respiratory rate, and historical blood oxygen saturation. The conditions for collecting the historical diving data samples include different diving mission types, different diving times, and different individual characteristics of divers; Normalizing the historical vital sign data collected from the historical diving data samples to obtain normalized historical vital sign data; Statistical analysis is performed on the normalized historical vital sign data and the corresponding historical environmental data. Correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response of various types of vital sign data to various types of environmental data, thereby forming a sensitivity matrix between the environment and vital signs. Each element of the matrix represents the sensitivity weight value of each type of vital sign data to each type of environmental data. Based on the sensitivity matrix, a multi-channel modeling framework is constructed, and mapping sub-models are constructed for each type of vital sign data. The mapping sub-model takes water temperature, water pressure, and water depth as input, and takes the environmental impact factors corresponding to the various types of vital sign data as output. The sub-model includes a weighted regression model, a neural network model, or an integrated learning model.

[0009] Preferably, the normalized historical vital sign data and the corresponding historical environmental data are statistically analyzed, and correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response degree of each type of vital sign data to each type of environmental data to form a sensitivity matrix between the environment and the vital signs, including: For each data sample in the normalized historical vital sign data, according to its collection time point, the water temperature value, water pressure value and water depth value at the corresponding time point are extracted from the normalized historical environmental data to form a vital sign environment sample pair; The physical sign and environmental sample pairs are classified according to the physical sign data type, and paired sample sets between various physical sign data and various environmental data are extracted respectively; Calculating response strength indices for each of the paired sample sets, wherein the response strength indices include Pearson correlation coefficients, covariance values, and mutual information values, and are used to characterize linear or nonlinear correlations between physical signs and environmental parameters; Performing weighted fusion on the response intensity indexes to obtain response sensitivity scores of various types of vital sign data to various types of environmental data; Arrange and combine the response sensitivity scores to construct an initial sensitivity matrix between physical signs and environments, wherein each row of the initial sensitivity matrix corresponds to a type of physical sign data, each column corresponds to a type of environmental data, and each matrix element is a sensitivity score of a corresponding physical sign to a corresponding environment; Linear normalization is performed on the elements in the initial sensitivity matrix, and the sensitivity scores are converted into standard weight value intervals to form a normalized sensitivity matrix between the environment and physical signs.

[0010] Preferably, the real-time vital sign data is corrected according to the environmental impact factor to obtain the corrected target vital sign data, which includes: Correcting the real-time heart rate according to the first influencing factor to obtain a target heart rate; Correcting the real-time respiratory rate according to the second influencing factor to obtain a target respiratory rate; Correcting the real-time blood oxygen saturation according to the third influencing factor to obtain a target blood oxygen saturation; The target vital sign data is determined according to the target heart rate, target respiratory rate and target blood oxygen saturation.

[0011] Preferably, the fusion analysis of the target vital sign data to obtain the diver's real-time status analysis results includes: Obtaining the current diving mission type information, and based on the type information, calling corresponding fusion configuration information from a preset fusion analysis parameter library, the fusion configuration information including the conditional probability distribution information of each vital sign data under each diver's state level, and the prior probability information of the state level; Extract target heart rate, target respiratory rate and target blood oxygen saturation from the corrected target vital sign data as observation input data for Bayesian inference; Calculating the likelihood probabilities of target heart rate, target respiratory rate, and target blood oxygen saturation at each diver's state level based on the conditional probability distribution information defined in the fusion configuration information; Multiplying the likelihood probability of the target physical sign data at each diver status level by the prior probability of the corresponding diver status level to obtain the unnormalized posterior probability corresponding to each diver status level; Normalizing the unnormalized posterior probability to obtain a normalized posterior probability distribution of each diver's status level; The diver status level corresponding to the maximum probability in the posterior probability distribution is selected as the real-time health status level of the current diver, and the level and its probability confidence are output to form a real-time status analysis result.

[0012] Preferably, calling corresponding fusion configuration information from a preset fusion analysis parameter library according to the type information includes: Acquire a historical diving mission dataset, the dataset including multiple diving mission types, environmental condition information of each diving mission, individual characteristic data of the diver, and actual status level label of the diver; Classifying the historical data set according to diving mission types to obtain multiple mission type subsets, each mission type subset corresponding to historical data of a type of diving mission; The vital sign data in each task type subset are classified and counted according to the corresponding state level labels, and the statistical characteristics of each type of vital sign data under various state levels are calculated, including expected value, variance, and distribution form, which are used to construct a conditional probability distribution model; Based on the statistical characteristics, a parameter fitting method is used to establish a conditional probability distribution model for each type of vital sign data at different state levels, wherein the conditional probability distribution model includes probability distribution functions of target heart rate, target respiratory rate, and target blood oxygen saturation at each state level; Count the historical occurrence frequencies of different status levels under each task type, calculate the prior probability of each status level based on the frequency distribution, and form a prior probability set of the status level; The conditional probability distribution model of the physical signs corresponding to each diving task type and the prior probability set of the state level are used as fusion configuration information, indexed by task type, and stored in a fusion analysis parameter library to complete the construction of the fusion analysis parameter library; The current diving task type information is obtained, and the type information is used as a search keyword to call the fusion configuration information corresponding to the current task type from the fusion analysis parameter library.

[0013] Preferably, the comparing the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data and performing a safety assessment to obtain a safety assessment result includes: Determine index conditions for a diver's health status template based on current diving mission type information and real-time environmental data, wherein the health status template includes status level threshold settings and response measures under different environmental conditions; Calling a target health status template that matches the index condition from a preset health status template library, wherein the target health status template records the safety recommendations and risk warning strategies corresponding to each status level; According to the real-time status analysis result, obtaining the real-time health status level and its corresponding probability confidence output based on Bayesian inference; Comparing the real-time health status level with the status level definition in the target health status template to determine whether the status level belongs to the warning level or danger level category; When the health status level is a warning level or a danger level, generating a safety intervention suggestion for the current diver according to the response strategy corresponding to the status level in the target health status template, the safety intervention suggestion including a buoyancy suggestion, enhanced monitoring, or mission termination; The health status level, probability confidence value and generated safety intervention suggestion are output as the safety assessment result of the current diver.

[0014] In a second aspect, the present invention provides a diver safety monitoring system based on data fusion analysis, the system comprising: A data acquisition module is used to obtain the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate and real-time blood oxygen saturation; An environmental data analysis module is used to analyze the real-time environmental data to obtain an environmental impact factor that reflects the degree of impact of the current environment on the vital sign data; A vital sign data correction module, configured to correct the real-time vital sign data according to the environmental influencing factors to obtain corrected target vital sign data; A fusion analysis module is used to perform fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; A safety assessment module is used to compare the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data and perform a safety assessment to obtain a safety assessment result; At least one processor, at least one memory, and computer program instructions stored in the memory, when the computer program instructions are executed by the processor, implement the method according to the first aspect.

[0015] In summary, the beneficial effects of the present invention are as follows: The present invention provides a diver safety monitoring method and system based on data fusion analysis, the method comprising: obtaining real-time vital sign data and real-time environmental data of the diver, the real-time environmental data comprising real-time water temperature, real-time water pressure and real-time water depth, the real-time vital sign data comprising real-time heart rate, real-time respiratory rate and real-time blood oxygen saturation; analyzing the real-time environmental data to obtain an environmental impact factor reflecting the degree of impact of the current environment on the vital sign data; correcting the real-time vital sign data according to the environmental impact factor to obtain corrected target vital sign data; performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; comparing and safety evaluating the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data to obtain a safety assessment result. To address the challenges posed by complex underwater environments to diver vital sign measurement and safety monitoring, the present invention proposes a safety monitoring method based on data fusion analysis, focusing on addressing the problems of vital sign measurement errors caused by environmental interference and the failure of safety monitoring using a single indicator. First, the diver's vital sign data (such as heart rate, respiratory rate, and blood oxygen saturation) and environmental data (such as water temperature, water pressure, and water depth) are collected in real time. By constructing an environmental template database and establishing a sensitivity matrix that reflects the relationship between the environment and vital sign responses, the degree of environmental impact on the vital signs is analyzed and corresponding environmental impact factors are generated. Next, by establishing a multi-channel mapping model, these environmental impact factors are used to correct the original vital sign data to obtain target vital sign data that is more consistent with the actual physiological state. Subsequently, using the Bayesian inference method, combined with the current diving mission type, the vital sign data of various dimensions are fused and analyzed to generate the diver's real-time status level. Finally, the analysis results are compared with the preset health status template, and a safety assessment result is generated based on the status level and confidence level. Specific risk intervention recommendations (such as ascent or mission termination) are output when necessary. Overall, through the technical path of environmental perception modeling + physical sign correction + probabilistic fusion reasoning + template comparison, accurate identification of the diver's safety status and safety warning are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work, and these are all within the scope of protection of the present invention.

[0017] Figure 1 Schematic diagram of the overall working process of the diver safety monitoring method based on data fusion analysis in Example 1 of the present invention; Figure 2 This is a flow chart of performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of a diver in Example 1 of the present invention; Figure 3 This is a structural block diagram of a diver safety monitoring system based on data fusion analysis in Example 2 of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the orientation or position relationship indicated by the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like is based on the orientation or position relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further limitations, elements defined by the phrase "comprising..." do not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising the elements. The embodiments of the present invention and the features thereof may be combined with each other if there is no conflict, and all are within the scope of protection of the present invention.

[0019] Example 1 See Figure 1 Embodiment 1 of the present invention discloses a diver safety monitoring method based on data fusion analysis, the method comprising: Acquire the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure, and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate, and real-time blood oxygen saturation; Specifically, real-time vital sign data refers to a diver's vital signs collected in real time via wearable or form-fitting sensors. These include heart rate (heart beats per unit time, typically expressed in bpm), respiratory rate (respirations per unit time, reflecting respiratory system activity), and blood oxygen saturation (SpO2, representing the proportion of oxygen carried by hemoglobin in the blood, typically expressed as a percentage, with a normal value generally above 95%). These data reflect the diver's current vital activity and are key indicators for assessing their physical condition. Real-time environmental data includes water temperature, water pressure, and water depth. Water temperature refers to the water temperature at the current diving depth; water pressure indicates the pressure exerted by the water on the human body, which increases with depth; and depth refers to the diver's position relative to the surface. These environmental parameters have a direct impact on human physiological status. For example, low water temperature may lead to a decrease in heart rate and slower breathing, while high water pressure may cause changes in gas exchange efficiency. The primary purpose of this step is to build the foundational data for subsequent multimodal fusion analysis. By simultaneously collecting physiological and environmental variables, unified modeling of human-environment status is achieved. Due to the extremely complex diving environment, the presentation and meaning of vital sign data may vary under varying depths, temperatures, and pressures. Relying solely on uncorrected vital sign data makes it difficult to accurately assess health status. Therefore, environmental information must be incorporated as a prerequisite for subsequent correction, modeling, analysis, and safety alerts. During implementation, smart wearable devices integrating multiple sensor types are employed, such as diving wristbands or wetsuit-embedded modules that integrate a heart rate monitor, pulse oximeter, and respiratory sensor. These devices adhere to the diver's skin and measure heart rate and blood oxygen saturation using photoplethysmography (PPG), and respiratory rate using chest strap inductance or volume change methods. Real-time environmental information is acquired through depth gauges, water temperature sensors, and pressure sensors, typically integrated into diving helmets, dials, or backpacks. All collected parameters undergo preliminary filtering (e.g., sliding averages, wavelet denoising) to remove signal noise before being transmitted in real time to a host computer or diving safety monitoring platform via low-power wireless communication modules (e.g., Bluetooth or ultrasound). By synchronously collecting real-time environmental data and real-time vital sign data, dynamic vital sign modeling under environmental perception can be achieved, providing a data basis for subsequent corrections and judgments.

[0020] Analyzing the real-time environmental data to obtain an environmental impact factor reflecting the degree of impact of the current environment on the vital sign data; Specifically, by analyzing the collected real-time environmental data, environmental impact factors are constructed to quantify the degree to which current environmental conditions affect the diver's real-time vital sign data. Specifically, different environmental variables can interfere with normal vital sign readings to varying degrees. For example, low water temperature may cause a decrease in heart rate and respiratory rate, and high water pressure may affect blood gas exchange efficiency. To quantitatively describe these potential interferences, based on historical diving data samples, statistical methods (such as correlation analysis and mutual information analysis) are used to calculate the sensitivity of environmental variables to vital sign indicators. A mapping model is then constructed to convert real-time environmental inputs into corresponding impact factors. This analysis provides a basis for subsequent environmental adaptability correction of vital sign data, thereby effectively avoiding misjudgments due to environmental anomalies and improving the accuracy and reliability of diver health status assessments.

[0021] Preferably, the analysis of the real-time environmental data to obtain the environmental impact factor reflecting the degree of impact of the current environment on the vital sign data includes: Constructing an environmental template database based on the diving mission type, diving time, and individual diver characteristics. The environmental template database includes ideal environmental parameter range information corresponding to multiple diving mission types. The ideal environmental parameter range information includes a safe water temperature range, a safe water pressure range, and a safe water depth range. Specifically, the environmental template database is a knowledge base used to store ideal environmental conditions for diving missions. Indexed by mission type (e.g., rescue, scientific research, training), mission duration (e.g., short, long), and individual diver characteristics (e.g., age, fitness level, and physical adaptability), this database records the ideal environmental parameter ranges for different mission types, including safe water temperature, safe water pressure, and safe water depth. The primary purpose of this step is to provide a reference for determining whether the current environment is within an acceptable range, thereby avoiding misjudgments of a diver's physical condition in abnormal circumstances. Because the tolerance environments corresponding to different diving missions and individual diver characteristics vary significantly, a unified standard cannot be used for judgment. Therefore, a refined conditional template database must be established to enable dynamic adaptability assessment. To implement this, historical mission archives and actual dive records are first collected to calculate the environmental ranges corresponding to successful dives for different mission types. In combination with physiological adaptability research, the extreme tolerances of different divers are modeled and safe parameter ranges for different mission types are defined. Finally, this template information is uniformly and structured and stored in the environmental template database to support subsequent real-time comparison and indexing. By combining task characteristics with individual differences, graded and typed environmental templates are formed, which not only improves the personalization and accuracy of system judgments, but also provides a standard reference for pre-dive task planning and dynamic monitoring, and enhances the system's adaptability to sudden environmental changes.

[0022] Comparing the real-time environmental data with the environmental template database to determine whether the real-time water temperature is within the safe water temperature range, whether the real-time water pressure is within the safe water pressure range, and whether the real-time water depth is within the safe water depth range; Specifically, the real-time water temperature, water pressure, and water depth values ​​collected by the sensor are matched item by item with the ideal parameter ranges defined in the environmental template database to determine whether they are currently within the safety boundary. By quickly identifying whether there are potential risks in the current diving environment, it serves as a prerequisite for whether to continue the status analysis. If the environment is obviously beyond the safety range, any correction or analysis of the vital sign data may lose its meaning, so a judgment must be made here to improve the reliability of subsequent analysis. In the specific implementation process, the corresponding safe interval template is called from the database according to the diving mission type and the individual information of the diver, and then the real-time environmental data is compared with it by threshold. The comparison process generally uses a floating-point range judgment algorithm to ensure compatibility with small measurement errors. If any environmental data exceeds the set range, the environmental anomaly mark can be triggered immediately, skipping the vital sign correction link and directly entering the risk warning process.

[0023] When the real-time water temperature is within the safe water temperature range, the real-time water pressure is within the safe water pressure range, and the real-time water depth is within the safe water depth range, a mapping model between environmental data and environmental influencing factors is established based on pre-collected historical environmental data and historical vital sign data, and the real-time environmental data is input into the mapping model to obtain the environmental influencing factors, wherein the environmental influencing factors include a first influencing factor corresponding to heart rate, a second influencing factor corresponding to respiratory rate, and a third influencing factor corresponding to blood oxygen saturation.

[0024] Specifically, a mapping model is a mathematical or machine learning model that maps environmental parameters (such as water temperature, water pressure, and water depth) to their impact on vital sign data. These models, such as multivariate linear regression models, feedforward neural network models, or ensemble regression tree models, are used to map environmental parameters (such as water temperature, water pressure, and water depth) to their influence on vital sign data. Environmental impact factors are intermediate quantities output by these models, representing the degree to which a particular environmental combination interferes with heart rate, respiratory rate, and blood oxygen saturation. By quantifying the impact of complex environmental variables, this provides a scientific basis for subsequent correction of vital sign data to remove environmental interference. This prevents misinterpretation of normal physiological environmental feedback as health abnormalities, contributing to more accurate individual status assessments. In implementation, statistical modeling and multi-channel learning techniques are used based on historical diving data samples to establish independent prediction sub-models for each vital sign. The model inputs are historical water temperature, historical water pressure, and historical water depth, and the output is an estimated value of the interference with the specific vital sign. During real-time monitoring, newly acquired environmental data is fed into the corresponding sub-model, which outputs three environmental impact factors: one for heart rate, one for respiratory rate, and one for blood oxygen saturation. By enabling quantitative modeling and rapid inference of environmental effects, the ability to interpret the sources of abnormal vital sign data is significantly improved, reducing false positives and laying the foundation for personalized health monitoring. Furthermore, it supports the model's adaptive learning capabilities, continuously optimizing assessment accuracy as new data accumulates, a key step in ensuring the long-term applicability of monitoring systems.

[0025] Preferably, the establishing of a mapping model between environmental data and environmental influencing factors based on pre-collected historical environmental data and historical vital sign data includes: Collecting historical diving data samples, the historical diving data samples including historical environmental data of historical water temperature, historical water pressure, and historical water depth, and historical vital sign data corresponding to historical heart rate, historical respiratory rate, and historical blood oxygen saturation. The conditions for collecting the historical diving data samples include different diving mission types, different diving times, and different individual characteristics of divers; Specifically, historical diving data samples refer to complete data records collected during past diving missions. These include environmental data (such as water temperature, water pressure, and depth) and corresponding vital signs (such as heart rate, respiratory rate, and blood oxygen saturation) at different points during the dive. These data samples cover a variety of diving scenarios, such as shallow-water photography missions, deep-sea scientific exploration missions, and high-energy training missions. They use dive times (e.g., 10 minutes, 30 minutes, and 1 hour) and the physical conditions of individual divers (e.g., young amateur divers versus older professional divers) as variables. This categorized collection maximizes the diversity of human physiological responses to real-world diving environments. By building a representative and comprehensive historical data base, this serves as the core foundation for subsequent modeling. Data quality and diversity directly impact the accuracy and universality of environmental influencing factors. Therefore, comprehensive specifications for classification dimensions, coverage boundaries, and recording frequency must be established at the source of data collection. This step also lays the foundation for consistent data dimensions for subsequent normalization and sensitivity modeling. During implementation, historical data is collected through the following channels: historical mission logs automatically recorded by existing diving equipment (such as integrated diving watches and smart diving suits); diving test data accumulated by scientific research institutions or the military; and manually entered mission documents and manual monitoring results. Before data collection, a unified data format should be established, for example, using a recording time step of one minute to ensure a one-to-one correspondence between environmental and vital sign data. Furthermore, data cleaning rules should be established (such as removing missing fields or abnormal jumps) to ultimately form a structured and labeled training sample set. The key technical advantage of this step lies in establishing a multi-dimensional, high-quality sample data source, which helps the model fully learn the true relationship between the environment and vital signs under different conditions during the training phase, avoiding the loss of generalization ability caused by insufficient or biased samples. During the subsequent system deployment phase, continuous updates to this database can also achieve long-term adaptive evolution of the model.

[0026] Normalizing the historical vital sign data collected from the historical diving data samples to obtain normalized historical vital sign data; Specifically, normalization involves mapping vital sign data of different dimensions (e.g., heart rate in beats per minute, oxygen saturation in percentage) to the same numerical range using standardized mathematical transformation methods (e.g., maximum-minimum normalization, Z-score standardization). For example, heart rate values ​​may originally range from 60–180 beats per minute. Normalization converts these values ​​to values ​​between 0 and 1 to ensure consistency in subsequent calculations. This step aims to eliminate computational bias caused by scale differences between vital sign data, ensuring that subsequent statistical analysis or model inputs do not overwhelm the overall trend due to the excessively large range of a single metric. Normalization also helps improve the convergence speed and stability of model training, and is an essential preprocessing step, particularly in multivariate analysis. Normalization typically involves two steps: the first is the statistical analysis phase, in which the system scans the maximum, minimum, mean, and standard deviation of each vital sign dimension across all historical samples; the second is the numerical transformation phase, in which all vital sign values ​​are transformed one by one. Normalization should be consistent, meaning that subsequent models should use the same normalization parameters for real-time data preprocessing. This improves the comparability and quality of vital sign data and prevents interference between different dimensions from misleading analytical models. This not only supports data cleaning during the modeling phase but also provides a unified data structure for subsequent sensitivity matrix construction, helping to develop more accurate environmental-vital sign response relationships.

[0027] Statistical analysis is performed on the normalized historical vital sign data and the corresponding historical environmental data. Correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response of various types of vital sign data to various types of environmental data, thereby forming a sensitivity matrix between the environment and vital signs. Each element of the matrix represents the sensitivity weight value of each type of vital sign data to each type of environmental data. Specifically, the sensitivity matrix is ​​a two-dimensional matrix structure whose rows represent different vital sign data (such as heart rate, respiratory rate, and blood oxygen saturation), and whose columns represent different environmental data (such as water temperature, water pressure, and water depth). Each element in the matrix represents the weight of the vital sign's response to that environmental variable (i.e., sensitivity). For example, a matrix element value of 0.72 indicates a high degree of heart rate responsiveness to water pressure, with significant changes. The goal of this step is to extract the patterns of influence of environmental variables on vital sign indicators, providing a mathematical basis for the subsequent construction of an environmental influencing factor model. Because the impact of environmental changes on human physiological state is nonlinear and varies greatly among individuals, it is necessary to combine multiple analytical methods to comprehensively assess the strength of the association to establish a more robust explanatory model. In this implementation, correlation coefficient analysis is first used to measure the strength of linear relationships, which is suitable for preliminary analysis. Second, covariance analysis is used to identify whether environmental changes lead to consistent trends in the direction of vital sign fluctuations. Finally, mutual information analysis is introduced to assess nonlinear correlations, which is particularly suitable for identifying data pairs with unclear implicit relationships but statistical dependence. Ultimately, the results of these three methods are integrated to calculate a weighted sensitivity index, forming a complete sensitivity matrix. This step, through multi-dimensional statistical extraction technology, not only improves the accuracy of modeling the environmental-physical-sign relationship but also implements parameter sensitivity ranking and variable priority management, helping the system dynamically adjust parameter collection priorities based on mission requirements during subsequent deployments. Furthermore, the sensitivity matrix can be used in risk assessment systems to provide early warnings for potentially highly sensitive individuals and high-risk environments, enhancing the system's preventative response capabilities.

[0028] Preferably, the normalized historical vital sign data and the corresponding historical environmental data are statistically analyzed, and correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response degree of each type of vital sign data to each type of environmental data to form a sensitivity matrix between the environment and the vital signs, including: For each data sample in the normalized historical vital sign data, according to its collection time point, the water temperature value, water pressure value and water depth value at the corresponding time point are extracted from the normalized historical environmental data to form a vital sign environment sample pair; Specifically, a vital sign-environment sample pair refers to a combination of normalized vital sign data and environmental data corresponding to the same time point. For example, in the 15th minute of a historical dive record, the recorded normalized heart rate value is 0.64, water temperature is 0.72, water pressure is 0.81, and water depth is 0.65. These data constitute a complete vital sign-environment sample pair. Time point matching is crucial to ensure a one-to-one correspondence, ensuring a clear input structure for subsequent related analyses. Time synchronization accurately matches vital sign data with the environmental influences at the time, providing a data foundation for calculating the degree of environmental influence in subsequent analyses. If vital sign data and environmental data are out of sync, analysis will be distorted. Therefore, time alignment is a prerequisite for establishing a reasonable causal inference chain. In implementation, all normalized vital sign samples are traversed, and based on their timestamps (e.g., minute t, minute t+1), the water temperature, water pressure, and water depth values ​​at the same time point are searched and extracted from the normalized environmental data. This process can be accelerated through dual pointers or indexed hashing. When environmental data at a particular point in time is missing, it should be supplemented through methods such as forward filling and linear interpolation to avoid omissions. Ultimately, each vital sign sample is accompanied by a set of environmental data values, forming a standardized analysis pair. This step constructs a set of sample pairs with unified data dimensions and consistent temporal semantics, greatly improving the consistency and effectiveness of data input. It also provides a clearly structured basic unit for subsequent dimensional analysis, response modeling, and matrix calculations, facilitating automated processing and visualization.

[0029] The physical sign and environmental sample pairs are classified according to the physical sign data type, and paired sample sets between various physical sign data and various environmental data are extracted respectively; Specifically, classification by vital sign data type in this step involves splitting sample pairs according to their vital sign variables (such as heart rate, respiratory rate, and blood oxygen saturation). Each vital sign data type is then paired with each environmental variable (such as water temperature, water pressure, and water depth) to form a separate sample set. For example, the system extracts all data pairs related to heart rate and water pressure to form one subset, respiratory rate and water temperature to form another subset, and so on, resulting in a total of 3 × 3 = 9 paired sample sets. The purpose of this step is to establish a dedicated statistical sample pool for each vital sign and environmental variable combination, facilitating subsequent independent analysis of their correlation and sensitivity. This classification process avoids cross-interference between different vital signs, making the analysis results for each category more targeted and explanatory. In specific implementation, the vital sign and environment sample pairs are iterated over, and each record is extracted and classified by field. The nine sample sets, such as "heart rate and water temperature," "heart rate and water pressure," and "heart rate and water depth," are stored in corresponding cache sets. The data structure can use a nested dictionary (e.g., dict vital sign type environment type) for efficient storage and access. This classification process not only supports parallel computing but also supports subsequent independent visualization operations. The resulting paired sample sets significantly improve the clarity and precision of statistical analysis. The system can develop specific strategies for each specific pair, such as using a nonlinear model for blood oxygen and water depth, or a linear fit for heart rate and water temperature. The advantage of classification strategies lies in the flexibility they provide for modeling.

[0030] Calculating response strength indices for each of the paired sample sets, wherein the response strength indices include Pearson correlation coefficients, covariance values, and mutual information values, and are used to characterize linear or nonlinear correlations between physical signs and environmental parameters; Specifically, response strength indices are statistical metrics that quantify whether a particular physical sign variable is affected by a particular environmental variable. These include: the Pearson correlation coefficient, which measures the strength of linear correlation and ranges from -1 to 1; the covariance value, which reflects the directional consistency of two variables; larger values ​​indicate stronger consistency; and the mutual information value, which measures the degree of nonlinear information sharing between two variables; higher values ​​indicate stronger correlation, even when there is no linear relationship. This comprehensive assessment of the physical sign's response to the environment through multiple statistical dimensions avoids bias caused by relying solely on a single indicator. Parallel linear and nonlinear analysis can help the system identify different types of influencing mechanisms, providing a high-dimensional data foundation for constructing sensitivity scores and final modeling. During implementation, each physical sign-environment sample set is fed into three separate analysis models, resulting in the calculation of three sets of indices. Numerical libraries such as NumPy and SciPy can be used for efficient implementation. Note that when the sample size is small, the bias-corrected Pearson formula should be used; and when calculating mutual information, discretization or kernel density estimation should be used to improve accuracy. The results for each set of indices are recorded and used in the next step of weighted fusion processing. This step covers the entire correlation structure, from explicit linearity to implicit nonlinearity, ensuring that both obvious correlations and subtle implicit relationships are captured. The results can be used for model selection (such as linear regression or tree models) and also contribute to model interpretation enhancement and risk control analysis.

[0031] Performing weighted fusion on the response intensity indexes to obtain response sensitivity scores of various types of vital sign data to various types of environmental data; Specifically, weighted fusion combines the three acquired response strength indicators (Pearson coefficient, covariance, and mutual information) according to preset or trained weighting factors, ultimately yielding a composite score, called the response sensitivity score. For example, for the analysis of heart rate and water pressure, the three indicators are 0.6, 0.4, and 0.7, respectively, and the fusion weights are 0.3, 0.2, and 0.5, resulting in a fusion score of 0.3 × 0.6 + 0.2 × 0.4 + 0.5 × 0.7. Integrating multi-dimensional statistical indicators into a unified, comparable value gives the response strength of physical signs and the environment a unified dimension, facilitating sorting and matrix construction. Weighting strategies can also reflect the priority of different analysis dimensions in real-world scenarios. For example, in strongly nonlinear environments, mutual information can be given a higher weight. During implementation, static weighting (i.e., manually set empirical weighting) or dynamic weighting (automatically learned weighting through data-driven methods such as principal component analysis and information gain) can be used. A static weighting strategy is generally adopted in the initial stages of system deployment. Once sufficient training data is available, a dynamic fusion strategy can be employed to improve generalization. The fusion process should retain all original indicators to facilitate subsequent anomaly analysis and tracing. This step integrates statistical indicators across complex dimensions, improving decision-making efficiency and comparative clarity. In particular, when constructing a matrix, a unified score can effectively avoid the trade-offs between multiple indicators, improving model stability and practicality.

[0032] Arrange and combine the response sensitivity scores to construct an initial sensitivity matrix between physical signs and environments, wherein each row of the initial sensitivity matrix corresponds to a type of physical sign data, each column corresponds to a type of environmental data, and each matrix element is a sensitivity score of a corresponding physical sign to a corresponding environment; Specifically, the initial sensitivity matrix is ​​a two-dimensional matrix with dimensions equal to the number of physical sign variables × the number of environmental variables. Each element represents a response sensitivity score. This matrix format not only clearly reflects the relationships between variables but also serves as the input for subsequent modeling or feature selection. By systematically and structuredly organizing the scattered sensitivity scores, a comprehensive physical sign-environment response map is constructed. This not only supports overall system evaluation but also provides a basis for anomaly detection and alerting for individual pathways. During implementation, the system can predefine a list of physical sign variables and a list of environmental variables, initialize an empty matrix (e.g., 3×3), and sequentially populate it with the fusion scores from the previous stage. The matrix can be stored in a two-dimensional array or Pandas DataFrame, facilitating subsequent normalization, sorting, and visualization. This step establishes an efficient indexing structure and a unified data view, providing structured input for AI models and facilitating the automatic generation and optimization of the model feature matrix. It also provides an intuitive data foundation for expert visual analysis, parameter tuning, and early warning strategy development.

[0033] Linear normalization is performed on the elements in the initial sensitivity matrix, and the sensitivity scores are converted into standard weight value intervals to form a normalized sensitivity matrix between the environment and physical signs.

[0034] Specifically, linear normalization involves mapping all values ​​in the sensitivity matrix to a standard range, typically 0 and 1, to enhance the comparability of the responses of different signs to different environmental parameters. This step aims to eliminate the impact of varying numerical scales on subsequent modeling or weighted analysis, ensuring that all sensitivity scores fall within a uniform range. This facilitates subsequent modeling weight assignment, priority sorting, and visualization (e.g., heatmaps). During implementation, the system first iterates over all elements in the matrix, recording the maximum and minimum values. A linear normalization formula is then applied to each element. Note that in the special case where the maximum and minimum values ​​are equal, to avoid division by zero errors, the values ​​can be uniformly set to 0.5 or the dimension can be skipped. It is recommended to cache intermediate states during the normalization process, retaining the original matrix for comparative analysis. The normalized sensitivity matrix not only improves computational stability but also facilitates integration into multi-source model systems, serving as input features or adjustment factors.

[0035] Based on the sensitivity matrix, a multi-channel modeling framework is constructed, and mapping sub-models are constructed for each type of vital sign data. The mapping sub-model takes water temperature, water pressure, and water depth as input, and takes the environmental impact factors corresponding to the various types of vital sign data as output. The sub-model includes a weighted regression model, a neural network model, or an integrated learning model.

[0036] Specifically, the multi-channel modeling framework separates vital sign data such as heart rate, respiratory rate, and blood oxygen saturation into multiple independent modeling channels. The mapping sub-model is a mathematical tool that converts the three environmental variables of water temperature, water pressure, and water depth into the magnitude of the interference with the corresponding vital signs. For example, the "heart rate channel" sub-model specifically outputs the interference value for heart rate. This design aims to leverage the prior information in the sensitivity matrix about which environmental variables are most sensitive to which vital signs to estimate the specific interference value in real time and with high precision. During implementation, historical vital signs are first matched one-to-one with environmental data using timestamps. Outliers are then removed through sliding average and wavelet threshold denoising. The most effective environmental inputs are screened based on the sensitivity matrix and scaled between zero and one. An appropriate model is then selected for each channel. If a linear relationship prevails, weighted least squares regression is used and the sensitivity weights are embedded in the coefficient initialization. If nonlinearity is present, a three-layer feedforward neural network is used and a weight penalty is added to the loss function. If there are few samples and complex variable coupling, a random forest or gradient boosting tree is used and a weight penalty is introduced in the split gain. Each channel is optimized using five-fold cross-validation and then solidified into an inference service. This channel-by-channel, weight-guided, and regularized modeling process not only ensures the accuracy and interpretability of the model, but also facilitates independent updating and maintenance by channel later, thereby improving the adaptability and reliability of the entire monitoring system in various diving environments.

[0037] Correcting the real-time vital sign data according to the environmental impact factors to obtain corrected target vital sign data; Specifically, the environmental impact factor is the interference output by the sub-model in the previous step, which is used to quantify the specific impact of the current water temperature, water pressure, and water depth on each vital sign reading. The target vital sign data is the original reading after removing these interferences to obtain a more realistic value. Its core purpose is to distinguish "environmental adaptation changes" from "physiological abnormalities" to avoid misinterpreting a decreased heart rate caused by low temperature or a deviation in blood oxygen readings caused by high pressure as a health issue. The implementation process is as follows: The monitoring platform receives raw vital sign and environmental data from wearable devices every second. The environmental variables are first processed using the same normalized parameters used during the training phase, and then fed into the corresponding channel model to calculate heart rate, respiration, and blood oxygen disturbances. At the data level, the raw vital sign signals are smoothed using a second-order unbiased Kalman filter. Compensation is then applied using either addition or multiplication depending on the vital sign type. For example, the interference value is directly subtracted from the heart rate and respiration values, while blood oxygen is converted back to a percentage. The corrected values ​​are then compared with a library of individual thresholds. Exceeding these limits triggers a voice or vibration alarm. The four-tuple (raw value, environmental variable, interference value, and corrected value) is written to a local cache and synchronized to the mothership server via underwater communication. If the interference value remains high for dozens of consecutive sampling cycles, the system automatically lowers the threshold and prompts the diver to surface. This link requires only lightweight mathematical operations and small model inference, with a total latency of less than 50 milliseconds. This significantly reduces false alarms and improves the detection rate of true anomalies, providing fast, accurate, and traceable health assurance for diving operations.

[0038] Preferably, the real-time vital sign data is corrected according to the environmental impact factor to obtain the corrected target vital sign data, which includes: Correcting the real-time heart rate according to the first influencing factor to obtain a target heart rate; Specifically, the first environmental impact factor is an estimate of the disturbance caused by the current environmental conditions on the diver's heart rate, such as the natural downward trend in heart rate caused by decreasing water temperature or increasing water pressure. This step aims to remove these non-pathological changes caused by the external physical environment from the raw heart rate reading, resulting in a target heart rate value that more closely reflects the diver's true physiological state. In its implementation, the system first filters the raw heart rate signal to reduce short-term jitter. It then inputs the current water temperature, water pressure, and water depth into the heart rate mapping model to obtain the first environmental impact factor. The system then subtracts this disturbance from the raw heart rate, or applies a proportional correction based on the model's recommendations, to generate the target heart rate. To prevent abnormal deviations, the corrected heart rate is compared with the individual's historical average heart rate to ensure it remains within an acceptable range. This processing method effectively avoids false positives caused by environmentally adaptive heart rate changes, helping to more accurately determine whether a true health risk exists.

[0039] Correcting the real-time respiratory rate according to the second influencing factor to obtain a target respiratory rate; Specifically, the second environmental impact factor reflects the degree of interference with respiratory rate under current environmental conditions. For example, high-pressure environments at depth can cause respiratory rate to naturally accelerate or decelerate. This correction step is intended to eliminate rhythmic variations caused by environmental factors, ensuring that the system's assessment of an excessively high or low respiratory rate is not influenced by these factors, thereby more accurately reflecting the diver's pulmonary and neurological status. During implementation, the platform captures the raw respiratory rate using a chest strap or a form-fitting sensor. The real-time environmental data is then fed into a mapping model corresponding to the respiratory rate to calculate the second environmental impact factor. Correction can be performed by subtracting the factor from the raw rate or adjusting it by a proportional coefficient. The corrected result becomes the target respiratory rate. This process also includes outlier removal, range limiting, and data writeback to ensure stable and reliable results. By introducing a respiratory rate correction mechanism, the system can still make reasonable judgments under varying water depths, temperatures, or gas concentrations, improving the stability of health status recognition.

[0040] Correcting the real-time blood oxygen saturation according to the third influencing factor to obtain a target blood oxygen saturation; Specifically, the third environmental impact factor is an estimate reflecting the impact of current environmental conditions on blood oxygen saturation measurements. This may result from changes in water temperature or pressure fluctuations that interfere with oxygen exchange efficiency or photoelectric measurement accuracy. The core purpose of this correction is to eliminate false dips or increases in blood oxygen readings caused by environmental factors, thereby more accurately determining whether a diver is at risk of hypoxia. In practice, the system first performs wavelet denoising and averaging filtering on the raw blood oxygen signal. The real-time environmental conditions are then input into the blood oxygen mapping model to obtain the third environmental impact factor. Since blood oxygen data is typically expressed as a percentage, the correction process often employs a multiplication compensation method: the raw percentage value is multiplied by a correction factor and then normalized to obtain the final target blood oxygen saturation value. This result is then cross-validated against the individual's historical range to ensure that there are no false positives or negatives due to model errors. This correction strategy effectively improves the interference resistance of blood oxygen determination, significantly reducing sources of error, particularly in complex underwater lighting and pressure environments.

[0041] The target vital sign data is determined according to the target heart rate, target respiratory rate and target blood oxygen saturation.

[0042] Specifically, target vital sign data refers to a set of core physiological status indicators formed after the three key vital signs are corrected for environmental interference. Together, they form the basic basis for judging the diver's current true health status. The main purpose of setting up this step is to unify the three independently corrected data of heart rate, respiratory rate and blood oxygen to form a comprehensive data package that can be directly used for health assessment and risk warning. During the implementation process, the system encapsulates the target heart rate, target respiratory rate and target blood oxygen saturation in a unified format, and records the related metadata such as timestamp, diving mission number, diver identity, etc. to form a complete data structure. Subsequently, the target vital sign data will be used in the state analysis model, abnormal judgment logic and visualization platform display. This unified processing not only facilitates subsequent algorithm processing, but also supports medical staff to track and analyze the diver's status at every moment and make remote intervention decisions, further improving the system's real-time response capability and overall health management level.

[0043] Performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; Specifically, the fusion analysis of target vital sign data refers to integrating the three core indicators of corrected heart rate, respiratory rate, and blood oxygen saturation together with their changing trends into an overall health portrait, and then outputting real-time status labels such as "safe", "mild warning", or "severe warning" based on this. The purpose of this approach is to overcome the limitation of a single indicator being easily affected by occasional fluctuations, and to improve the accuracy and robustness of judgments through the mutual verification of multiple indicators. In specific implementation, the system first calculates derivative features such as the short-term average, fluctuation amplitude, and rise or fall rate for each vital sign, and then sets thresholds and weights according to the diving operation specifications, and uses a hierarchical weighted algorithm to synthesize a comprehensive score. If the amount of on-site data is sufficient, a classification model based on decision trees or naive Bayes can also be introduced to automatically learn the mapping relationship between each indicator combination and health status within the model, and continuously update the judgment result using a sliding time window. Finally, the platform double-checks the comprehensive score and model output to generate real-time status analysis results, and prompts the diver or the command end through voice or vibration. Because this fusion analysis simultaneously considers instantaneous values, trend values, and multi-indicator coupling characteristics, it not only significantly reduces single-point failures and false alarm rates, but also enables early detection of slowly accumulating risks, making rescue and intervention more targeted and timely. Preferably, see Figure 2 The fusion analysis of the target vital sign data to obtain the diver's real-time status analysis results includes: Obtaining the current diving mission type information, and based on the type information, calling corresponding fusion configuration information from a preset fusion analysis parameter library, the fusion configuration information including the conditional probability distribution information of each vital sign data under each diver's state level, and the prior probability information of the state level; Specifically, the system first identifies the diving mission type, such as scientific sampling, equipment maintenance, or emergency rescue. This is then used as a search key to retrieve the corresponding fusion configuration from the fusion analysis parameter library. This fusion configuration refers to a set of parameters pre-derived from historical data from a large number of similar missions. These parameters include the conditional probability distributions of heart rate, respiratory rate, and blood oxygen saturation at the safe, mild warning, and severe warning levels, as well as the a priori probabilities of each level within the mission scenario. This approach allows subsequent state determination calculations to fully account for the varying intensity and risk of each mission, enabling intelligent threshold adjustment tailored to each mission. In implementation, the mission type identifier is automatically passed to the mission management module before diving or upon entry. The parameter library uses a key-value index to quickly retrieve the corresponding entry and load it into memory as input for the real-time fusion algorithm. If a mission changes mid-dive (e.g., from reconnaissance to rescue), a hot update is triggered to reload the new configuration. The beneficial effects of this technical design are: on the one hand, the correspondence between physical signs and status levels is accurately characterized through conditional probability distribution, which can improve the judgment accuracy of Bayesian inference or weighted scoring models; on the other hand, the scenario weights brought by the prior probability are used to achieve sensitive response to special high-risk tasks, thereby significantly reducing false alarms and missed alarms, and ensuring the real-time safety of divers in changing operating scenarios.

[0044] Preferably, calling corresponding fusion configuration information from a preset fusion analysis parameter library according to the type information includes: Acquire a historical diving mission dataset, the dataset including multiple diving mission types, environmental condition information of each diving mission, individual characteristic data of the diver, and actual status level label of the diver; Specifically, the historical diving mission dataset refers to a comprehensive data package collected and archived during past actual dives. This data includes mission type (e.g., scientific sampling, equipment maintenance, emergency rescue), environmental conditions such as temperature, pressure, and depth, as well as diver's vital signs during the mission, such as heart rate, respiratory rate, and blood oxygen saturation, along with the actual status level (safe, mild warning, severe warning) annotated by medical personnel. This step aims to provide a rich, authentic, and traceable sample source for subsequent statistical modeling, ensuring the representativeness of the probabilistic model. During implementation, the project management platform exports raw files from dive computers, smart wearable devices, and operation logs, interpolates missing items, manually reviews and removes significant outliers, and stores them in the analysis database using a unified field format. Mission numbers and diver identities are also desensitized to protect privacy. This ensures that subsequent analysis is based on high-quality data, reduces model bias, and improves the reliability of safety assessments.

[0045] Classifying the historical data set according to diving mission types to obtain multiple mission type subsets, each mission type subset corresponding to historical data of a type of diving mission; Specifically, task type classification refers to splitting mixed historical records into multiple subsets according to predefined work scenario labels, such as storing data from scientific research tasks separately from data from rescue tasks. Its purpose is to solve the problem that different task intensities and work postures have different effects on physical signs, so that subsequent statistical features are more homogeneous. During implementation, the "task type" field or keyword matching in the task management system can be used to batch filter data entries, and then scripts can be used to automatically write the filtering results into separate data tables or folders; if the task type is missing or confused, the log records will be manually checked for labeling or merging. The classified subsets enable the model to more accurately learn the specific relationship between physical signs and risks in each task scenario, thereby improving the targeted nature of the analysis.

[0046] The vital sign data in each task type subset are classified and counted according to the corresponding state level labels, and the statistical characteristics of each type of vital sign data under various state levels are calculated, including expected value, variance, and distribution form, which are used to construct a conditional probability distribution model; Specifically, within each task type subset, the heart rate, respiratory rate, and blood oxygen saturation are statistically analyzed for both centralized and discrete indicators based on the three medically labeled status levels of "safe," "mild warning," and "severe warning," forming a summary table of features such as expected value, variance, and skewness. The purpose of this is to extract the core features of the distribution of physical signs under "the same task type and the same risk level" to serve as a priori morphological reference for the conditional probability distribution. The specific operations include: first aggregating by status level using a database grouping statement or data analysis script, then calling a statistical function to calculate the mean, variance, and distribution form (such as whether it is approximately normal) of each indicator; if the sample size is insufficient, a self-service sampling method is used to expand the confidence interval. The statistical results provide a quantitative basis for subsequent model fitting and can reduce the uncertainty caused by loose parameters.

[0047] Based on the statistical characteristics, a parameter fitting method is used to establish a conditional probability distribution model for each type of vital sign data at different state levels, wherein the conditional probability distribution model includes probability distribution functions of target heart rate, target respiratory rate, and target blood oxygen saturation at each state level; Specifically, the conditional probability distribution model uses a mathematical function to describe the probability of a certain value for a given state indicator. Its purpose is to enable the system to infer the risk level from the value of the physical sign during real-time analysis. During implementation, the first step is to select a distribution type that matches the statistical morphology: if the heart rate is approximately normal, use the normal distribution function; if the respiratory rate is biased towards the right long tail, use the lognormal or gamma distribution; then use maximum likelihood estimation or least squares to solve for the distribution parameters. To prevent overfitting, cross-validation or the Bayesian Information Criterion can be used to select the optimal model. After completion, three combinations of the probability density function for heart rate, respiratory rate, and blood oxygen saturation are obtained for different state levels. This model enables the system to quickly calculate the posterior risk through methods such as Bayesian inference, thereby improving the accuracy of real-time judgments.

[0048] Count the historical occurrence frequencies of different status levels under each task type, calculate the prior probability of each status level based on the frequency distribution, and form a prior probability set of the status level; Specifically, the prior probability refers to the proportion of independent historical occurrences of each risk level within a specific task type, reflecting the inherent risk of the task. This is intended to assign higher or lower initial risk weights during real-time inference, ensuring that the results are realistic. The implementation process is to count the number of occurrences of the three status levels within each task subset and divide it by the total number of records in that subset to obtain a percentage. If the data distribution is extremely unbalanced, a Laplace correction can be introduced to prevent the probability from being zero. The resulting set of prior probabilities can be directly applied to the Bayesian formula, improving the robustness of the overall model to extreme situations.

[0049] The conditional probability distribution model of the physical signs corresponding to each diving task type and the prior probability set of the state level are used as fusion configuration information, indexed by task type, and stored in a fusion analysis parameter library to complete the construction of the fusion analysis parameter library; Specifically, the conditional probability distribution model of the physical signs and the prior probability of the state level corresponding to each task type are combined into a fusion configuration record and written into the parameter library with the task type as the index key. The purpose of this library is to provide a standard template for one-click call for real-time monitoring, avoiding the delays caused by temporary modeling. During implementation, a document database or key-value database can be used to serialize model parameters into structured data; version numbers and timestamps are also recorded to support rolling model upgrades. The existence of the parameter library enables the system to load on demand and take effect immediately, significantly reducing the configuration overhead during task switching.

[0050] The current diving task type information is obtained, and the type information is used as a search keyword to call the fusion configuration information corresponding to the current task type from the fusion analysis parameter library.

[0051] Specifically, before the diver goes into the water, the task management module will send the current task type identifier to the monitoring system. The system uses this as a search keyword to quickly read the corresponding fusion configuration from the parameter library, including the conditional probability distribution function of the three types of physical signs and the prior probabilities of the three risk levels, and loads it into the memory for the real-time fusion algorithm to call. The purpose of this design is to achieve different thresholds for different tasks and ensure that the evaluation standards are synchronized with the intensity of the work. The specific implementation can use a message queue to trigger loading and a cache hot update mechanism to ensure that the switching delay is controlled at the millisecond level. With this step, the system can maintain a dual balance between judgment sensitivity and false alarm rate in various scenarios such as scientific research and rescue, significantly improving the intelligence level of diving safety assurance.

[0052] Extract target heart rate, target respiratory rate and target blood oxygen saturation from the corrected target vital sign data as observation input data for Bayesian inference; Specifically, the target heart rate, target respiratory rate, and target blood oxygen saturation are three core physiological indicators after the environmental interference has been removed in the previous step. For example, a diver's current heart rate is 78 beats per minute, respiratory rate is 17 times per minute, and blood oxygen saturation is 97 percentage points. The purpose of setting this step is to convert the vital sign readings into usable observations for subsequent probability calculations, ensuring that the analysis only focuses on real physiological fluctuations rather than external noise. In specific implementation, within each sampling cycle, these three values ​​are directly read from the corrected data structure, and the timestamp and diver's identification number are synchronously marked to ensure subsequent multi-source data alignment. By clarifying the observation entry, it can not only reduce repeated data analysis, but also facilitate quick calling, thereby improving the timeliness and traceability of real-time inference.

[0053] Calculating the likelihood probabilities of target heart rate, target respiratory rate, and target blood oxygen saturation at each diver's state level based on the conditional probability distribution information defined in the fusion configuration information; Specifically, likelihood probability refers to the probability of observing the current vital sign value when assuming the diver is in a certain state level. The goal is to use statistical models to quantify the degree of fit between the observed values ​​and each risk level. During implementation, the fusion configuration is first called according to the current task type to obtain the conditional probability distribution function of the three types of vital sign indicators at each state level. Then, the target heart rate, target respiratory rate, and target blood oxygen saturation are substituted into the corresponding functions to obtain three values. The three results are then multiplied together (assuming the conditions are independent) or a table lookup is used to obtain the joint likelihood. This method fully utilizes historical statistical patterns, avoids misjudgments caused by artificially set thresholds, and makes the assessment of the diver's state level more consistent with objective laws.

[0054] Multiplying the likelihood probability of the target physical sign data at each diver status level by the prior probability of the corresponding diver status level to obtain the unnormalized posterior probability corresponding to each diver status level; Specifically, the prior probability reflects the proportion of each risk level in similar tasks. For example, in rescue tasks, the proportion of mild warning or severe warning is naturally higher. The purpose of this step is to combine the inherent risk level with the degree of match between the current physical signs to form the original support for each state level. The specific method is: read the prior probabilities of the three risk levels one by one, multiply them by the corresponding likelihood probabilities obtained in the previous step, and obtain three unnormalized posterior probabilities. This operation follows the Bayesian principle, which can give rare but dangerous states appropriate weight in high-risk tasks and reduce missed reports.

[0055] Normalizing the unnormalized posterior probability to obtain a normalized posterior probability distribution of each diver's status level; Specifically, the unnormalized posterior probabilities are normalized. Normalization involves dividing the three unnormalized posterior probabilities simultaneously by their sum, so that the sum of the three is exactly one, resulting in a standardized posterior probability distribution. This provides directly comparable probability values ​​for the next step and ensures mathematical consistency. Implementation requires only proportional scaling of the three sums; to prevent the denominator from approaching zero, the system adds a very small constant for smoothing in extreme cases. The normalized probability distribution clearly demonstrates the credibility of each state level and facilitates intuitive display of risk trends on a visual interface.

[0056] The diver status level corresponding to the maximum probability in the posterior probability distribution is selected as the real-time health status level of the current diver, and the level and its probability confidence are output to form a real-time status analysis result.

[0057] Specifically, the status level corresponding to the maximum value in the normalized posterior distribution is found and determined as the diver's real-time health status. The purpose of setting this step is to make the most reasonable diagnosis based on the principle of maximum probability, and at the same time give the confidence level for the commander to quickly judge whether intervention is needed. In the specific implementation, the platform records the selected level, corresponding probability, sampling time, and triggers different levels of sound, light or vibration prompts; if the confidence level is lower than the preset threshold, it will also be marked as uncertain and a retest will be recommended. Through this probability-based automatic judgment method, the missed reports and false alarms caused by the simple threshold method can be significantly reduced, and the safety assurance efficiency of underwater operations can be improved.

[0058] The real-time status analysis result is compared with a preset diver health status template corresponding to the real-time environmental data and a safety assessment is performed to obtain a safety assessment result.

[0059] Specifically, the system compares real-time health analysis results with pre-set health status templates for safety assessment. This means that after determining a diver's current health status (e.g., "safe," "mild warning," or "severe warning"), the system will further reference a standard health status template that matches the current environmental conditions to confirm whether the diver's health status falls within the normal, acceptable, or high-risk ranges under the current circumstances, thereby generating a more robust safety assessment. The health status template is a pre-established rule library based on extensive historical diving data. It lists safety ranges for corresponding vital signs under varying water depths, temperatures, and pressures. For example, at depths exceeding 30 meters and temperatures below 15 degrees Celsius, a slightly elevated heart rate and respiratory rate may be a normal compensatory response, while the same values ​​in shallow, high-temperature conditions may indicate fatigue or overheating. This step is intended to prevent the system from misjudging risks based solely on vital sign status levels, thereby incorporating environmental adaptability into the comprehensive assessment and improving the accuracy and explanatory power of the safety analysis. In specific implementation, after completing status identification, the platform will simultaneously extract real-time environmental data such as current water temperature, water pressure, and water depth, and match the corresponding standard template entries from the health template library; then compare the currently identified status level with the recommended status level range indicated in the template to determine whether it is within an acceptable range. If the analysis result is higher than the template recommended level, the system will output a "status abnormality, there is an environmental adaptation risk prompt"; if it is consistent or lower than the standard, the output status is normal or slightly high but acceptable, and other level annotations will be recorded in the safety assessment report. This dual cross-comparison mechanism that combines environmental conditions and physiological status can effectively improve the accuracy of diving safety assessments, especially in complex environments or special tasks, significantly reducing the possibility of false alarms and neglected risks.

[0060] Preferably, the comparing the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data and performing a safety assessment to obtain a safety assessment result includes: Determine index conditions for a diver's health status template based on current diving mission type information and real-time environmental data, wherein the health status template includes status level threshold settings and response measures under different environmental conditions; Specifically, health status templates are pre-set, structured sets of rules used to guide divers' health assessments and risk responses under various environmental conditions, such as water depth, water pressure, and water temperature. These templates typically include threshold ranges for vital signs, health status classifications (e.g., normal, warning, critical), and corresponding responses (e.g., reducing intensity, surfacing, or aborting the mission). For example, a template specifies that a water temperature <10°C and a heart rate above 120 bpm constitute a moderate warning state. This step aims to quickly identify the most applicable health assessment template based on actual diving mission scenarios (e.g., deep-sea salvage, equipment maintenance, diving training) and current environmental conditions (e.g., high pressure and low temperature), providing an accurate reference system for subsequent real-time health assessments. Different missions correspond to different risk tolerances and warning mechanisms, so templates must be indexable and task-sensitive. During implementation, a health status template library should be established, with index fields defined for each template, such as mission type = training, water temperature range = 5-15°C, and water pressure range = 1.2-3 atm. After obtaining the current mission type and environmental data, a conditional matching algorithm (such as a multi-field joint hash index and a conditional filtering expression tree) is used to locate the template that best meets the current conditions. This template serves as a reference for the next step. This step transitions from generic data to a personalized assessment strategy, ensuring that the system can dynamically adjust health assessment rules based on the actual complexity and risk level of the diving mission. This dynamic adaptation mechanism enhances the flexibility and practicality of the assessment system, avoiding the risk of misjudgments or omissions.

[0061] Calling a target health status template that matches the index condition from a preset health status template library, wherein the target health status template records the safety recommendations and risk warning strategies corresponding to each status level; Specifically, calling a template involves extracting a template from an established library of health status templates that fully or most closely matches the current index criteria and loading its content for subsequent real-time status comparison and decision support. The target health status template not only includes the corresponding vital sign threshold ranges for each health status level (e.g., normal, mild warning, severe warning, and critical), but also includes corresponding safety recommendations and risk mitigation measures, such as the need to immediately terminate the mission and ascend if the critical level is reached.

[0062] The purpose of this step is to quickly read and load the complete template content after selecting appropriate template index conditions. This allows subsequent real-time health level assessment and risk response to be based on clear, unambiguous, and structured rules. The template call mechanism provides the system with a standardized and modular risk assessment basis. In practice, template content in the target format can be retrieved from the template database through path mapping or index reference. The template structure typically includes three fields: 1) acclimatization interval, 2) a table of physical signs and levels, and 3) level-corresponding recommendations. For example, a template specifies that "heart rate >130 bpm and water pressure >2.5 atm" correspond to a hazardous level and recommends "mission abort + rapid ascent." This template is loaded into the cache or model module for subsequent use. This step ensures a standardized basis for the health assessment process, enabling the entire assessment system to support mission customization and policy reasoning, while also providing the necessary foundation for a closed-loop risk response system. The template-based design also facilitates subsequent updates and expansions, such as the addition of templates for extreme cold diving or hyperbaric experimentation.

[0063] According to the real-time status analysis result, obtaining the real-time health status level and its corresponding probability confidence output based on Bayesian inference; Specifically, Bayesian inference is a probability-based reasoning method used to calculate the probability of each health status level (such as normal, warning, and dangerous) under the current observed vital sign data and environmental conditions, and output the most likely status level and its corresponding confidence level. For example, the system calculates that the current state is "alert" with a confidence level of 87%. This step aims to input observed vital sign parameters (such as real-time heart rate and respiratory rate) into the model, integrating them with historical data distributions and thresholds in the state template for inference, resulting in a dynamic, data-driven health status assessment. Using Bayesian methods, prior knowledge and current data can be integrated to provide results with confidence assessments rather than absolute value judgments, enhancing system robustness. In implementation, the system uses the current vital sign value as the observation variable and invokes a Bayesian classifier or Bayesian network model trained based on a prior probability distribution to output the posterior probability of each state level. For example, normal (10%), warning (87%), and critical (3%) are used. The current state level is determined as "alert" based on the maximum a posteriori principle (MAP). This step transforms traditional rule-based judgments into probabilistic modeling, enhancing the system's robustness in complex environments and noisy data, and improving the reliability and interpretability of the diagnosis. The probabilistic output also facilitates integration with subsequent risk strategy systems, such as setting intervention intensity or escalation mechanisms based on confidence levels.

[0064] Comparing the real-time health status level with the status level definition in the target health status template to determine whether the status level belongs to the warning level or danger level category; Specifically, status level comparison involves matching the health level output by the Bayesian model with the various status levels defined in the template to determine whether it represents a level requiring intervention (such as a warning level or a critical level). For example, if the output level is "Severe Warning," but the template specifies it as "Immediate Intervention Required," the level will be labeled "Intervention Level." This step determines whether the current health status has triggered the warning or intervention threshold and is a key decision point in the risk closed-loop process. Only when the system confirms the current status is at a warning or critical level will the corresponding policy generation logic be initiated; otherwise, the status is considered safe and only routine records are made. In implementation, the system compares the output level label (e.g., level=2) with the level number table in the template (e.g., 0=Normal, 1=Mild Warning, 2=Severe Warning, 3=Critical) and reads the corresponding "Triggers Intervention" Boolean flag, or directly determines whether it exceeds a set threshold (e.g., level≥2). The result of this determination serves as the trigger for generating recommendations. This step allows the system to maintain high sensitivity while avoiding unnecessary overreactions, ensuring a stable and reliable health assessment system. The comparison mechanism can also be designed to support individualized strategies, such as lowering the trigger threshold for high-risk individuals to achieve customized risk prevention and control.

[0065] When the health status level is a warning level or a danger level, generating a safety intervention suggestion for the current diver according to the response strategy corresponding to the status level in the target health status template, the safety intervention suggestion including a buoyancy suggestion, enhanced monitoring, or mission termination; Specifically, safety intervention recommendations automatically invoke the corresponding policy in the template based on the health status level and its meaning, generating specific action recommendations to guide the diver or command system. For example, if the level is "dangerous," the system might recommend "immediately and slowly ascend to a safe depth and abort the current mission." This step aims to establish a closed-loop risk response, automating the entire process from status assessment to policy output. This ensures that when health risks arise during a diving mission, the system can provide actionable recommendations to safeguard personnel safety. In practice, the system reads the response policy field in the target template that matches the current level. The policy content can include multiple dimensions: recommended action (ascent, monitoring, mission termination); priority (high, medium, low); execution method (manual operation, system prompt, remote command); and auxiliary information (e.g., estimated ascent time, target depth). The recommendation generation module formats this information and outputs it as structured commands or graphical prompts. Through this step, the system implements intelligent intervention capabilities from data perception to proactive feedback, significantly reducing the risk of accidents caused by delayed response or human misjudgment. The modular design of intervention recommendations also supports subsequent interface docking with smart bracelets, wearable devices, remote control platforms, etc.

[0066] The health status level, probability confidence value and generated safety intervention suggestion are output as the safety assessment result of the current diver.

[0067] Specifically, the safety assessment results are a comprehensive, structured output of the diver's current health status, consisting of three core fields: current status level (e.g., "warning"), confidence level (e.g., 87%), and recommended safety strategies (e.g., "immediately ascend and exit the mission"). This output serves as a risk notification for task performers or supervisory platforms. This step aims to integrate these judgment and strategy information into standardized results, enabling the system to provide clear, actionable health assessment information for decision-making, automated control, or archiving. During implementation, each data field is formatted into a JSON structure or UI-displayable data and transmitted to a front-end interface, console, or remote monitoring platform. Some systems also synchronize assessment results to the diver's wearable device, providing real-time notifications and vibration alerts for enhanced scenario adaptability. This output mechanism builds a bridge between the system and humans, facilitating the rapid implementation of health assessment results in real-world operations. It also provides a complete data loop for subsequent statistical analysis, model optimization, and risk assessment.

[0068] Example 2 See Figure 3 Embodiment 2 of the present invention further provides a diver safety monitoring system based on data fusion analysis, the system comprising: A data acquisition module is used to obtain the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate and real-time blood oxygen saturation; An environmental data analysis module is used to analyze the real-time environmental data to obtain an environmental impact factor that reflects the degree of impact of the current environment on the vital sign data; A vital sign data correction module, configured to correct the real-time vital sign data according to the environmental influencing factors to obtain corrected target vital sign data; A fusion analysis module is used to perform fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; A safety assessment module is used to compare the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data and perform a safety assessment to obtain a safety assessment result; At least one processor, at least one memory, and computer program instructions stored in the memory, which implement the method as described in embodiment 1 when the computer program instructions are executed by the processor.

[0069] Specifically, a diver safety monitoring system based on data fusion analysis provided by an embodiment of the present invention is adopted, and the system includes: a data acquisition module for acquiring real-time vital sign data and real-time environmental data of the diver, the real-time environmental data including real-time water temperature, real-time water pressure and real-time water depth, and the real-time vital sign data including real-time heart rate, real-time respiratory rate and real-time blood oxygen saturation; an environmental data analysis module for analyzing the real-time environmental data to obtain environmental impact factors reflecting the degree of impact of the current environment on the vital sign data; a vital sign data correction module for correcting the real-time vital sign data according to the environmental impact factors to obtain corrected target vital sign data; a fusion analysis module for performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; a safety assessment module for comparing and safety assessing the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data to obtain a safety assessment result; this system is designed for the safety assessment of the diver in a complex underwater environment. To address the challenges brought by the measurement and safety monitoring of divers' vital signs, a safety monitoring method based on data fusion analysis is proposed, focusing on addressing the problems of vital sign measurement errors caused by environmental interference and the failure of safety monitoring using a single indicator: First, the diver's vital sign data (such as heart rate, respiratory rate, and blood oxygen saturation) and environmental data (such as water temperature, water pressure, and water depth) are collected in real time. By constructing an environmental template database and a sensitivity matrix that reflects the relationship between the environment and vital sign responses, the degree of environmental impact on the vital signs is analyzed and corresponding environmental impact factors are generated. Then, by establishing a multi-channel mapping model, these environmental impact factors are used to correct the original vital sign data to obtain target vital sign data that is more consistent with the actual physiological state. Subsequently, the Bayesian inference method is used to fuse and analyze the vital sign data of various dimensions in combination with the current diving mission type to generate the diver's real-time status level. Finally, the analysis results are compared with the preset health status template, and a safety assessment result is generated based on the status level and confidence level. Specific risk intervention recommendations (such as ascent or mission termination) are output when necessary. Overall, through the technical path of environmental perception modeling + physical sign correction + probability fusion reasoning + template comparison, accurate identification of the diver's safety status and safety warning are achieved. Specifically, the processor may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits for implementing the embodiments of the present invention.

[0070] The memory may include a large-capacity storage for data or instructions. By way of example, and not limitation, the memory may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disk, a magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory may include removable or non-removable (or fixed) media. Where appropriate, the memory may be internal or external to the data processing device. In certain embodiments, the memory is non-volatile solid-state memory. In certain embodiments, the memory includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0071] The processor implements any one of the diver safety monitoring methods based on data fusion analysis in the above embodiments by reading and executing computer program instructions stored in the memory.

[0072] In summary, the embodiments of the present invention provide a diver safety monitoring method and system based on data fusion analysis.

[0073] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0074] The functional blocks shown in the block diagrams described above can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they may be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, and the like. When implemented in software, the elements of the present invention are programs or code segments used to perform the desired tasks. Programs or code segments may be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave over a transmission medium or communication link. "Machine-readable medium" may include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memory, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like. Code segments may be downloaded via a computer network such as the Internet or an intranet.

[0075] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant location, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0076] It should also be noted that the exemplary embodiments described herein describe methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the steps described above. In other words, the steps may be performed in the order described in the embodiments, or in a different order, or several steps may be performed simultaneously.

[0077] The above description is only a specific embodiment of the present invention. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention.

Claims

1. A diver safety monitoring method based on data fusion analysis, characterized in that: The method comprises: Acquire the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure, and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate, and real-time blood oxygen saturation; Analyzing the real-time environmental data to obtain an environmental impact factor reflecting the degree of impact of the current environment on the vital sign data; Correcting the real-time vital sign data according to the environmental impact factors to obtain corrected target vital sign data; Performing fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; The real-time status analysis result is compared with a preset diver health status template corresponding to the real-time environmental data and a safety assessment is performed to obtain a safety assessment result.

2. The diver safety monitoring method based on data fusion analysis according to claim 1 is characterized in that: The analysis of the real-time environmental data to obtain environmental impact factors reflecting the degree of impact of the current environment on the vital sign data includes: Constructing an environmental template database based on the diving mission type, diving time, and individual diver characteristics. The environmental template database includes ideal environmental parameter range information corresponding to multiple diving mission types. The ideal environmental parameter range information includes a safe water temperature range, a safe water pressure range, and a safe water depth range. Comparing the real-time environmental data with the environmental template database to determine whether the real-time water temperature is within the safe water temperature range, whether the real-time water pressure is within the safe water pressure range, and whether the real-time water depth is within the safe water depth range; When the real-time water temperature is within the safe water temperature range, the real-time water pressure is within the safe water pressure range, and the real-time water depth is within the safe water depth range, a mapping model between environmental data and environmental influencing factors is established based on pre-collected historical environmental data and historical vital sign data, and the real-time environmental data is input into the mapping model to obtain the environmental influencing factors, wherein the environmental influencing factors include a first influencing factor corresponding to heart rate, a second influencing factor corresponding to respiratory rate, and a third influencing factor corresponding to blood oxygen saturation.

3. The diver safety monitoring method based on data fusion analysis according to claim 2 is characterized in that: The establishment of a mapping model between environmental data and environmental impact factors based on pre-collected historical environmental data and historical physical sign data includes: Collecting historical diving data samples, the historical diving data samples including historical environmental data of historical water temperature, historical water pressure, and historical water depth, and historical vital sign data corresponding to historical heart rate, historical respiratory rate, and historical blood oxygen saturation. The conditions for collecting the historical diving data samples include different diving mission types, different diving times, and different individual characteristics of divers; Normalizing the historical vital sign data collected from the historical diving data samples to obtain normalized historical vital sign data; Statistical analysis is performed on the normalized historical vital sign data and the corresponding historical environmental data. Correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response of various types of vital sign data to various types of environmental data, thereby forming a sensitivity matrix between the environment and vital signs. Each element of the matrix represents the sensitivity weight value of each type of vital sign data to each type of environmental data. Based on the sensitivity matrix, a multi-channel modeling framework is constructed, and mapping sub-models are constructed for each type of vital sign data. The mapping sub-model takes water temperature, water pressure, and water depth as input, and takes the environmental impact factors corresponding to the various types of vital sign data as output. The sub-model includes a weighted regression model, a neural network model, or an integrated learning model.

4. The diver safety monitoring method based on data fusion analysis according to claim 3 is characterized in that: The normalized historical vital sign data and the corresponding historical environmental data are statistically analyzed, and correlation coefficient analysis, covariance analysis, and mutual information analysis are used to calculate the response degree of various vital sign data to various environmental data, and form a sensitivity matrix between the environment and the vital sign, including: For each data sample in the normalized historical vital sign data, according to its collection time point, the water temperature value, water pressure value and water depth value at the corresponding time point are extracted from the normalized historical environmental data to form a vital sign environment sample pair; The physical sign and environmental sample pairs are classified according to the physical sign data type, and paired sample sets between various physical sign data and various environmental data are extracted respectively; Calculating response strength indices for each of the paired sample sets, wherein the response strength indices include Pearson correlation coefficients, covariance values, and mutual information values, and are used to characterize linear or nonlinear correlations between physical signs and environmental parameters; Performing weighted fusion on the response intensity indexes to obtain response sensitivity scores of various types of vital sign data to various types of environmental data; Arrange and combine the response sensitivity scores to construct an initial sensitivity matrix between physical signs and environments, wherein each row of the initial sensitivity matrix corresponds to a type of physical sign data, each column corresponds to a type of environmental data, and each matrix element is a sensitivity score of a corresponding physical sign to a corresponding environment; Linear normalization is performed on the elements in the initial sensitivity matrix, and the sensitivity scores are converted into standard weight value intervals to form a normalized sensitivity matrix between the environment and physical signs.

5. The diver safety monitoring method based on data fusion analysis according to claim 2 is characterized in that: The real-time vital sign data is corrected according to the environmental impact factor to obtain the corrected target vital sign data, which includes: Correcting the real-time heart rate according to the first influencing factor to obtain a target heart rate; Correcting the real-time respiratory rate according to the second influencing factor to obtain a target respiratory rate; Correcting the real-time blood oxygen saturation according to the third influencing factor to obtain a target blood oxygen saturation; The target vital sign data is determined according to the target heart rate, target respiratory rate and target blood oxygen saturation.

6. The diver safety monitoring method based on data fusion analysis according to any one of claims 1 to 5, characterized in that: The fusion analysis of the target vital sign data to obtain the real-time status analysis result of the diver includes: Obtaining the current diving mission type information, and based on the type information, calling corresponding fusion configuration information from a preset fusion analysis parameter library, the fusion configuration information including the conditional probability distribution information of each vital sign data under each diver's state level, and the prior probability information of the state level; Extract target heart rate, target respiratory rate and target blood oxygen saturation from the corrected target vital sign data as observation input data for Bayesian inference; Calculating the likelihood probabilities of target heart rate, target respiratory rate, and target blood oxygen saturation at each diver's state level based on the conditional probability distribution information defined in the fusion configuration information; Multiplying the likelihood probability of the target physical sign data at each diver status level by the prior probability of the corresponding diver status level to obtain the unnormalized posterior probability corresponding to each diver status level; Normalizing the unnormalized posterior probability to obtain a normalized posterior probability distribution of each diver's status level; The diver status level corresponding to the maximum probability in the posterior probability distribution is selected as the real-time health status level of the current diver, and the level and its probability confidence are output to form a real-time status analysis result.

7. The diver safety monitoring method based on data fusion analysis according to claim 6 is characterized in that: The calling corresponding fusion configuration information from a preset fusion analysis parameter library according to the type information includes: Acquire a historical diving mission dataset, the dataset including multiple diving mission types, environmental condition information of each diving mission, individual characteristic data of the diver, and actual status level label of the diver; Classifying the historical diving mission data set according to diving mission types to obtain multiple mission type subsets, each mission type subset corresponding to historical data of a type of diving mission; The vital sign data in each task type subset are classified and counted according to the corresponding state level labels, and the statistical characteristics of each type of vital sign data under various state levels are calculated, including expected value, variance, and distribution form, which are used to construct a conditional probability distribution model; Based on the statistical characteristics, a parameter fitting method is used to establish a conditional probability distribution model for each type of vital sign data at different state levels, wherein the conditional probability distribution model includes probability distribution functions of target heart rate, target respiratory rate, and target blood oxygen saturation at each state level; Count the historical occurrence frequencies of different status levels under each task type, calculate the prior probability of each status level based on the frequency distribution, and form a prior probability set of the status level; The conditional probability distribution model of the physical signs corresponding to each diving task type and the prior probability set of the state level are used as fusion configuration information, indexed by task type, and stored in a fusion analysis parameter library to complete the construction of the fusion analysis parameter library; The current diving task type information is obtained, and the type information is used as a search keyword to call the fusion configuration information corresponding to the current task type from the fusion analysis parameter library.

8. The diver safety monitoring method based on data fusion analysis according to claim 6, characterized in that: The real-time status analysis result is compared with a preset diver health status template corresponding to the real-time environmental data and a safety assessment is performed to obtain a safety assessment result, including: Determine index conditions for a diver's health status template based on current diving mission type information and real-time environmental data, wherein the health status template includes status level threshold settings and response measures under different environmental conditions; Calling a target health status template that matches the index condition from a preset health status template library, wherein the target health status template records the safety recommendations and risk warning strategies corresponding to each status level; According to the real-time status analysis result, obtaining the real-time health status level and its corresponding probability confidence output based on Bayesian inference; Comparing the real-time health status level with the status level definition in the target health status template to determine whether the status level belongs to the warning level or danger level category; When the health status level is a warning level or a danger level, generating a safety intervention suggestion for the current diver according to the response strategy corresponding to the status level in the target health status template, the safety intervention suggestion including a buoyancy suggestion, enhanced monitoring, or mission termination; The health status level, probability confidence value and generated safety intervention suggestion are output as the safety assessment result of the current diver.

9. A diver safety monitoring system based on data fusion analysis, characterized in that: The system comprises: A data acquisition module is used to obtain the diver's real-time vital sign data and real-time environmental data, wherein the real-time environmental data includes real-time water temperature, real-time water pressure and real-time water depth, and the real-time vital sign data includes real-time heart rate, real-time respiratory rate and real-time blood oxygen saturation; An environmental data analysis module is used to analyze the real-time environmental data to obtain an environmental impact factor that reflects the degree of impact of the current environment on the vital sign data; A vital sign data correction module, configured to correct the real-time vital sign data according to the environmental influencing factors to obtain corrected target vital sign data; A fusion analysis module is used to perform fusion analysis on the target vital sign data to obtain a real-time status analysis result of the diver; A safety assessment module is used to compare the real-time status analysis result with a preset diver health status template corresponding to the real-time environmental data and perform a safety assessment to obtain a safety assessment result; At least one processor, at least one memory, and computer program instructions stored in the memory, which implement the method according to any one of claims 1 to 8 when the computer program instructions are executed by the processor.

Images