Underwater treadmill emergency control method and system based on user physiological abnormality monitoring

Abstract

The invention discloses an underwater treadmill emergency control method and system based on user physiological anomaly monitoring. The method comprises the steps that time sequence data and water flow state parameters of physiological parameters generated when a user runs on an underwater treadmill are obtained; according to the water flow state parameters, the exercise load level generated by the current water flow to the user is determined; inputting the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data, and generating a corresponding physiological parameter normal range; and calculating the change trend of each physiological parameter based on the time sequence data of the physiological parameters, and generating an emergency control instruction and sending the emergency control instruction to the underwater treadmill in combination with the corresponding normal range of the physiological parameters when it is pre-judged that any physiological parameter exceeds the corresponding normal range of the physiological parameters, so as to control the underwater treadmill to execute emergency operation. According to the invention, physiological abnormality can be accurately pre-judged and emergency control can be triggered in advance according to a dynamic water flow environment during underwater running.

Description

Technical Field

[0001] This invention relates to the field of underwater treadmill technology, and in particular to an emergency control method and system for underwater treadmills based on monitoring user physiological abnormalities. Background Technology

[0002] In underwater fitness equipment such as underwater treadmills, existing technologies typically use physiological sensors to collect users' physiological data in real time and combine this data with preset fixed thresholds to determine whether physiological abnormalities have occurred. Some devices also incorporate water flow sensors to monitor the flow velocity within the water chamber; however, this flow data is generally only used to adjust the water flow intensity to match the user's set training level, without establishing a dynamic correlation between it and the user's actual exercise load. When physiological parameters exceed the fixed threshold, the system usually immediately triggers emergency operations such as shutdown, speed reduction, or alarms. However, this judgment mechanism does not consider the direct impact of water flow conditions in the underwater environment on the user's physiological response. For example, in high-flow-rate water, a reasonable increase in the user's heart rate and blood pressure is a normal physiological adaptation; if the fixed thresholds of land-based equipment are still used for judgment, it is very easy to lead to false alarms. Conversely, in low-flow water, a sudden increase in heart rate may reflect a real pathological risk, but it may be missed because it has not reached a high threshold. Therefore, existing technologies cannot adaptively determine the normal range of physiological parameters according to the dynamic underwater water flow environment, resulting in an inability to accurately reflect the normal range of physiological parameters, thus making it difficult to effectively prevent risks during underwater exercise. Summary of the Invention

[0003] This invention provides an emergency control method and system for underwater treadmills based on monitoring user physiological abnormalities. It can accurately reflect the normal range of physiological parameters based on the dynamic water flow environment during underwater running, thereby accurately judging physiological abnormalities during underwater exercise and reducing the risks of underwater exercise.

[0004] An embodiment of the present invention provides an emergency control method for an underwater treadmill based on user physiological abnormality monitoring, comprising: The system acquires time-series data of physiological parameters and water flow parameters generated by the user while running on the underwater treadmill; wherein the physiological parameters include heart rate, blood pressure, and blood oxygen saturation. Based on the water flow state parameters, determine the level of motion load that the current water flow generates for the user; The exercise load level is input into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters; wherein, the normal ranges of physiological parameters include normal ranges of heart rate, normal ranges of blood pressure, and normal ranges of blood oxygen saturation; Based on the time-series data of the physiological parameters, the changing trend of each physiological parameter is calculated, and combined with the normal range of the corresponding physiological parameters, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter in the future preset period of time. If it is predicted that any of the physiological parameters will exceed the normal range of its corresponding physiological parameter, an emergency control command is generated and sent to the underwater treadmill to control it to perform an emergency operation.

[0005] As an improvement to the above solution, the step of acquiring time-series data of physiological parameters and water flow state parameters generated by the user while running on the underwater treadmill includes the following sub-steps: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of the physiological parameters; The water flow sensing unit installed in the water cavity is used to acquire signals to obtain the water flow state parameters; The time-series data of the physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of the physiological parameters and the water flow state parameters.

[0006] As an improvement to the above solution, determining the motion load level of the current water flow on the user based on the water flow state parameters includes the following sub-steps: The fluid load intensity of the water flow acting on the user's body is calculated based on the water flow state parameters. The fluid load intensity is compared with a plurality of preset load levels; Based on the comparison results, the exercise load level is determined.

[0007] As an improvement to the above solution, the step of inputting the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters includes the following sub-steps: The exercise load level is input as an input variable into the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the physiological parameter distribution characteristics in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of the physiological parameters, the normal range of the physiological parameters is generated.

[0008] As an improvement to the above scheme, the step of calculating the changing trend of each physiological parameter based on the time-series data of the physiological parameters, and predicting whether any physiological parameter will exceed its corresponding normal range within a preset period of time, includes the following sub-steps: Extract a local data segment within a preset time window from the time-series data of the physiological parameters; Linear regression processing is performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as the trend of change of each of the physiological parameters; Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter within a preset period of time in the future.

[0009] Another embodiment of the present invention provides an emergency control device for an underwater treadmill based on user physiological abnormality monitoring, characterized in that it includes: The acquisition module is used to acquire time-series data of physiological parameters and water flow state parameters generated by the user while running on the underwater treadmill; wherein, the physiological parameters include heart rate, blood pressure and blood oxygen saturation; The determination module is used to determine the level of motion load on the user caused by the current water flow based on the water flow state parameters; The generation module is used to input the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters; wherein, the normal ranges of physiological parameters include normal ranges of heart rate, blood pressure and blood oxygen saturation. The judgment module is used to calculate the changing trend of each physiological parameter based on the time series data of the physiological parameters, and, in combination with the normal range of the corresponding physiological parameters, predict whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter in the future preset period of time. The execution module is used to generate an emergency control command and send it to the underwater treadmill if it is predicted that any of the physiological parameters will exceed the normal range of the corresponding physiological parameter, so as to control the treadmill to perform an emergency operation.

[0010] As an improvement to the above solution, the acquisition module is specifically used for: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of the physiological parameters; The water flow sensing unit installed in the water cavity is used to acquire signals to obtain the water flow state parameters; The time-series data of the physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of the physiological parameters and the water flow state parameters.

[0011] As an improvement to the above solution, the determining module is specifically used for: The fluid load intensity of the water flow acting on the user's body is calculated based on the water flow state parameters. The fluid load intensity is compared with a plurality of preset load levels; Based on the comparison results, the exercise load level is determined.

[0012] As an improvement to the above solution, the generation module is specifically used for: The exercise load level is input as an input variable into the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the physiological parameter distribution characteristics in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of the physiological parameters, the normal range of the physiological parameters is generated.

[0013] As an improvement to the above solution, the judgment module is specifically used for: Extract a local data segment within a preset time window from the time-series data of the physiological parameters; Linear regression processing is performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as the trend of change of each of the physiological parameters; Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter within a preset period of time in the future.

[0014] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention first acquires time-series data of physiological parameters and water flow state parameters generated when a user runs on an underwater treadmill. Then, based on the water flow state parameters, it determines the exercise load level of the user caused by the current water flow and inputs the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters. This allows the water flow state parameters to be used as the basis for determining the user's exercise load level, freeing the judgment of physiological abnormalities from the limitation of fixed thresholds. Furthermore, the load-physiological normal range prediction model trained based on historical exercise data is used to dynamically generate normal ranges of heart rate, blood pressure, and blood oxygen saturation that match the exercise load level, so as to accurately distinguish between physiological fluctuations caused by water flow and real abnormalities. Furthermore, this embodiment of the invention calculates the changing trends of each physiological parameter based on time-series data and, combined with the corresponding normal range of the physiological parameters, predicts whether any physiological parameter will exceed its corresponding normal range within a preset time period. When it is predicted that any physiological parameter will exceed its corresponding normal range, an emergency control command is generated and sent to the underwater treadmill to control it to perform an emergency operation. This achieves proactive identification of abnormal risks, generating emergency control commands and controlling the underwater treadmill to perform emergency operations before an abnormality is predicted, avoiding reliance on post-event response. Therefore, this embodiment of the invention can effectively and accurately predict physiological abnormalities and trigger emergency control in advance based on the dynamic water flow environment during underwater running. Attached Figure Description

[0015] Figure 1 is a flowchart illustrating an emergency control method for an underwater treadmill based on monitoring user physiological abnormalities, according to an embodiment of the present invention. Figure 2 is a schematic diagram of the structure of an underwater treadmill emergency control system based on user physiological abnormality monitoring provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Referring to Figure 1, it is a flowchart illustrating an emergency control method for an underwater treadmill based on user physiological anomaly monitoring according to an embodiment of the present invention. The emergency control method for an underwater treadmill based on user physiological anomaly monitoring includes: S10: Acquire time-series data of physiological parameters and water flow parameters generated when the user runs on the underwater treadmill; among which, physiological parameters include heart rate, blood pressure and blood oxygen saturation; S11, Determine the motion load level of the current water flow on the user based on the water flow state parameters; S12, input the exercise load level into the load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters; among which, the normal range of physiological parameters includes the normal range of heart rate, normal range of blood pressure and normal range of blood oxygen saturation; S13, calculate the changing trend of each physiological parameter based on the time series data of the physiological parameters, and predict whether any physiological parameter will exceed its corresponding normal range in the future within a preset period of time, in combination with the corresponding normal range of the physiological parameters. S14. If it is predicted that any physiological parameter will exceed the normal range of its corresponding physiological parameter, an emergency control command is generated and sent to the underwater treadmill to control it to perform emergency operations.

[0018] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention first acquires time-series data of physiological parameters and water flow state parameters generated when a user runs on an underwater treadmill. Then, based on the water flow state parameters, it determines the exercise load level of the user caused by the current water flow and inputs the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters. This allows the water flow state parameters to be used as the basis for determining the user's exercise load level, freeing the judgment of physiological abnormalities from the limitation of fixed thresholds. Furthermore, the load-physiological normal range prediction model trained based on historical exercise data is used to dynamically generate normal ranges of heart rate, blood pressure, and blood oxygen saturation that match the exercise load level, so as to accurately distinguish between physiological fluctuations caused by water flow and real abnormalities.

[0019] Furthermore, this embodiment of the invention also calculates the changing trend of each physiological parameter based on the time-series data of the physiological parameters, and combines the corresponding normal range of the physiological parameters to predict whether any physiological parameter will exceed its corresponding normal range within a preset period of time in the future. When it is predicted that any physiological parameter will exceed its corresponding normal range, an emergency control command is generated and sent to the underwater treadmill to control it to perform emergency operations. This achieves the forward identification of abnormal risks, generating emergency control commands and controlling the underwater treadmill to perform emergency operations before an abnormality is predicted, thus avoiding reliance on post-event response.

[0020] Therefore, the embodiments of the present invention can effectively predict physiological abnormalities and trigger emergency control in advance based on the dynamic water flow environment during underwater running.

[0021] As one embodiment, acquiring time-series data of physiological parameters and water flow state parameters generated when a user runs on an underwater treadmill includes the following sub-steps: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of physiological parameters; Signals are acquired from the water flow sensing unit installed in the water cavity to obtain water flow state parameters; The time-series data of physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of physiological parameters and water flow state parameters.

[0022] In this embodiment, signals are first acquired from the physiological sensing unit worn by the user to obtain time-series data of physiological parameters. The physiological sensing unit is a flexible wearable device adapted to human skin, with built-in heart rate, blood pressure, and blood oxygen saturation sensors. It captures physiological signals during the user's exercise in real time. During the acquisition process, an adaptive filtering algorithm filters out vibration interference from running movements and water environment noise to ensure continuous and stable signals, ultimately forming time-series data of heart rate, blood pressure, and blood oxygen saturation indexed by timestamps. Next, signals are acquired from the water flow sensing unit located within the water cavity to obtain water flow state parameters. The water flow sensing units are evenly distributed at the front, middle, and rear of the underwater treadmill's water cavity, covering different water depths, and simultaneously acquire data on water velocity, flow direction angle, and water pressure acting on the human body area. These data together constitute a complete set of water flow state parameters. Finally, the time-series data of physiological parameters and the water flow state parameters are synchronized based on a common time reference to obtain two types of time-aligned data. During synchronous processing, the system clock of the underwater treadmill is used as a unified reference. Timestamps of the same precision are added to each set of collected physiological parameters and water flow state parameters. The timestamp matching algorithm is used to associate and bind the physiological parameters and water flow state parameters corresponding to the same time node, ensuring strict correspondence between the two types of data in the time dimension during subsequent analysis.

[0023] In this embodiment, the time-series data of physiological parameters and water flow state parameters are synchronized based on a common time reference. During synchronization, the system reference clock of the underwater treadmill is used as the common time reference. First, the deviation value ΔT between the original timestamp of each set of physiological parameter data and water flow state parameter data and the reference clock is calculated. Then, the deviation correction weight ω is determined according to the corresponding transmission delay, and the formula is used to correct the deviation. The original timestamps are corrected to ensure that the timestamps of all data are uniformly calibrated to the system reference clock. The standard timestamp after synchronization Here, ΔT represents the original timestamp of the physiological parameters or water flow state parameters, ΔT represents the deviation between the original timestamp and the system reference clock (calculated by comparing the original timestamp with the system reference clock in real time), and ω represents the deviation correction weight (dynamically adjusted based on the data transmission delay, which is calculated through the communication feedback signal between the main controller and the sensing unit). After correction, the time-series data of physiological parameters (including heart rate, blood pressure, and blood oxygen saturation data) and water flow state parameters (including flow velocity, flow direction, and water pressure data) at the same time are bound one by one according to the order of the timestamps, forming a joint dataset with complete time alignment. This ensures that the two types of data can strictly correspond in the time dimension in subsequent steps such as calculating the exercise load level and generating the normal physiological range, thus ensuring the logical coherence and calculation accuracy of the entire emergency control method.

[0024] As one embodiment, determining the motion load level of the current water flow on the user based on water flow state parameters includes the following sub-steps: Calculate the fluid load intensity of the water flow acting on the user's body based on the water flow state parameters; The fluid load intensity is compared with multiple preset load levels; Based on the comparison results, the exercise load level is determined.

[0025] In this embodiment, firstly, the fluid load intensity exerted by the water flow on the user's body is calculated based on water flow state parameters. Next, the fluid load intensity is compared with multiple preset load levels. These preset load levels are continuous intervals defined based on exercise tolerance test data from different populations. Each interval corresponds to a clearly defined load intensity range, and the interval boundary values ​​have been statistically verified to ensure coverage of load differences under different exercise scenarios. Finally, the exercise load level is determined based on the comparison results. If the calculated fluid load intensity falls within a certain preset interval, the level corresponding to that interval is defined as the current exercise load level exerted by the water flow on the user.

[0026] In this embodiment, the fluid load intensity F is calculated using the following formula based on the water flow velocity ν and the angle θ between the water flow direction and the user's running direction.

[0027]

[0028] Wherein, α is the weighting coefficient for the influence of flow velocity, which is determined based on the user's body posture; β is the weighting coefficient for the influence of water pressure, which is determined based on the underwater force-bearing area of ​​the user's body; and P is the average water pressure exerted on the user's body by the water flow. This formula comprehensively considers the influence of flow velocity, the angle of flow direction on motion resistance, and the direct effect of water pressure, which is closer to the actual load conditions of underwater motion, making the judgment of motion load level more accurate.

[0029] The preset load levels are based on a large amount of measured data from users of different ages, genders, and exercise abilities on underwater treadmills, and are continuous intervals defined through statistical analysis. Each load level corresponds to a specific range of fluid load intensity, and the boundaries between adjacent levels have undergone significance testing to ensure that the load differences between different levels are statistically significant. For example, the level classification considers the tolerance thresholds of different groups of people when exercising underwater. Low load levels correspond to the load range that the human body can easily bear, medium load levels correspond to the normal load range under moderate exercise intensity, and high load levels correspond to the load range close to the upper limit of human tolerance. During the comparison process, the system compares the calculated fluid load intensity F with these preset intervals one by one to determine the specific interval in which F falls.

[0030] When the fluid load intensity F falls within a preset load level range, the system determines the motion load level generated by the current water flow on the user to be the level corresponding to that range. For example, if F falls within the low load level range, the motion load level is low; if it falls within the medium load level range, it is medium; and if it falls within the high load level range, it is high. Once this level is determined, it will serve as the input parameter for the load-physiological normal range prediction model in subsequent steps, providing a basis for generating targeted normal ranges of physiological parameters and ensuring the logical consistency of motion load assessment and physiological state monitoring throughout the entire emergency control process.

[0031] As one embodiment, the exercise load level is input into a load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters, including the following sub-steps: The exercise load level is used as an input variable to the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the distribution characteristics of physiological parameters in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of physiological parameters, normal ranges for physiological parameters are generated.

[0032] In this embodiment, firstly, the exercise load level is input as an input variable to the load-physiological normal range prediction model. This model is an improved random forest regression model, and the training data comes from the historical exercise data of different users at various exercise load levels. After receiving the exercise load level, the model retrieves the corresponding historical physiological parameter dataset for that level through index matching and extracts the distribution characteristics of the dataset, including the mean, variance, and quantile distribution. Then, the normal range of physiological parameters is generated based on the distribution characteristics of the physiological parameters.

[0033] In this embodiment, the load-physiological normal range prediction model is an improved random forest regression model. Its training data comes from a large amount of historical exercise data from different users exercising on underwater treadmills. This historical data is categorized and stored according to exercise load levels. Each category contains complete time-series records of physiological parameters such as heart rate, blood pressure, and blood oxygen saturation at the corresponding load level, as well as the user's basic information (such as age, gender, and exercise habits). When the exercise load level is input into the model, the model first uses a level matching algorithm to locate the dataset in the historical data that matches the current load level. Then, it extracts features from this dataset to obtain the distribution characteristics of physiological parameters, including but not limited to the mean heart rate, standard deviation of blood pressure, interquartile range of blood oxygen saturation, and the frequency distribution of fluctuations of different physiological parameters at that load level. Simultaneously, the model combines the current user's basic information (pre-entered into the system) to initially screen the extracted distribution features, removing data components that significantly differ from the characteristics of the current user group. This ensures that the retrieved distribution features are more closely aligned with the current user's physiological characteristics, laying the foundation for subsequently generating targeted normal ranges for physiological parameters.

[0034] Through the formula R= The normal range R of physiological parameters is determined by the following formula: μ is the mean of historical physiological parameters; k is the confidence level coefficient, which is set according to safety redundancy requirements; σ is the standard deviation of historical physiological parameters; and У is the individual fit coefficient, which is set based on the deviation of the user's recent exercise data. The normal range R of physiological parameters generated by this formula not only conforms to the general pattern of the current exercise load level, but also takes into account individual differences, thus achieving higher accuracy.

[0035] As one embodiment, the changing trend of each physiological parameter is calculated based on time-series data of physiological parameters, and combined with the corresponding normal range of physiological parameters, it is predicted whether any physiological parameter will exceed its corresponding normal range within a preset period in the future, including the following sub-steps: Extract a local data segment within a preset time window from the time-series data of physiological parameters; Linear regression was performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as a trend of change for each physiological parameter; Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any physiological parameter will exceed its corresponding normal range within a preset period of time in the future.

[0036] In this embodiment, firstly, a local data segment within a preset time window is extracted from the time-series data of physiological parameters. The length of the preset time window is set according to the response speed of physiological parameter changes to ensure a complete reflection of recent parameter changes. The window employs a sliding update mechanism to capture the latest data features in real time. Secondly, linear regression processing is performed on the local data segment to obtain the slope value of each physiological parameter. The least squares method is used to fit the trend line of the local data segment, and the slope of the trend line quantifies the rate and direction of parameter change. Finally, the slope value is used as the trend of each physiological parameter. A positive slope indicates an upward trend, while a negative slope indicates a downward trend. The absolute value of the slope corresponds to the rate of change. Based on the trend and the normal range of the physiological parameter, a prediction is made. If the predicted value exceeds the normal range of the corresponding physiological parameter, it is determined that the physiological parameter will be abnormal.

[0037] The time-series data of the physiological parameters include time-series sequences of heart rate, blood pressure, and blood oxygen saturation. Each sequence is indexed by a timestamp, recording the physiological parameter values ​​at various moments during the user's exercise. The length of the preset time window is dynamically set according to the change characteristics of different physiological parameters. For example, heart rate changes are relatively rapid, so the window length is relatively short, while blood pressure and blood oxygen saturation change more gradually, so the window length can be appropriately extended to ensure that the local data segment can fully reflect the recent change trend of the parameter. The window adopts a sliding update mechanism, that is, at fixed intervals (such as a short period matching the data sampling frequency), the window moves forward by one unit, discarding the oldest data and incorporating the most recently collected physiological parameter data, so that the captured local data segment always contains the latest data near the current moment, ensuring the timeliness and accuracy of subsequent trend calculations.

[0038] This algorithm employs a weighted linear regression approach. Unlike traditional least squares, this algorithm assigns different weights to data points at different times within a local data segment—data points closer to the current time have higher weights—to strengthen the influence of recent data on the trend and weaken the interference of earlier data. Specifically, for a selected local data segment containing a physiological parameter (such as heart rate), let the time variable be t (relative time starting from the earliest time within the window), and the parameter value be x. The objective function is then solved. Obtain the regression line Where n is the number of data points within the window. The weight of the i-th data point ( It increases as i increases, and (a) represents the intercept, and b is the slope value. The slope value calculated by this algorithm can more accurately reflect the recent rate of change of physiological parameters.

[0039] In this embodiment, the process of using slope values ​​as the trend of various physiological parameters is as follows: the sign and magnitude of the slope value directly characterize the direction and speed of parameter change: when the slope value is positive, it indicates that the physiological parameter is showing an upward trend in the near term; the larger the slope value, the faster the increase. When the slope value is negative, it indicates that the parameter is showing a downward trend; the larger the absolute value of the slope value, the faster the decrease. When the slope value is close to zero, it indicates that the parameter is stabilizing in the near term with a small change. For example, a positive slope value for heart rate means that the user's heart rate is gradually increasing, while a negative slope value for blood oxygen saturation means that the blood oxygen level is gradually decreasing. This trend information will provide key basis for subsequent anomaly prediction.

[0040] In this embodiment, based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any physiological parameter will exceed its corresponding normal range within a preset future time period. During the prediction, the calculated physiological parameter will be used within the preset future time period. The predicted value after The normal range corresponding to this physiological parameter (such as the normal range of heart rate) ) compare, if < or > If a physiological parameter is determined to exceed the normal range within a preset time period, it is considered to pose an abnormal risk. This method is used to predict the risk for heart rate, blood pressure, and blood oxygen saturation. If any of these parameters is determined to exceed the normal range, the subsequent emergency control procedures are triggered. Specifically, if a certain physiological parameter is determined to exceed the normal range within a preset time period... The predicted value after Through formula It is obtained through calculation. For a certain physiological parameter, a preset time period in the future The predicted value after; b is the value of the physiological parameter at the current time t; b is the slope value (trend of change) obtained in the second step. The preset prediction period length (set according to the advance advance requirement of emergency response); This is a trend correction factor, the value of which is dynamically adjusted according to the parameter type (e.g., heart rate). Slightly greater than 1, to amplify the predicted risk of an upward trend, blood oxygen saturation (Also, appropriate adjustments were made to improve the sensitivity of early warnings).

[0041] In one embodiment, if it is predicted that any physiological parameter will exceed its corresponding normal range, an emergency control command is generated and sent to the underwater treadmill to control it to perform emergency operations. The process is as follows: If it is predicted that any physiological parameter will exceed its corresponding normal range, an emergency control command is immediately generated and sent to the control unit of the underwater treadmill. The emergency control command is set according to the urgency level of the predicted anomaly. After receiving the command, the control unit executes the corresponding emergency operation to ensure the user's exercise safety. The entire process forms a closed loop, with continuous real-time monitoring and response. Emergency operations include stopping the machine, reducing speed, or issuing an alarm, with the specific emergency operation determined according to the actual urgency level.

[0042] Referring to Figure 2, it is a structural schematic diagram of an underwater treadmill emergency control device based on user physiological abnormality monitoring according to an embodiment of the present invention. The underwater treadmill emergency control device based on user physiological abnormality monitoring includes: The acquisition module 10 is used to acquire time-series data of physiological parameters and water flow state parameters generated when the user runs on the underwater treadmill; among which, the physiological parameters include heart rate, blood pressure and blood oxygen saturation; The determination module 11 is used to determine the motion load level of the current water flow to the user based on the water flow state parameters; The generation module 12 is used to input the exercise load level into the load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters; wherein, the normal range of physiological parameters includes the normal range of heart rate, normal range of blood pressure and normal range of blood oxygen saturation. The judgment module 13 is used to calculate the changing trend of each physiological parameter based on the time series data of the physiological parameters, and, in combination with the normal range of the corresponding physiological parameters, predict whether any physiological parameter will exceed the normal range of its corresponding physiological parameter in the future preset period. The execution module 14 is used to generate an emergency control command and send it to the underwater treadmill if it is predicted that any physiological parameter will exceed the normal range of its corresponding physiological parameter, so as to control it to perform emergency operations.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In this embodiment of the invention, the acquisition module 10 first acquires the time-series data of physiological parameters and water flow state parameters generated when the user runs on the underwater treadmill. Then, the determination module 11 determines the exercise load level of the user based on the water flow state parameters. The generation module 12 inputs the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters. This allows the water flow state parameters to be used as the basis for determining the user's exercise load level, freeing the judgment of physiological abnormalities from the limitation of fixed thresholds. Furthermore, the load-physiological normal range prediction model trained based on historical exercise data dynamically generates normal ranges of heart rate, blood pressure, and blood oxygen saturation that match the exercise load level, so as to accurately distinguish between physiological fluctuations caused by water flow and real abnormalities.

[0044] Furthermore, this embodiment of the invention also utilizes the judgment module 13 to calculate the changing trend of each physiological parameter based on the time-series data of the physiological parameters, and combines it with the corresponding normal range of the physiological parameters to predict whether any physiological parameter will exceed its corresponding normal range within a preset period of time in the future. The execution module 14 generates an emergency control command and sends it to the underwater treadmill when it is predicted that any physiological parameter will exceed its corresponding normal range, so as to control it to perform emergency operations. This achieves the forward identification of abnormal risks, generates emergency control commands and controls the underwater treadmill to perform emergency operations before an abnormality is predicted to occur, and avoids relying on post-event response.

[0045] Therefore, the embodiments of the present invention can effectively predict physiological abnormalities and trigger emergency control in advance based on the dynamic water flow environment during underwater running.

[0046] As an improvement to the above solution, the acquisition module is specifically used for: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of physiological parameters; Signals are acquired from the water flow sensing unit installed in the water cavity to obtain water flow state parameters; The time-series data of physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of physiological parameters and water flow state parameters.

[0047] As an improvement to the above solution, the determination module is specifically used for: Calculate the fluid load intensity of the water flow acting on the user's body based on the water flow state parameters; The fluid load intensity is compared with multiple preset load levels; Based on the comparison results, the exercise load level is determined.

[0048] As an improvement to the above solution, the generation module is specifically used for: The exercise load level is used as an input variable to the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the distribution characteristics of physiological parameters in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of physiological parameters, normal ranges for physiological parameters are generated.

[0049] As an improvement to the above solution, the judgment module is specifically used for: Extract a local data segment within a preset time window from the time-series data of physiological parameters; Linear regression was performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as the trend of change of each of the physiological parameters. Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any physiological parameter will exceed its corresponding normal range within a preset period of time in the future.

[0050] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0051] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An emergency control method for an underwater treadmill based on monitoring user physiological abnormalities, characterized in that, include: The system acquires time-series data of physiological parameters and water flow parameters generated by the user while running on the underwater treadmill; wherein the physiological parameters include heart rate, blood pressure, and blood oxygen saturation. Based on the water flow state parameters, determine the level of motion load that the current water flow generates for the user; The exercise load level is input into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters; wherein, the normal ranges of physiological parameters include normal ranges of heart rate, normal ranges of blood pressure, and normal ranges of blood oxygen saturation; Based on the time-series data of the physiological parameters, the changing trend of each physiological parameter is calculated, and combined with the normal range of the corresponding physiological parameters, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter in the future preset period of time. If it is predicted that any of the physiological parameters will exceed the normal range of its corresponding physiological parameter, an emergency control command is generated and sent to the underwater treadmill to control it to perform an emergency operation.

2. The emergency control method for an underwater treadmill based on user physiological abnormality monitoring as described in claim 1, characterized in that, The process of acquiring time-series data of physiological parameters and water flow state parameters generated by the user while running on the underwater treadmill includes the following sub-steps: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of the physiological parameters; The water flow sensing unit installed in the water cavity is used to acquire signals to obtain the water flow state parameters; The time-series data of the physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of the physiological parameters and the water flow state parameters.

3. The emergency control method for an underwater treadmill based on user physiological abnormality monitoring as described in claim 1, characterized in that, The step of determining the motion load level of the current water flow on the user based on the water flow state parameters includes the following sub-steps: The fluid load intensity of the water flow acting on the user's body is calculated based on the water flow state parameters. The fluid load intensity is compared with a plurality of preset load levels; Based on the comparison results, the exercise load level is determined.

4. The emergency control method for an underwater treadmill based on user physiological abnormality monitoring as described in claim 1, characterized in that, The step of inputting the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate the corresponding normal range of physiological parameters includes the following sub-steps: The exercise load level is input as an input variable into the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the physiological parameter distribution characteristics in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of the physiological parameters, the normal range of the physiological parameters is generated.

5. The emergency control method for an underwater treadmill based on user physiological abnormality monitoring as described in claim 1, characterized in that, The step of calculating the changing trend of each physiological parameter based on the time-series data of the physiological parameters, and predicting whether any physiological parameter will exceed its corresponding normal range within a preset future time period, includes the following sub-steps: Extract a local data segment within a preset time window from the time-series data of the physiological parameters; Linear regression processing is performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as the trend of change of each of the physiological parameters; Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter within a preset period of time in the future.

6. An emergency control device for an underwater treadmill based on monitoring user physiological abnormalities, characterized in that, include: The acquisition module is used to acquire time-series data of physiological parameters and water flow state parameters generated by the user while running on the underwater treadmill; wherein, the physiological parameters include heart rate, blood pressure and blood oxygen saturation; The determination module is used to determine the level of motion load on the user caused by the current water flow based on the water flow state parameters; The generation module is used to input the exercise load level into a load-physiological normal range prediction model trained based on historical exercise data to generate corresponding normal ranges of physiological parameters; wherein, the normal ranges of physiological parameters include normal ranges of heart rate, blood pressure and blood oxygen saturation. The judgment module is used to calculate the changing trend of each physiological parameter based on the time series data of the physiological parameters, and, in combination with the normal range of the corresponding physiological parameters, predict whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter in the future preset period of time. The execution module is used to generate an emergency control command and send it to the underwater treadmill if it is predicted that any of the physiological parameters will exceed the normal range of the corresponding physiological parameter, so as to control the treadmill to perform an emergency operation.

7. The underwater treadmill emergency control device based on user physiological abnormality monitoring as described in claim 6, characterized in that, The acquisition module is specifically used for: Signals are collected from the physiological sensing unit worn by the user to obtain time-series data of the physiological parameters; The water flow sensing unit installed in the water cavity is used to acquire signals to obtain the water flow state parameters; The time-series data of the physiological parameters and the water flow state parameters are synchronized according to a common time reference to obtain time-aligned time-series data of the physiological parameters and the water flow state parameters.

8. The underwater treadmill emergency control device based on user physiological abnormality monitoring as described in claim 6, characterized in that, The determining module is specifically used for: The fluid load intensity of the water flow acting on the user's body is calculated based on the water flow state parameters. The fluid load intensity is compared with a plurality of preset load levels; Based on the comparison results, the exercise load level is determined.

9. The underwater treadmill emergency control device based on user physiological abnormality monitoring as described in claim 6, characterized in that, The generation module is specifically used for: The exercise load level is input as an input variable into the load-physiological normal range prediction model, so that the load-physiological normal range prediction model can retrieve the physiological parameter distribution characteristics in the corresponding historical exercise data based on the exercise load level. Based on the distribution characteristics of the physiological parameters, the normal range of the physiological parameters is generated.

10. The underwater treadmill emergency control device based on user physiological abnormality monitoring as described in claim 6, characterized in that, The judgment module is specifically used for: Extract a local data segment within a preset time window from the time-series data of the physiological parameters; Linear regression processing is performed on the local data segment to obtain the slope values ​​of each physiological parameter; The slope value is used as the trend of change of each of the physiological parameters; Based on the changing trends of each physiological parameter and in conjunction with the corresponding normal range of the physiological parameter, it is predicted whether any of the physiological parameters will exceed the normal range of its corresponding physiological parameter within a preset period of time in the future.

Images