A method and system for managing exercise fatigue

Abstract

This application relates to the field of medical technology, specifically to a method and system for managing exercise fatigue. The method synchronously collects heart rate, electromyography, movement, and respiratory sound signals using a multimodal wearable device, extracts characteristic parameters from each signal, performs feature fusion using a weighted average method, predicts fatigue risk types using a time series model, and outputs personalized recovery suggestions accordingly. This application overcomes the shortcomings of traditional single-parameter monitoring methods—poor specificity and inability to comprehensively reflect the user's physical condition—through the synchronous acquisition and fusion analysis of multimodal physiological signals, achieving accurate assessment of complex physiological states. It can predictively assess fatigue status, solving the problems of lag and strong subjectivity in existing subjective assessment methods. Through targeted recovery suggestions, it achieves proactive and targeted physiological regulation intervention, making it particularly suitable for scenarios requiring precise load management, such as cardiopulmonary rehabilitation for the elderly.

Description

Technical Field

[0001] This application relates to the field of medical technology, specifically to a method and system for managing exercise fatigue. Background Technology

[0002] In the field of cardiopulmonary rehabilitation training for the elderly, cardiopulmonary diseases are common ailments that seriously threaten human health. While cardiopulmonary surgery can significantly improve patients' conditions, the accuracy of postoperative rehabilitation management directly affects treatment outcomes and patient safety. However, traditional cardiopulmonary function monitoring methods (such as conventional electrocardiogram and blood pressure monitoring equipment) can only provide limited basic physiological data, making it difficult to comprehensively analyze the dynamic functional state of the cardiopulmonary system. This results in rehabilitation programs lacking scientific basis, easily leading to the risks of overtraining or insufficient recovery, and failing to meet individualized rehabilitation needs.

[0003] In related technologies, rehabilitation assessment and monitoring techniques mainly rely on two types of methods: subjective assessment and single physiological parameter monitoring. Subjective assessment is entirely based on patient self-reports or therapist's direct observation, which suffers from significant lag, strong subjective bias, and difficulty in quantification. While single physiological parameter monitoring can track specific indicators (such as heart rate or blood oxygen saturation) in real time, it focuses on only a single physiological dimension and cannot integrate complex physiological signals, leading to biased and easily misinterpreted assessment results. It is difficult to provide early warnings before fatigue risks appear, severely restricting the proactive intervention capability in the rehabilitation process. Therefore, practitioners urgently need to construct an intelligent closed-loop system that can deeply integrate multimodal data to achieve fatigue prediction and proactive recovery, in order to overcome existing technological bottlenecks and improve the accuracy and efficiency of cardiopulmonary rehabilitation for the elderly. Summary of the Invention

[0004] In response to the problem that related technologies lack the ability to accurately predict exercise fatigue and provide personalized active recovery guidance based on multimodal physiological signal coupling analysis in cardiopulmonary rehabilitation training for the elderly.

[0005] In a first aspect, embodiments of this application provide a method for managing exercise fatigue, the method comprising: Collect heart rate measurement signals, electromyography signals, movement signals, and respiratory sound signals of the target subject; Modal characteristics of heart rate measurement signal, electromyography signal, action signal, and breath sound signal were calculated respectively; A weighted average method is used to fuse all modal features to form a fused feature; Predict the types of fatigue risks for the target object based on fusion characteristics; Recovery recommendations are generated based on the type of fatigue risk.

[0006] In conjunction with the first aspect, in one embodiment, the calculation of modal features of the heart rate measurement signal, electromyography signal, action signal, and respiratory sound signal respectively includes: Calculate the time-domain characteristics and frequency-domain characteristics of heart rate and heart rate variability based on the heart rate measurement signal; Calculate the median frequency characteristics and root mean square characteristics of electromyographic signals; Calculate respiratory rate characteristics and respiratory ratio characteristics based on respiratory sound signals.

[0007] In conjunction with the first aspect, in one embodiment, the acquisition of the heart rate measurement signal of the target object includes: acquiring the electrocardiogram signal or photoplethysmography (PPG) signal of the target object.

[0008] In conjunction with the first aspect, in one implementation, predicting the type of fatigue risk of the target object based on fusion features includes: A time series model is used to analyze the fusion features in order to obtain the probability of fatigue risk within a preset future time period; The fatigue risk of the target object is classified according to the probability of fatigue risk.

[0009] In conjunction with the first aspect, in one implementation, classifying the fatigue risk of the target object based on the fatigue risk probability includes: Based on the probability of fatigue risk, the target subjects are classified into one of the following types: lung-dominant, muscle-dominant, and mixed.

[0010] In conjunction with the first aspect, in one implementation, the method of fusing all modal features using a weighted average to form a fused feature includes: Modal features are fused using a weighted average method based on preset weight coefficients; The fused features are normalized to form fused features.

[0011] In conjunction with the first aspect, in one implementation, the step of outputting recovery recommendations based on the type of fatigue risk includes: The corresponding recovery protocol is extracted from the preset recovery protocol library based on the type of fatigue risk; The corresponding recovery suggestions are output based on the recovery protocol.

[0012] In conjunction with the first aspect, in one implementation, after outputting recovery recommendations based on the type of fatigue risk, the method further includes: Analyze the physiological signals of the target object during the recovery process and generate a recovery progress report; And adjust the recovery recommendations based on the recovery progress report.

[0013] Secondly, embodiments of this application provide a sports fatigue management system, which includes: The data acquisition module is used to acquire the target object's heart rate measurement signal, electromyography signal, movement signal, and respiratory sound signal; The data processing module is used to calculate the modal characteristics of heart rate measurement signals, electromyography signals, action signals, and respiratory sound signals, respectively. The analysis module is used to fuse all modal features using a weighted average method to form a fused feature; The classification module is used to predict the type of fatigue risk of a target object based on fusion features. The recovery guidance module is used to output recovery suggestions based on the type of fatigue risk.

[0014] In conjunction with the second aspect, in one implementation, the recovery guidance module is further configured to output voice and vibration reminders to the target object based on the output recovery suggestions.

[0015] The beneficial effects of the technical solutions provided in this application include: This application overcomes the shortcomings of traditional single physiological parameter monitoring methods, such as poor specificity and inability to fully reflect the user's physical condition, by synchronously acquiring and fusing multimodal physiological signals, thus achieving accurate assessment of complex physiological states. It can predictively assess fatigue status, solving the problems of lag and strong subjectivity in existing subjective assessment methods. Through targeted recovery suggestions, it achieves proactive and targeted physiological regulation intervention, which is particularly suitable for scenarios requiring precise load management, such as cardiopulmonary rehabilitation for the elderly. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the first embodiment of the exercise fatigue management method of this application; Figure 2 This is a flowchart of the first embodiment of the exercise fatigue management method of this application; Figure 3 This is a schematic diagram of the hardware structure of the sports fatigue management device involved in the embodiment of this application. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0018] Among related technologies, there is a lack of intelligent closed-loop systems in cardiopulmonary rehabilitation training for the elderly that can accurately predict exercise fatigue and provide personalized active recovery guidance based on multimodal physiological signal coupling analysis.

[0019] Firstly, this application provides a method for managing exercise fatigue, the method comprising: Step S1: Collect the target subject's heart rate measurement signal, electromyography signal, motion signal, and respiratory sound signal.

[0020] The above step S1 specifically includes: Step S1a: Configure the basic parameters of the multimodal wearable sensing terminal, including sampling frequency, data buffer size, etc. Initialize the communication connection of the edge-cloud collaborative computing platform and load the pre-trained multimodal signal processing model.

[0021] Step S1b: Acquire the ECG signal or PPG signal (photoplethysmography signal) of the target object through the ECG / photoplethysmography module.

[0022] Understandably, the above embodiments provide two non-invasive heart rate monitoring methods to suit different usage scenarios and user preferences, improving the system's applicability and user acceptance. ECG signals provide more accurate information on cardiac activity and are suitable for high-precision monitoring scenarios, while photoplethysmography (PPG) signals are more suitable for daily monitoring, offering better comfort and convenience. This flexibility ensures that the system can operate stably in different environments. In particular, PPG signal acquisition is more comfortable for elderly users, lowering the barrier to entry and improving the feasibility of long-term monitoring.

[0023] Step S1c: Acquire electromyographic signals through the surface electromyography module.

[0024] Step S1d: Collect information such as motion type and speed through the inertial measurement unit; Step S1e: Acquire respiratory sound signals through the bioacoustic module; Step S1f: The acquired signal is initially filtered and its features are extracted by the data processing and communication unit; Step S1g: Compress the processed data and upload it to the edge-cloud collaborative computing platform via Bluetooth / Low Power Wide Area Network.

[0025] Understandably, a multimodal data acquisition system, which uses a surface electromyography module to collect electromyographic signals, an inertial measurement unit to acquire motion information, and a bioacoustic module to monitor respiratory sound signals, combined with preliminary filtering feature extraction and efficient data compression and transmission technology at the edge end, enables a comprehensive and accurate assessment of fatigue status during cardiopulmonary rehabilitation training in the elderly.

[0026] Step S2: Calculate the modal characteristics of the heart rate measurement signal, electromyography signal, action signal, and respiratory sound signal respectively.

[0027] Specifically, step S2 includes: Step S2a: Calculate the time-domain characteristics and frequency-domain characteristics of heart rate and heart rate variability based on the heart rate measurement signal. Calculate the median frequency characteristics and root mean square characteristics of the electromyographic signal; calculate the respiratory rate characteristics and respiratory ratio characteristics based on the breath sound signal.

[0028] Specifically, each acquired modal signal undergoes independent signal enhancement and normalization processing. The normalization method is used to normalize the signal amplitude to the [0, 1] interval. Time-domain and frequency-domain features of each modal signal are extracted, and time-domain features (such as SDNN) and frequency-domain features (VLF / TNF) of heart rate (HR) and heart rate variability (HRV) are calculated. Median frequency (MF) and root mean square (RMS) features of electromyography (EMG) signals are extracted, as are respiratory rate (RF) and inspiratory-to-expiratory ratio (I:E) features.

[0029] Step S3: Use a weighted average method to fuse all modal features to form a fused feature.

[0030] Specifically, based on preset weight settings, a weighted average method is used to fuse multimodal features. The fused features are then normalized using the Min-Max normalization method, normalizing the feature values ​​to the [0, 1] interval.

[0031] It is worth noting that by setting reasonable weighting coefficients, physiological signal features that contribute more to fatigue assessment can be highlighted, thus improving the representativeness of the fused features. The weights can be dynamically adjusted according to individual user differences and historical data, enhancing the system's personalized adaptability. Normalization eliminates the differences in the dimensions and numerical ranges of different features, improving the rationality of feature fusion and the accuracy of subsequent analysis. This feature fusion strategy ensures the effective integration of multimodal information, avoids information redundancy and conflicts, and greatly improves the accuracy of fatigue prediction.

[0032] Step S4: Predict the type of fatigue risk of the target object based on the fusion characteristics.

[0033] Step S4 above includes: Step S4a: Use a time series model to analyze the fusion features to obtain the fatigue risk probability within a preset time period in the future.

[0034] Specifically, the normalized features are input into a pre-trained time-series prediction model (the model can use a Transformer network structure or an LSTM network structure), and the time-series prediction model is used to perform in-depth mining analysis of the feature sequence. The preset sequence length is input to output the fatigue risk probability within a preset time period in the future.

[0035] Step S4b: Classify the fatigue risk of the target object based on the fatigue risk probability obtained in the previous steps.

[0036] Specifically, based on the probability of fatigue risk, the target subjects are classified into one of the following types: lung-dominant, muscle-dominant, and mixed.

[0037] It is worth noting that by distinguishing different types of fatigue, the system can identify the main physiological drivers of fatigue, laying the foundation for providing targeted recovery suggestions. This classification method overcomes the limitations of the "one-size-fits-all" approach in traditional fatigue assessment, enabling more refined fatigue management. In particular, for specific scenarios such as cardiopulmonary rehabilitation for the elderly, it can differentiate between the fatigue states of the cardiopulmonary and muscular systems, providing safer and more effective training guidance. This avoids inappropriate training or recovery measures due to confusion of fatigue types, significantly improving the safety and effectiveness of rehabilitation training.

[0038] Step S5: Output recovery suggestions based on the type of fatigue risk.

[0039] Specifically, step S5 includes: Step S5a: Extract the corresponding recovery protocol from the preset recovery protocol library according to the type of fatigue risk.

[0040] Specifically, select the appropriate recovery protocol library based on the classification results. For example, select the "neuromuscular relaxation technique" recovery protocol for muscle-dominant types.

[0041] Step S5b: Output the corresponding recovery suggestions according to the recovery protocol.

[0042] It is worth noting that personalized recovery guidance plans are generated, such as recommending 15 minutes of neuromuscular relaxation exercises, performing recovery operations through vibration feedback, and using a micro-vibration module with a vibration frequency of 50Hz for assisted relaxation.

[0043] In some preferred embodiments, the exercise fatigue management method further includes: Step S6: Recovery status assessment.

[0044] Specifically, step S6 includes: Step S6a: Analyze the physiological signals of the target object during the recovery process.

[0045] Specifically, multimodal physiological signals continued to be collected. Physiological signals during the recovery process were processed and analyzed in real time, using real-time heart rate monitoring and respiratory rate analysis. Physiological recovery indices during the recovery process were calculated, such as respiratory rate recovery rate and heart rate recovery speed; the recovery results were compared with baseline data, such as physiological indicators at the initial state.

[0046] Step S6b: Generate a recovery progress report based on the indicator data calculated in step S6a.

[0047] It is worth noting that the recovery progress report includes trends in physiological indicators and recommendations.

[0048] Step S6c: Adjust the recovery recommendations based on the recovery progress report.

[0049] It should be noted that the process will determine whether the recovery target has been achieved. If not, return to step S1; if the target is met, proceed with the next steps.

[0050] Step S6c specifically includes: determining the recovery effect over two consecutive recovery cycles (e.g., recovery effect over two consecutive days); adjusting recovery plan parameters based on the determination results (e.g., adjusting the recovery time length); updating the individual user model in the cloud database (including physiological indicators and recovery effects); and optimizing local model parameters by updating the model using transfer learning techniques.

[0051] Step S7: Update configuration.

[0052] Specifically, update the local model according to the latest model version and test and verify the new version. Adjust system configuration parameters based on the test results, such as adjusting feature extraction algorithm parameters; update help documentation and version history, recording update time and changes.

[0053] In a first specific embodiment of the exercise fatigue management method of this application, the following steps are included: Step 1: Initialize system configuration.

[0054] Specifically, configure the basic parameters of the multimodal wearable sensing terminal, including setting the sampling frequency to 125Hz and the data cache size to 4MB; initialize the communication connection of the edge-cloud collaborative computing platform and establish a data transmission channel using a Wi-Fi network; further, load the pre-trained multimodal signal processing model, including the MobileNetV3 algorithm model, and modify the input layer size to 1x1x3.

[0055] Step 2: Data Acquisition and Preprocessing.

[0056] Specifically, the system acquires ECG or PPG signals via an ECG / photoplethysmography (PPG) module for 60 seconds. It also acquires electromyography (EMG) signals via a surface electromyography (SEM) module for 60 seconds; motion type and speed information are acquired via an inertial measurement unit for 60 seconds; and respiratory sound signals are acquired via a bioacoustic module for 60 seconds. The acquired signals undergo preliminary filtering and feature extraction using a bandpass filter with a cutoff frequency of 15-150Hz via a data processing and communication unit. The processed data is then compressed and uploaded to the edge-cloud collaborative computing platform via a Wi-Fi network.

[0057] Step 3: Multimodal feature extraction and fusion.

[0058] Specifically, step 3 includes: Step A: Perform independent signal enhancement and normalization processing on each modal signal, and use the normalization processing method to normalize the signal amplitude to the [0, 1] interval; Step B: Extract the time-domain and frequency-domain features of each modality signal, calculate the time-domain features (such as SDNN) and frequency-domain features (VLF / TNF) of heart rate (HR) and heart rate variability (HRV), extract the median frequency (MF) and root mean square (RMS) features of electromyography signals, and extract the respiratory rate (RF) and inspiratory-to-expiratory ratio (I:E Ratio) features. Step C: Use a weighted average method to fuse multimodal features, with weights set to 0.35:0.35:0.25:0.05:0.05; Step D: Use the Min-Max normalization method to normalize the fused features, normalizing the feature values ​​to the [0, 1] interval.

[0059] Step 4: Fatigue prediction and classification.

[0060] Specifically, step 4 includes: Step A: Input the normalized features into the pre-trained temporal prediction model. The model adopts the Transformer network structure. Step B: Use the Transformer network to perform deep mining analysis of the feature sequence. The input sequence length is 90 seconds, and the output is the fatigue risk probability in the next 45 seconds. Step C: Output the fatigue risk probability and classification results for the next 45 seconds. The classification results include cardiopulmonary dominant type, muscle dominant type, and mixed type. Step D: Select the appropriate recovery protocol library based on the classification results. For example, for muscle-dominant types, select the "Neuromuscular Relaxation Techniques" recovery protocol.

[0061] Step 5: Restore boot execution.

[0062] Specifically, step 5 includes: Step A: Determine the main physiological driving factors of fatigue based on the prediction results. For example, in muscle-dominant fatigue, the main factor is abnormal electromyographic signals. Step B: Select the appropriate recovery protocol library; Step C: Generate a personalized recovery guidance plan, such as suggesting 15 minutes of neuromuscular relaxation exercises; Step D: Perform the recovery operation through vibration feedback, using a micro-vibration module with a vibration frequency of 50Hz for assisted relaxation.

[0063] Step 6: Recovery status assessment.

[0064] Specifically, step 6 includes: Step A: Continue to collect multimodal physiological signals for 15 minutes; Step B: Real-time processing and analysis of physiological signals during the recovery process, using real-time heart rate monitoring and respiratory rate analysis; Step C: Calculate the physiological recovery index during the recovery process, such as the respiratory rate recovery rate and heart rate recovery speed; Step D: Compare the recovery results with baseline data, such as physiological indicators at the initial state; Step E: Generate a recovery progress report, including trends in physiological indicators and recommendations.

[0065] Step 7: Determine if the recovery goal has been achieved. If not, return to step 2. If the recovery goal has been achieved, proceed to step 8.

[0066] Specifically, step 7 includes: Step A: Determine the recovery effect over two consecutive recovery cycles, for example, the recovery effect over two consecutive days; Step B: Adjust the recovery plan parameters based on the judgment results, such as adjusting the recovery time length; Step C: Update the individual user model in the cloud database, including physiological indicators and recovery results; Step D: Optimize the local model parameters and update the model using transfer learning techniques.

[0067] Step 8: Update system configuration and finish.

[0068] Specifically, step 8 includes: Step A: Update the local model according to the latest model version, with a version update frequency of once a month; Step B: Test and verify the new version with 200 test samples. Step C: Adjust system configuration parameters based on test results, such as adjusting feature extraction algorithm parameters; Step D: Update the help documentation and version history, recording the update time and changes.

[0069] In a second specific embodiment of the exercise fatigue management method of this application, the following steps are included: Step 1: Initialize system configuration.

[0070] Specifically, step 1 includes: Step A: Configure the basic parameters of the multimodal wearable sensing terminal, including setting the sampling frequency to 100Hz and the data buffer size to 2MB; Step B: Initialize the communication connection of the edge-cloud collaborative computing platform and establish a data transmission channel using Bluetooth Low Energy mode; Step C: Load the pre-trained multimodal signal processing model, including the MobileNetV3 algorithm model.

[0071] Step 2: Data Acquisition and Preprocessing.

[0072] Specifically, step 2 includes: Step A: Acquire ECG or PPG signals using the ECG / photoplethysmography (PPG) module for 30 seconds; Step B: Acquire electromyography (EMG) signals using a surface EMG module for 30 seconds. Step C: Collect motion type and speed information using an inertial measurement unit for 30 seconds; Step D: Collect respiratory sound signals using the bioacoustic module for 30 seconds; Step E: The acquired signal is initially filtered and its features are extracted by the data processing and communication unit using a bandpass filter with a cutoff frequency of 20-100Hz. Step F: Compress the processed data and upload it to the edge-cloud collaborative computing platform via Bluetooth Low Energy network.

[0073] Step 3: Multimodal feature extraction and fusion.

[0074] Specifically, step 3 includes: Step A: Perform independent signal enhancement and normalization processing on each modal signal, and use the normalization processing method to normalize the signal amplitude to the [-1, 1] interval; Step B: Extract the time-domain and frequency-domain features of each modality signal, calculate the time-domain features (such as RMSSD) and frequency-domain features (LF / HF) of heart rate (HR) and heart rate variability (HRV), extract the median frequency (MF) and root mean square (RMS) features of electromyography signals, and extract the respiratory rate (RF) and inspiratory-to-expiratory ratio (I:E Ratio) features. Step C: Use a weighted average method to fuse multimodal features, with weights set to 0.3:0.3:0.3:0.1:0.1; Step D: Use the Max-Min normalization method to normalize the fused features, normalizing the feature values ​​to the [0, 1] interval.

[0075] Step 4: Fatigue prediction and classification.

[0076] Specifically, step 4 includes: Step A: Input the normalized features into the pre-trained time series prediction model. The model uses an LSTM network structure. Step B: Use an LSTM network to perform deep mining analysis of the feature sequence. The input sequence length is 60 seconds, and the output is the fatigue risk probability for the next 30 seconds. Step C: Output the fatigue risk probability and classification results for the next 30 seconds. The classification results include cardiopulmonary dominant type, muscle dominant type, and mixed type. Step D: Select the appropriate recovery protocol library based on the classification results. For example, for cardiopulmonary dominant type, select the "heart rate regulation breathing method" recovery protocol.

[0077] Step 5: Restore boot execution.

[0078] Specifically, step 5 includes: Step a: Determine the main physiological driving factors of fatigue based on the prediction results. For example, in the cardiopulmonary dominant type, the main driving factors are abnormalities in heart rate and respiratory rate. Step b: Select the appropriate recovery protocol library; Step c: Generate a personalized recovery guidance plan, such as suggesting 10 minutes of heart rate regulation breathing exercises; Step d: Perform the recovery operation via voice guidance, using natural language processing technology to generate voice prompts.

[0079] Step 6: Recovery status assessment.

[0080] Specifically, step 6 includes: Step A: Continue to collect multimodal physiological signals for 10 minutes; Step B: Real-time processing and analysis of physiological signals during the recovery process, using real-time heart rate monitoring and electromyography signal analysis; Step C: Calculate the physiological recovery index during the recovery process, such as heart rate recovery rate and electromyographic signal recovery rate; Step D: Compare the recovery results with baseline data, such as physiological indicators at the initial state; Step E: Generate a recovery progress report, including trends in physiological indicators and recommendations.

[0081] Step 7: Determine if the recovery goal has been achieved. If not, return to step 2; otherwise, proceed to step 8.

[0082] Specifically, step 7 includes: Step a: Determine the recovery effect over three consecutive recovery cycles, for example, the recovery effect over three consecutive days; Step b: Adjust the recovery plan parameters based on the judgment results, such as adjusting the recovery time length; Step c: Update the individual user model in the cloud database, including physiological indicators and recovery results; Step d: Optimize the local model parameters and update the model using transfer learning techniques.

[0083] Step 8: Update system configuration and finish.

[0084] Specifically, step 8 includes: Step A: Update the local model according to the latest model version, with a version update frequency of once a week; Step B: Test and verify the new version with 100 test samples. Step C: Adjust system configuration parameters based on test results, such as adjusting feature extraction algorithm parameters; Step D: Update the help documentation and version history, recording the update time and changes.

[0085] In summary, this invention utilizes a multimodal wearable sensing terminal to simultaneously collect various physiological signals, including electrocardiogram (ECG), photoplethysmography (PPG), surface electromyography (EMG), inertial measurement unit (IMU), and bioacoustics. This constructs a comprehensive monitoring system covering multiple systems, including the cardiovascular, muscular, and respiratory systems. It overcomes the shortcomings of traditional single-parameter monitoring methods, which suffer from poor specificity and cannot fully reflect the user's physical condition, thus improving fatigue assessment accuracy to 85-90%. Furthermore, by employing time-series models such as LSTM and Transformer to deeply mine and analyze multimodal feature sequences, it can identify subtle changes in physiological parameters, detecting them before subjective fatigue symptoms appear and before abnormalities in traditional single indicators are detected. It accurately predicts the fatigue inflection point within 15-30 minutes, effectively solving the problem of lack of predictability in existing technologies. The warning time is 10-15 minutes earlier than traditional methods. Based on the prediction results, the system can intelligently distinguish between cardiopulmonary-dominant, muscle-dominant, or mixed fatigue types, and automatically generate differentiated recovery guidance plans based on individual user differences. This realizes the transformation from passive rest to active intervention, avoiding the simplistic treatment of "rest is stillness" in traditional methods. It improves recovery efficiency by 25-30%, and is particularly suitable for scenarios requiring precise load management, such as cardiopulmonary rehabilitation for the elderly. It significantly reduces exercise risks and improves training safety and effectiveness.

[0086] Secondly, this application provides a sports fatigue management system, which includes: a data acquisition module, a data processing module, an analysis module, a classification module, and a recovery guidance module; wherein, The system comprises the following modules: a data acquisition module for acquiring heart rate, electromyography (EMG), motion, and respiratory sound signals of the target object; a data processing module for calculating the modal features of the heart rate, EMG, motion, and respiratory sound signals; an analysis module for fusing all modal features using a weighted average method to form a fused feature; a classification module for predicting the type of fatigue risk of the target object based on the fused feature; and a recovery guidance module for outputting recovery suggestions based on the type of fatigue risk.

[0087] In some preferred embodiments, the recovery guidance module is also used to output voice and vibration reminders to the target object based on the output recovery suggestions.

[0088] The functions of each module in the above-mentioned sports fatigue management system correspond to the steps in the above-mentioned sports fatigue management method embodiment, and their functions and implementation processes will not be described in detail here.

[0089] Thirdly, embodiments of this application provide a sports fatigue management device, which can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0090] Reference Figure 3 , Figure 3 This is a schematic diagram of the hardware structure of the sports fatigue management device involved in the embodiments of this application. In the embodiments of this application, the sports fatigue management device may include a processor, a memory, a communication interface, and a communication bus.

[0091] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0092] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting internal components of the motion fatigue management device, as well as interfaces used for interconnecting the motion fatigue management device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0093] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0094] The processor can be a general-purpose processor, which can call the exercise fatigue management program stored in the memory and execute the exercise fatigue management method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the exercise fatigue management program is called can be referred to in the various embodiments of the exercise fatigue management method of this application, and will not be repeated here.

[0095] Those skilled in the art will understand that Figure 3 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0096] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0097] The present application provides a computer-readable storage medium storing a sports fatigue management program, wherein when the sports fatigue management program is executed by a processor, it implements the steps of the sports fatigue management method described above.

[0098] The method implemented when the exercise fatigue management procedure is executed can be referred to in the various embodiments of the exercise fatigue management method of this application, and will not be repeated here.

[0099] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0100] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0101] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0102] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0103] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0105] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for managing exercise fatigue, characterized in that, The exercise fatigue management method includes: Collect heart rate measurement signals, electromyography signals, movement signals, and respiratory sound signals of the target subject; Modal characteristics of heart rate measurement signal, electromyography signal, action signal, and breath sound signal were calculated respectively; A weighted average method is used to fuse all modal features to form a fused feature; Predict the types of fatigue risks for the target object based on fusion characteristics; Recovery recommendations are generated based on the type of fatigue risk.

2. The method for managing exercise fatigue as described in claim 1, characterized in that, The calculation of modal features of heart rate measurement signals, electromyography signals, action signals, and breath sound signals includes: Calculate the time-domain characteristics and frequency-domain characteristics of heart rate and heart rate variability based on the heart rate measurement signal; Calculate the median frequency characteristics and root mean square characteristics of electromyographic signals; Calculate respiratory rate characteristics and respiratory ratio characteristics based on respiratory sound signals.

3. The method for managing exercise fatigue as described in claim 1, characterized in that, The acquisition of the target object's heart rate measurement signal includes: acquiring the target object's electrocardiogram signal or photoplethysmography (PPG) pulse wave signal.

4. The method for managing exercise fatigue as described in claim 1, characterized in that, The method of predicting the type of fatigue risk of the target object based on fusion features includes: A time series model is used to analyze the fusion features in order to obtain the probability of fatigue risk within a preset future time period; The fatigue risk of the target object is classified according to the probability of fatigue risk.

5. The method for managing exercise fatigue as described in claim 4, characterized in that, The classification of fatigue risk of the target object based on fatigue risk probability includes: Based on the probability of fatigue risk, the target subjects are classified into one of the following types: lung-dominant, muscle-dominant, and mixed.

6. The method for managing exercise fatigue as described in claim 1, characterized in that, The method of fusing all modal features using a weighted average method to form a fused feature includes: Modal features are fused using a weighted average method based on preset weight coefficients; The fused features are normalized to form fused features.

7. The method for managing exercise fatigue as described in claim 1, characterized in that, The recovery recommendations output based on the type of fatigue risk include: The corresponding recovery protocol is extracted from the preset recovery protocol library based on the type of fatigue risk; The corresponding recovery suggestions are output based on the recovery protocol.

8. The method for managing exercise fatigue as described in claim 1, characterized in that, After outputting recovery suggestions based on the type of fatigue risk, the method also includes: Analyze the physiological signals of the target object during the recovery process and generate a recovery progress report; And adjust the recovery recommendations based on the recovery progress report.

9. A sports fatigue management system, characterized in that, include: The data acquisition module is used to acquire the target object's heart rate measurement signal, electromyography signal, movement signal, and respiratory sound signal; The data processing module is used to calculate the modal characteristics of heart rate measurement signals, electromyography signals, action signals, and respiratory sound signals, respectively. The analysis module is used to fuse all modal features using a weighted average method to form a fused feature; The classification module is used to predict the type of fatigue risk of a target object based on fusion features. The recovery guidance module is used to output recovery suggestions based on the type of fatigue risk.

10. The sports fatigue management system as described in claim 9, characterized in that: The recovery guidance module is also used to output voice and vibration reminders to the target object based on the output recovery suggestions.

Images