Methods and systems for sports training and risk assessment

Abstract

This invention relates to methods and systems for sports training and risk assessment, encompassing the fields of sports monitoring, wireless communication, and intelligent analysis. Its aim is to promptly identify and prevent sports risks during exercise. The methods include: real-time acquisition of sports data collected from the subject; acquisition of a parameter set corresponding to the type of exercise currently being performed by the subject; determination of the subject's respiratory-gait coupling stability based on respiratory and lower limb movement data, and determination of the subject's impact load index based on lower limb movement data; extraction of respiratory sound features from the subject's respiratory sound data and determination of respiratory sound risk features; and output of sports-related prompts based on the relationships between respiratory-gait coupling stability and its threshold, the impact load index and its threshold, and the respiratory sound risk features and their corresponding parameter thresholds.

Description

Technical Field

[0001] This disclosure relates to the fields of motion monitoring, wireless communication and intelligent analysis technology, and in particular to a method and system for motion training and risk assessment. Background Technology

[0002] With the development of national fitness and competitive sports, the scientific nature and safety of training processes are receiving increasing attention. Generally, trainees wear wearable devices to collect vital signs during exercise to prevent sports injuries.

[0003] However, existing wearable devices mostly focus on recording single vital signs such as heart rate and steps; however, relying solely on heart rate or steps makes it difficult to accurately assess sports injuries and also fails to provide feedback on the exercise being performed. Summary of the Invention

[0004] Firstly, a method for sports training and risk assessment is provided, the method comprising: Real-time acquisition of motion data collected from a moving object, the motion data including respiratory data and lower limb motion data, the respiratory data including at least one of respiratory rate, respiratory amplitude and / or inspiratory-to-expiratory ratio, the lower limb motion data including at least one of cadence, ground contact time, swing time and left-right symmetry parameters; Obtain a parameter set corresponding to the current movement type of the moving object. The movement type is used to characterize the current movement scenario of the moving object. Different movement types correspond to different parameter sets. The parameter set includes multiple parameter thresholds. The multiple parameter thresholds include at least a breathing-gait coupling stability threshold corresponding to the breathing data and an impact load threshold corresponding to the lower limb movement data. Based on the respiratory data and the lower limb movement data, the respiratory-gait coupling stability of the exercise object is determined, and based on the lower limb movement data, the impact load index of the exercise object is determined; wherein, the respiratory-gait coupling stability is used to characterize the relative relationship between breathing and gait; Based on the respiratory sound data, the respiratory sound features of the moving object are extracted, and the respiratory sound risk features are determined; Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, the relationship between the impact load index and the impact load threshold, and the relationship between the breathing sound risk characteristics and the corresponding parameter threshold, motion prompt information is output for the moving object; The exercise prompt information includes at least one of voice prompt information, text prompt information, beat sound and vibration beat, and the exercise prompt information is used to prompt the user to adjust the cadence and / or breathing rhythm.

[0005] In an exemplary embodiment, determining the respiratory-gait coupling stability of the moving object based on the respiratory data and the lower limb movement data includes: Based on the respiratory data within the sliding time window, determine the respiratory phase within the sliding time window; Based on the lower limb movement data within the sliding time window, determine the gait phase within the sliding time window; Based on the respiratory data and lower limb movement data within the sliding time window, parameter K is determined, wherein parameter K represents the integer ratio between breathing and gait. Based on the parameter K, the breathing phase, and the gait phase, determine the phase difference between the breathing phase and the gait phase within the sliding time window; The breathing-gait coupling stability is determined based on the phase difference corresponding to each of the sliding time windows.

[0006] In an exemplary embodiment, determining the impact load index of the exercise object based on the lower limb movement data includes: From the lower limb motion data, obtain the peak tibial / ankle acceleration and impact rise slope of the moving object at the moment of foot contact with the ground; The impact load index is determined based on the peak tibial / ankle acceleration, the slope of the impact rise, the ground contact time, and their respective weights.

[0007] In an exemplary embodiment, determining the impact load index based on the peak tibial / ankle acceleration, the impact rise slope, and the ground contact time, and their respective weights, includes: Obtain individual parameters of the moving object, wherein the individual parameters include at least one of the moving object's weight and walking speed; Based on the individual parameters, the peak tibial / ankle acceleration, the impact rise slope, and the ground contact time were normalized respectively. The impact load index is determined based on the tibial / ankle peak acceleration, impact rise slope, ground contact time, and their respective weights obtained after the normalization process.

[0008] The method described in the exemplary embodiments further includes: Acquire historical motion data collected within a preset time period. The historical motion data includes the cadence data and respiratory rhythm data of the moving object in a natural state. The respiratory rhythm data includes respiratory rate. Based on the gait frequency and respiratory rhythm data under natural conditions, the distributions of baseline gait frequency, baseline respiratory rate, and baseline respiratory-gait coupling stability are determined. Based on the baseline cadence, baseline respiratory rate, baseline respiratory-gait coupling stability distribution, and a preset forgetting factor, at least some of the parameter thresholds are updated; the forgetting factor is used to adjust the influence weight of the historical motion data on the parameter thresholds. Among them, the multiple parameter thresholds also include at least one of the following: cadence threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0009] In an exemplary embodiment, the method further includes: Based on the type of exercise, scene parameters corresponding to the type of exercise are obtained, and different types of exercise correspond to different scene parameters; wherein, the scene parameters include at least one of the following: load weight, type of equipment worn, altitude, slope, and training stage; Based on the scenario parameters, determine the adjustment weights corresponding to the multiple parameter thresholds in the parameter set; Based on the adjusted weights, the threshold values ​​of multiple parameters are updated.

[0010] In an exemplary embodiment, the step of outputting motion prompt information for the moving object based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, includes: Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the gait frequency in the lower limb motion data, a first motion prompt is output; wherein, the first motion prompt includes an inhalation / exhalation prompt synchronized with the gait frequency beat; Based on the relationship between the impact load index and the impact load threshold, a second motion prompt is output, which includes a correction prompt for any of the stride length, stride frequency, and landing mode.

[0011] In an exemplary embodiment, the method further includes: During the process of receiving the motion data, at least two of the following are detected: quality flag bit, packet loss rate, signal-to-noise ratio estimate, and artifacts corresponding to the motion data; wherein, the quality flag bit is carried in the motion data and is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference. A quality score is generated based on at least two of the quality flag, packet loss rate, signal-to-noise ratio estimation, and artifacts; the quality score is used to characterize at least one of the acquisition quality of the motion data, the transmission quality of the motion data, and the data quality of the motion data. Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, motion prompts are output for the moving object, including: Based on the relationship between the mass fraction, the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, the motion prompt information is output.

[0012] In an exemplary embodiment, the exercise type includes rehabilitation exercise types, and the plurality of parameter thresholds further include: cadence threshold range, inhalation / exhalation step ratio and inhalation / exhalation duration ratio range, gait symmetry threshold and ground contact time threshold; the method further includes: Based on multiple parameter thresholds, standard rhythm guidance information is output to the moving object; the standard rhythm guidance information includes gait frequency rhythm information and respiratory rhythm information, to instruct the moving object to move according to the gait frequency rhythm information and respiratory rhythm information; Based on the motion data of the moving object and the standard rhythm guidance information, determine the motion difference between the motion performed by the moving object and the standard motion corresponding to the standard rhythm guidance information; Based on the motion differences, at least some of the parameter thresholds are adjusted.

[0013] In an exemplary embodiment, the extracted respiratory sound features include at least one of the following: respiratory cycle sound energy distribution, dominant frequency band energy, frequency band ratio, respiratory phase transition features, continuity features, abnormal crackles, dry rales, wet rales, wheezing, and / or respiratory recovery time; the determination of respiratory sound risk features includes: Based on the extracted respiratory sound features, the ventilation load status, airway patency status, respiratory recovery status, and / or probability of abnormal respiratory events of the moving object are determined.

[0014] In an exemplary embodiment, the exercise prompting information includes at least one of rhythm prompting information, load adjustment prompting information, rehabilitation guidance information, and / or risk warning information; The load adjustment prompts and / or rehabilitation guidance information include at least one of the following: adjusting cadence, adjusting breathing rhythm, reducing training intensity, shortening training duration, extending recovery time, switching to recovery training mode, switching to rehabilitation guidance mode, and prompting to pause training.

[0015] In an exemplary embodiment, the method further includes: Based on the combined changing trends of the breathing-gait coupling stability, the impact load index, and the respiratory sound risk characteristics, the level of excessive fatigue risk and / or the level of sports injury risk are determined. When the risk level of excessive fatigue and / or the risk level of sports injury exceed a preset threshold, the load adjustment prompt information and / or the risk warning information are output.

[0016] Secondly, a sports training and risk assessment system is provided, the assessment system comprising a data acquisition module, a communication module, a data processing and intelligent analysis module, and a training feedback and risk alert module: wherein, The acquisition module includes multiple sensors, which are used to acquire motion data of the moving object. The communication module is connected to the acquisition module and configured to send the motion data to the data processing and intelligent analysis module, add a sensor channel identifier, sampling frequency, sequence number, timestamp, and quality flag to each frame of motion data, and when a disconnection from the data processing and intelligent analysis module is detected, store the motion data to be sent in a local ring buffer, and when the connection with the data processing and intelligent analysis module is restored, resend the missing motion data according to the sequence number and timestamp; wherein, the quality flag is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference; The data processing and intelligent analysis module is configured to perform the sports training and risk assessment method described in any exemplary embodiment of the first aspect; The training feedback and risk warning module is configured to send the exercise warning information to the terminal held by the exercise object.

[0017] The exercise training and risk assessment method provided in this embodiment can acquire exercise data collected from the exercise subject in real time, including respiratory data and lower limb movement data; and can acquire a parameter set corresponding to the type of exercise currently being performed by the exercise subject. The parameter set includes multiple parameter thresholds, including at least a respiratory-gait coupling stability threshold corresponding to the respiratory data and an impact load threshold corresponding to the lower limb movement data. Then, based on the respiratory data and lower limb movement data, the respiratory-gait coupling stability of the exercise subject can be determined, and based on the lower limb movement data, the impact load index of the exercise subject can be determined. Based on the relationship between the respiratory-gait coupling stability and the respiratory-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, exercise prompt information for the exercise subject is output. The exercise prompt information includes at least one of voice prompt information, text prompt information, beat sound, and vibration beat, and the exercise prompt information is used to prompt the user to adjust the cadence and / or breathing rhythm.

[0018] Because the exercise training and risk assessment method provided in this application acquires exercise data including respiratory data and lower limb movement data, it can assess the coordination of breathing and movement during exercise from both respiratory and lower limb movement dimensions. This yields the respiratory-gait coupling stability (e.g., assessing the matching degree between breathing and gait) and the impact load index. The respiratory-gait coupling stability and impact load index are then compared with corresponding thresholds in a parameter set, and exercise prompts are generated based on the comparison results. Therefore, this method not only integrates multimodal data such as breathing and lower limb gait to achieve a comprehensive assessment of training intensity and fatigue status, avoiding the limitations of single-data assessments in related technologies and improving assessment accuracy, but also provides guidance on appropriate movement frequency and breathing patterns for different exercise modes. This allows for exercise assessment and closed-loop feedback across different training subjects and rehabilitation stages, enabling timely correction of exercise and prevention of exercise-induced injuries.

[0019] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, specific embodiments of this disclosure are described below. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or related technologies of this disclosure, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the scale in the drawings is for illustration only and does not represent the actual scale.

[0021] Figure 1 This diagram illustrates the steps involved in sports training and risk assessment. Figure 2 A schematic diagram of the steps for determining respiratory-gait coupling stability is shown. Figure 3 A flowchart illustrating the steps for updating the parameter set in this embodiment is shown. Figure 4 A schematic diagram of the update steps for another parameter threshold is shown; Figure 5 A schematic diagram of the framework of a sports training and risk assessment system is shown; Figure 6 A schematic diagram showing the positional distribution of the acquisition modules on the moving object is shown. Detailed Implementation

[0022] To make the above-mentioned objectives, features, and advantages of this disclosure more apparent and understandable, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0023] In related technologies, wearable devices used in motion monitoring can only monitor single indicators such as heart rate and step count. While heart rate and step count monitoring technologies are relatively mature, they are insufficient to reflect information closely related to fatigue and injury risk, such as changes in respiratory efficiency, degeneration of movement rhythm, and asymmetry in load on the left and right limbs.

[0024] For example, in endurance sports such as running, marathons, and mountaineering, decreased gait symmetry, prolonged ground contact time, increased cadence fluctuations, and respiratory rhythm disorders often precede obvious heart rate abnormalities. In sports requiring high stability, such as shooting, respiratory phase and trunk micro-movements directly affect performance. In aquatic sports such as swimming, movement frequency and breathing rhythm need to be coordinated. Without comprehensive monitoring, trainees are prone to continuing to exercise under high load or fatigue, increasing the risk of overtraining and sports injuries.

[0025] In addition, existing motion monitoring has the following shortcomings: (1) the data from multiple sensors are not synchronized in time and the noise and packet loss processing is not perfect, resulting in unstable evaluation; (2) the analysis is mostly post-event statistics, lacking real-time early warning and executable training rhythm guidance; (3) there is a lack of individualized parameters and threshold adaptive mechanisms for different types of exercise; (4) the remote monitoring and guidance capabilities are insufficient, making it difficult to meet the needs of professional training and rehabilitation.

[0026] Furthermore, training for military and police units, special posts, and professional teams is often characterized by high intensity, long duration, heavy loads, and complex environments (such as high temperature, low temperature, and high altitude), and often adopts a collective or squad-based training management model. Training organizers need to objectively quantify and classify the real-time load, fatigue accumulation, and risk status of multiple trainees, while also meeting management requirements such as data security, access control, and training record keeping.

[0027] For example, in medical and rehabilitation settings, postoperative or chronic disease rehabilitation training emphasizes training prescriptions, standardized movements, and adherence monitoring. The training volume needs to be gradually increased within safe thresholds, and remote follow-up and dynamic adjustments should be conducted by doctors or rehabilitation therapists. Traditional rehabilitation assessments often rely on in-person observation or single tests, failing to provide continuous, real-time guidance and risk warnings in a home or community setting.

[0028] In view of this, this disclosure proposes a method and system for sports training and risk assessment, which can at least solve one of the problems described in the background art. The sports training and risk assessment method can collect respiratory data and lower limb movement data, thereby forming multimodal data. This allows for a more comprehensive assessment of the exercise process, moving beyond a single data set to include both movement and respiration.

[0029] Next, a parameter set corresponding to the type of exercise the subject is currently performing can be obtained. This parameter set may include an impact load threshold and a breathing-gait coupling stability threshold. After determining the breathing-gait coupling stability and impact load index based on the aforementioned breathing data and lower limb movement data, exercise prompts can be output to the subject based on the differences between the breathing-gait coupling stability threshold and the impact load index, as well as the differences between the impact load index and the impact load threshold. This allows for real-time execution of training rhythm guidance and real-time warnings, thus achieving real-time remote guidance and preventing trainees from continuing to exercise under high load or fatigue conditions.

[0030] As mentioned above, phenomena such as decreased gait symmetry, prolonged ground contact time, increased gait frequency fluctuations, and respiratory rhythm disorders often precede obvious heart rate abnormalities. By using the difference between the respiratory-gait coupling stability threshold and the respiratory-gait coupling stability, as well as the difference between the impact load index and the impact load threshold, this application enables the monitoring of both exercise and respiration, allowing for timely detection of abnormalities and avoiding the risks of overtraining and sports injuries.

[0031] The following description, in conjunction with the accompanying drawings, illustrates the sports training and risk assessment method and system of this disclosure.

[0032] Please combine Figure 1 As shown, Figure 1 A flowchart illustrating the steps of sports training and risk assessment methods is shown, such as... Figure 1 As shown, this sports training and risk assessment method includes the following steps: Step S100: Real-time acquisition of motion data collected from the moving object, including respiratory data and lower limb motion data.

[0033] The respiratory data includes at least one of respiratory rate, respiratory amplitude, and inspiratory-to-expiratory ratio, while the lower limb movement data includes at least one of cadence, ground contact time, swing time, and left-right symmetry parameters.

[0034] In this embodiment, the object of movement can be a person or other living organism.

[0035] In this embodiment, the motion data can be collected by multiple wearable sensors worn on the moving object.

[0036] For example, multiple wearable sensors may include a breathing signal sensor, a left lower limb inertial sensor, and a right lower limb inertial sensor.

[0037] Among them, the respiratory signal sensor is used to collect respiratory data, which may include at least one of respiratory rate, respiratory amplitude and inspiratory-to-expiratory ratio. For example, respiratory rate, respiratory amplitude or respiratory ratio may be collected.

[0038] Breathing amplitude is used to indicate the depth of breathing of a moving object; for example, it can indicate deep breathing and shallow breathing.

[0039] The breathing ratio can be the percentage of time spent exhaling versus the percentage of air intake.

[0040] Among them, the left lower limb inertial sensor and the right lower limb inertial sensor are used to collect lower limb motion data. The left lower limb inertial sensor is used to collect motion data of the left lower limb, and the right lower limb inertial sensor is used to collect motion data of the right lower limb.

[0041] The lower limb motion data may include multiple lower limb motion parameters, which may include at least one of the following: cadence, ground contact time, swing time, and left-right symmetry parameters.

[0042] Among them, step frequency can represent the frequency of lower limb movement on the ground, swing time can represent the time of one swing of the lower limb, and left-right symmetry parameter can represent the symmetry between the movement of the left and right lower limbs, also known as gait symmetry, which can include the symmetry parameter of the swing time of the left and right lower limbs, the symmetry parameter of the step frequency of the left and right lower limbs, and the symmetry parameter of the ground contact time of the left and right lower limbs.

[0043] The symmetry parameter can be represented by the absolute value of the difference between the left and right lower limbs in the same motion parameter. For example, the symmetry parameter of the gait frequency of the left and right lower limbs refers to the absolute value of the difference between the gait frequencies of the left and right lower limbs.

[0044] The symmetry parameters can be obtained by simple difference calculation after being collected by the left lower limb inertial sensor and the right lower limb inertial sensor.

[0045] In this embodiment, respiratory data can be used to reflect the respiratory state of the moving object during the movement process, and can promptly reflect whether the moving object has experienced respiratory rhythm disorder.

[0046] In this embodiment, lower limb motion data is used to reflect the gait symmetry, ground contact time, and gait frequency fluctuations of the moving object during the movement process. It can detect situations such as decreased gait symmetry, prolonged ground contact time, and increased gait frequency fluctuations that occur during the movement of the moving object.

[0047] Therefore, the respiratory data and lower limb movement data collected in this embodiment can be called multimodal data, which improves the richness of the collected data and avoids the problem of single data collection.

[0048] In other examples, respiratory data may also include respiratory sound data, which can be acoustic data such as sound frequency data. Analysis of this data yields a comprehensive dataset containing multiple dimensions, including sound type, acoustic characteristics, spatiotemporal attributes, and audio phase changes. This respiratory sound data can be used to assess the lung condition of an individual during exercise, thereby enabling the timely detection of lung health problems. It can also help monitor potential risks in exercise rehabilitation training for patients with existing lung conditions.

[0049] Accordingly, this example may also include sensors for collecting respiratory sound data, such as smart electronic stethoscopes and throat-specific microphones, wearable patches, etc. The wearable patch can be gently attached to the skin and wirelessly and continuously track respiratory sounds simultaneously at multiple locations on the body. It may also be equipped with a miniature microphone to separate internal and external sounds, enabling dynamic lung health assessment. Furthermore, the wearable patch can be integrated with devices such as ECG and blood oxygen monitoring systems, allowing multiple sensors to work collaboratively.

[0050] In some other examples, the collected motion data may also include heart rate data, so the wearable sensor may also include a heart rate sensor.

[0051] Heart rate data can include both exercise heart rate and resting heart rate, which can help detect situations where the heart rate may be too fast during exercise.

[0052] Of course, in some other examples, the collected motion data may also include blood oxygen saturation data, and the wearable sensor may also include a blood oxygen saturation sensor, wherein the blood oxygen saturation data is used to reflect the blood oxygen saturation of the moving object during the exercise process, so as to detect hypoxia in a timely manner.

[0053] The heart rate sensor, respiration sensor, and blood oxygen saturation sensor mentioned above can be integrated into the same device.

[0054] It should be noted that the above motion data can be collected in real time. Real time means that the data is collected at very short intervals, such as once every 10ms.

[0055] The lower limb movement data and respiratory data were collected at the same frequency and were synchronized, meaning that the difference between their collection times was very small, such as less than 1 microsecond, to ensure strict synchronization.

[0056] Step S200: Obtain the parameter set corresponding to the current motion type of the moving object. The motion type is used to characterize the current motion scene of the moving object. Different types of exercise correspond to different parameter sets, which include multiple parameter thresholds. These multiple parameter thresholds include at least the respiratory-gait coupling stability threshold corresponding to respiratory data and the impact load threshold corresponding to lower limb movement data.

[0057] In one example given, the parameter set includes at least one set of the following fields: (1) Target action frequency range F_step_target: including the lower limit F_min and the upper limit F_max of the target step frequency, and the maximum step ΔF_max for each rhythm adjustment; (2) Breathing mode target set R_br: including the set of inhalation-exhalation step ratios (such as 2:2, 2:3, 3:2, etc.) and / or the interval of inhalation-exhalation duration ratio; (3) Coupling stability target C_target and allowable deviation: including target interval [C_low, C_high] or target value and minimum stability threshold; (4) Impact load threshold L_th: including reminder threshold L1, warning threshold L2 and corresponding action correction strategies (such as increasing step frequency, reducing stride length, increasing landing cushioning, etc.). (5) Gait symmetry threshold A_th and ground contact time threshold CT_th: used to identify the compensation and fatigue stages and trigger the "stop / recovery / technical adjustment" prompt; (6) Safety constraint thresholds: including blood oxygen saturation, body surface temperature, etc., for safety gate thresholds used in extreme training and rehabilitation phases of military and police personnel (optional); (7) Output strategy: including prompting format (voice / beat / vibration), intervention maintain Re-intervention strategy, cooldown time, and hysteresis coefficient, etc.

[0058] In this embodiment, different parameter sets can be configured for different types of sports; that is, multiple sports types can be configured with their own independent parameter sets.

[0059] Among these, the types of parameter thresholds in the parameter sets corresponding to different motion types can be different, and the values ​​of the same parameter threshold can be different in the parameter sets corresponding to different motion types.

[0060] The parameter set may include multiple parameter thresholds, such as the respiratory-gait coupling stability threshold corresponding to respiratory data and the impact load threshold corresponding to lower limb movement data. Different parameter sets may have different respiratory-gait coupling stability thresholds and different parameter sets may have different impact load thresholds.

[0061] As mentioned earlier, when respiratory data includes breath sound data, the parameter set can also include parameter thresholds corresponding to the breath sound data. For example, the parameter set includes parameter thresholds corresponding to each of the following: dry rales, wet rales, pleural friction rubs, vesicular breath sounds, bronchial breath sounds, and bronchovesicular breath sounds. When making comparisons, the intensity values ​​of the real-time detected dry rales, wet rales, pleural friction rubs, vesicular breath sounds, bronchial breath sounds, and bronchovesicular breath sounds can be compared with the parameter thresholds in the parameter set, respectively. The combined impact load threshold and the respiratory-gait coupling stability threshold are used to jointly determine the output motion prompt information.

[0062] In this embodiment, the types of exercise may include running, marathon, mountain climbing, military and police training, special post and professional team training, and exercise with medical rehabilitation as the main purpose.

[0063] Among them, the training of military and police personnel, special posts and professional teams is often characterized by high intensity, long duration, heavy load, and complex environment (such as high temperature, low temperature, high altitude). Running, marathon and mountain climbing are endurance sports. Therefore, the impact load threshold of the parameter set corresponding to the training of military and police personnel, special posts and professional teams can be greater than the impact load threshold of the parameter set corresponding to running, marathon and mountain climbing.

[0064] Of course, the above is only an example. The specific impact load threshold and breathing-gait coupling stability threshold can be set according to the type of exercise.

[0065] In some embodiments, an independent set of parameters can be set for each moving object. For the same type of movement, the impact load threshold and the breathing-gait coupling stability threshold of different moving objects may be different.

[0066] For example, there are two moving objects, A and B. Moving object A corresponds to parameter set A and moving object B corresponds to parameter set B. The impact load threshold in parameter set A may be different from the impact load threshold in parameter set B, and the breathing-gait coupling stability threshold in parameter set A may be different from the breathing-gait coupling stability threshold in parameter set B.

[0067] This allows for customized parameter sets to be implemented for each moving object, enabling personalized motion monitoring.

[0068] In this embodiment, the impact load threshold can characterize the lower limb load during exercise. The larger the impact load threshold, the greater the load that the lower limbs of the exerciser can bear.

[0069] In this embodiment, the respiratory-gait coupling stability threshold can characterize the coordination stability of respiratory rhythm and gait rhythm within a sliding time window, which can be characterized by the phase difference distribution between the respiratory phase and the gait phase.

[0070] Step S300: Based on respiratory data and lower limb movement data, determine the respiratory-gait coupling stability of the exercise object, and based on lower limb movement data, determine the impact load index of the exercise object; Among them, the respiratory-gait coupling stability is used to characterize the relative relationship between breathing and gait. The higher the respiratory-gait coupling stability, the more stable and coordinated the respiratory rhythm and gait rhythm are.

[0071] In this embodiment, the stability of the breathing-gait coupling can be characterized by the phase difference distribution between the breathing phase and the gait phase. For example, the breathing phase can be determined based on the breathing data, and the gait phase can be determined based on the lower limb movement data. The difference between the breathing phase and the gait phase within the sliding time window is taken as the breathing-gait coupling stability within that time.

[0072] The respiratory phase can be determined based on respiratory rate and respiratory amplitude, and the respiratory phase can reflect the characteristics of breathing within a sliding time window.

[0073] Among them, when the gait phase can be determined based on the cadence, ground contact time, swing time, and left-right symmetry parameters, the gait phase can reflect the characteristics of the lower limb movement of the moving object.

[0074] By subtracting the gait phase and respiratory phase within a sliding time window, the coordination between breathing and lower limb movement can be statistically determined. For example, if the distribution of gait phase and respiratory phase within the sliding time window is relatively regular, it can be determined that there is coordination between breathing and lower limb movement. On the other hand, if the distribution of gait phase and respiratory phase within the sliding time window is relatively chaotic, it can be determined that there is poor coordination between breathing and lower limb movement, indicating that the breathing and movement of the exercise subject are disordered during the exercise, and there may be a risk of sports injury.

[0075] The sliding time window can be set according to requirements and can represent a continuous time period.

[0076] The sliding time window can be longer than the time interval between the collection of respiratory data and lower limb movement data.

[0077] Step S400: Based on the respiratory sound data, extract the respiratory sound features of the moving object and determine the respiratory sound risk features; In this embodiment, spectral analysis can be performed on the respiratory sound data, such as performing a fast Fourier transform on the respiratory sound data to obtain its spectrum. After analyzing the spectrum, respiratory sound risk features can be extracted, such as respiratory cycle sound energy distribution, main frequency band energy, frequency band ratio, respiratory phase transition features, continuity features, abnormal crackles, dry rales, wet rales, wheezing, and / or respiratory recovery time.

[0078] For example, respiratory sound data can be preprocessed, denoised, segmented, and feature extracted to obtain the acoustic energy distribution of the respiratory cycle, the energy of the main frequency band, the frequency band ratio, the difference in acoustic characteristics between the inspiratory and expiratory phases, the frequency of abnormal respiratory sound events, and the respiratory sound recovery time during the post-exercise recovery phase.

[0079] Step S500: Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, the relationship between the impact load index and the impact load threshold, and the relationship between the breathing sound risk characteristics and the corresponding parameter threshold, output motion prompt information for the moving object; The exercise prompts include at least one of voice prompts, text prompts, beat sounds, and vibration beats, and are used to prompt users to adjust their cadence and / or breathing rhythm.

[0080] In this embodiment, the breathing-gait coupling stability can be compared with the breathing-gait coupling stability threshold in the parameter set to obtain the comparison result of breathing-gait. For example, the difference between the breathing-gait coupling stability and the breathing-gait coupling stability threshold can be used as the comparison result of breathing-gait.

[0081] In this embodiment, the impact load index and the impact load threshold can be compared to obtain the comparison result of the impact load index. For example, the difference between the impact load index and the impact load threshold can be used as the comparison result of the impact load index.

[0082] Specifically, when the breathing-gait coupling stability is greater than the breathing-gait coupling stability threshold, the degree of incoordination between breathing and gait increases; when the breathing-gait coupling stability is less than or equal to the breathing-gait coupling stability threshold, the degree of coordination between breathing and gait increases.

[0083] Specifically, when the impact load index is greater than the impact load threshold, it indicates that the lower limb load exceeds the expected load; when the impact load index is less than or equal to the impact load threshold, it indicates that the lower limb load does not exceed the expected load.

[0084] When outputting motion cues based on the comparison results of breathing-gait and impact load index, the following scenarios are possible: Scenario 1: If the breathing-gait coupling stability is greater than the breathing-gait coupling stability threshold, regardless of whether the impact load index is greater than the impact load threshold, the output motion prompt information can be: adjusting the breathing rhythm, such as outputting at least one of the beat sound and vibration beat, or outputting voice prompt information to prompt the breathing rhythm.

[0085] Scenario 2: If the respiratory-gait coupling stability is less than the respiratory-gait coupling stability threshold and the impact load index is greater than the impact load threshold, the output prompt information can be voice prompt information and text prompt information, which can prompt the user to reduce the range of motion of the lower limbs.

[0086] Scenario 3: If the breathing-gait coupling stability is less than the breathing-gait coupling stability threshold and the impact load index is greater than the impact load threshold, the output prompt information can be at least one of a beat sound and a vibration beat. The frequency of the output beat sound and vibration beat can be greater than the step frequency of the current lower limb movement of the moving object, in order to prompt the user to increase the step frequency and reduce the ground contact time, so as to reduce the impact load index.

[0087] In some examples, when the impact load index or contact time continues to rise, the frequency of the beat and vibration beat can be increased again, first by increasing the step frequency with small steps to reduce the stride and impact.

[0088] In some examples, when a continuous increase in the impact load index is detected, a prompt message can be output to slow down or change road sections.

[0089] In some examples, when an increase in high-frequency abnormal components in breath sounds, unstable respiratory phase transitions, an increase in abnormal breath sound events, or prolonged respiratory recovery time is detected, it can be determined that the exercise subject has an increased ventilatory load, increased respiratory muscle fatigue, insufficient recovery, or a trend towards movement rhythm imbalance. At this time, the system can output prompts such as decreased cadence, prolonged expiration, reduced training load, prolonged recovery time, or alternative rehabilitation movements.

[0090] In some examples, prompts can be output based on the first comparison between the extracted breath sound features and the corresponding parameter thresholds in the parameter set, the second comparison between the shock load index and the shock load threshold, and the third comparison between the breath-gait coupling stability and the breath-gait coupling stability threshold.

[0091] For example, a health tag can be set for the exercise object. The health tag is used to identify the physical condition type of the exercise object. For example, if the exercise object is a professional athlete, its health tag can be 1. When outputting the prompt information, the second comparison result and the third comparison result shall prevail. If the exercise object is an ordinary person without lung disease, its health tag can be 2. When outputting the prompt information, the second comparison result and the third comparison result shall prevail. After the exercise time exceeds the preset time, the prompt information shall be output based on the first comparison result, the second comparison result and the third comparison result.

[0092] If the exercise subject is a person with lung disease, whose health label can be 3, then when outputting prompts, the first comparison result should be used first, followed by the second and third comparison results. Alternatively, the first and second comparison results can be used as the basis for outputting prompts. For example, if the first comparison result indicates that the exercise subject has strong dry rales or wet rales, the prompt should be to stop the exercise immediately. If the first comparison result indicates that the exercise subject has weak dry rales or wet rales, the prompt can be output in conjunction with the second comparison result. If the second comparison result indicates that breathing is coordinated, the monitoring can continue. If the second comparison result indicates that breathing is uncoordinated, breathing exercise prompts can be output to guide the exercise subject's breathing and relieve lung pressure.

[0093] In this embodiment, respiratory data and lower limb movement data can be sent to a server. The server has a built-in parameter set and can calculate a first comparison result, a second comparison result, and a third comparison result. The server then sends the first comparison result, the second comparison result, and the third comparison result to a decision module corresponding to the health identifier of the exercise object. Different health identifiers correspond to different decision modules. The decision module outputs exercise prompt information based on the corresponding strategy and feeds it back to the terminal held by the exercise object, thereby realizing closed-loop feedback of exercise.

[0094] The exercise training and risk assessment method of this embodiment, since the exercise data includes respiratory data and lower limb movement data, can assess the coordination of breathing and movement during exercise from both respiratory and lower limb movement dimensions. This yields the respiratory-gait coupling stability (e.g., assessing the matching degree between breathing and gait) and the impact load index. The respiratory-gait coupling stability and impact load index are then compared with corresponding thresholds in a parameter set, and exercise prompts are generated based on the comparison results. Therefore, this method not only integrates multimodal data such as breathing and lower limb gait to achieve a comprehensive assessment of training intensity and fatigue status, avoiding the limitations of single-data assessments in related technologies and improving assessment accuracy, but also provides guidance on movement frequency and breathing patterns for different exercise modes. This allows for exercise assessment and closed-loop feedback across different training subjects and rehabilitation stages, enabling timely correction of movement and reducing the risk of exercise-induced injuries.

[0095] In some embodiments, the exercise prompting information includes at least one of rhythm prompting information, load adjustment prompting information, rehabilitation guidance information, and / or risk warning information.

[0096] In some examples of this embodiment, the load adjustment prompts and / or rehabilitation guidance information include at least one of the following: adjusting cadence, adjusting breathing rhythm, reducing training intensity, shortening training duration, extending recovery time, switching to recovery training mode, switching to rehabilitation guidance mode, and prompting to pause training.

[0097] Specifically, it can be determined by comprehensively considering the aforementioned first comparison result, second comparison result, and third comparison result.

[0098] In some examples of this embodiment, the risk warning information includes the results of potential disease risk screening. The results of potential disease risk screening can be determined based on the risk characteristics of respiratory sounds, changes in respiratory rate, respiratory recovery time, changes in blood oxygen saturation, trends in heart rate changes, and / or changes in exercise tolerance. These results are used to indicate that the exercise subject has abnormal risks in the respiratory system and / or abnormal risks in the circulatory system, thereby timely monitoring of the health risks that occur to the exercise subject during exercise.

[0099] In some examples of this embodiment, the risk level of over-fatigue and / or the risk level of sports injury can also be determined based on the combined changing trend of the breathing-gait coupling stability, the impact load index, and the risk characteristics of the breathing sounds; when the risk level of over-fatigue and / or the risk level of sports injury exceeds a preset threshold, the load adjustment prompt information and / or the risk warning information are output.

[0100] In this embodiment, the combined changing trends of the breathing-gait coupling stability, the impact load index, and the respiratory sound risk characteristics can reflect the frequency and magnitude of change of each of the breathing-gait coupling stability, the impact load index, and the respiratory sound risk characteristics within a time window. Thus, based on the frequency and magnitude of change of each of the breathing-gait coupling stability, the impact load index, and the respiratory sound risk characteristics within a time window, a profile of the health status of the exercise object during exercise can be created, thereby depicting the coupling state of lung respiration, gait, and lower limb load during exercise, and further determining at least one of the excessive fatigue risk level and the sports injury risk level.

[0101] Specifically, when the risk level of excessive fatigue or sports injury exceeds a preset threshold, or both exceed the preset threshold, load adjustment prompts and / or risk warnings will be output.

[0102] For example, if the risk level of excessive fatigue exceeds a preset threshold, a load adjustment prompt will be output; if the risk level of sports injury exceeds a preset threshold, a risk warning will be output. If both exceed their respective preset thresholds, both a load adjustment prompt and a risk warning will be output.

[0103] In some embodiments, the extracted breath sound features include at least one of the following: respiratory cycle sound energy distribution, dominant frequency band energy, frequency band ratio, respiratory phase transition features, continuity features, abnormal crackles, dry rales, wet rales, wheezing, and / or respiratory recovery time; then, when determining breath sound risk features, at least one of the following breath sound risk features can be determined based on the extracted breath sound features: The ventilation load status, airway patency status, respiratory recovery status, and / or probability of abnormal respiratory events of the exercised object.

[0104] Ventilation load refers to the total resistance or work required by the respiratory muscles to maintain effective ventilation. It comprehensively reflects the mechanical characteristics of the respiratory system and is the key to assessing whether breathing is strenuous.

[0105] Among them, the airway patency represents the "unobstructedness of the passageway" for air to enter and exit. It specifically refers to the degree to which the entire respiratory tract from the nasal cavity to the alveoli remains open and free from physical obstruction. It is the anatomical and functional basis for ensuring that air can enter and exit freely. Among them, the respiratory recovery state represents a "positive signal" of improved lung function. The "respiratory recovery state" describes the dynamic process of respiratory function returning to normal after treatment of lung diseases or during the body's self-healing process. It is a comprehensive manifestation of a series of positive physiological changes.

[0106] Based on the above-mentioned ventilation load status, airway patency status, respiratory recovery status, and / or probability of abnormal respiratory events, the lung status of the exercise subject during exercise can be assessed.

[0107] Accordingly, the parameter set includes parameter thresholds corresponding to ventilation load status, airway patency status, respiratory recovery status, and / or the probability of abnormal respiratory events.

[0108] In some embodiments, please refer to Figure 2 As shown, Figure 2 A schematic diagram of the steps for determining respiratory-gait coupling stability is shown, as follows: Figure 2 As shown, determining the stability of the breathing-gait coupling may include the following steps: Step S301: Determine the respiratory phase within the sliding time window based on the respiratory data within the sliding time window; Step S302: Determine the gait phase within the sliding time window based on the lower limb motion data within the sliding time window; Step S303: Based on the respiratory data and lower limb movement data within the sliding time window, determine the parameter K, which represents the integer ratio between breathing and gait; Step S304: Based on parameter K, respiratory phase, and gait phase, determine the phase difference between the respiratory phase and gait phase within the sliding time window; Step S305: Determine the breathing-gait coupling stability based on the phase difference corresponding to each sliding time window.

[0109] In this embodiment, the respiratory phase φ_breath(t) within the sliding time window can be determined based on the respiratory rate, respiratory amplitude, and respiratory ratio within the sliding time window.

[0110] In this embodiment, the gait phase φ_step(t) within the sliding time window can be determined based on the step frequency, ground contact time, swing time, and left-right symmetry parameters within the sliding time window.

[0111] Next, parameter K can be determined based on respiratory data and lower limb movement data within the sliding time window, where K is an integer ratio between breathing and gait. In one example, it could be the ratio between respiratory frequency and gait frequency within the sliding time window.

[0112] In this embodiment, the phase difference within the sliding time window can be determined according to the following formula (1): Δφ(t) = φ_breath(t) k·φ_step(t) formula (1).

[0113] Next, the breathing-gait coupling stability can be determined based on the phase difference corresponding to each sliding time window. For example, the mean of the phase difference corresponding to each sliding time window can be used to determine the breathing-gait coupling stability. Alternatively, in some examples, the mean square error of the phase difference corresponding to each sliding time window can be used to determine the breathing-gait coupling stability.

[0114] In some embodiments, when determining the impact load index, the peak tibial / ankle acceleration and the impact rise slope at the moment of foot contact with the ground can be obtained from the lower limb motion data; and the impact load index can be determined based on the peak tibial / ankle acceleration, the impact rise slope, the contact time, and their respective weights.

[0115] In this embodiment, the lower limb motion data can be interpreted to obtain the peak tibial / ankle acceleration at the moment of foot contact with the ground. For example, the left and right lower limb inertial sensors can continuously collect the peak tibial / ankle acceleration. Then, based on the step frequency, ground contact time, and swing time within the continuous time, the impact rise slope can be determined. The impact rise slope can represent a proxy characterizing the vertical loading rate AVLR.

[0116] Next, the impact load index can be determined based on the peak tibial / ankle acceleration, the slope of the impact rise, the ground contact time, and their respective weights.

[0117] For example, the impact load index L can be calculated according to the following formula (2): L=w1·AVLR_norm+w2·PTA_norm+w3·CT_norm formula (2); In formula (2), AVLR_norm represents the impact rise slope, PTA_norm represents the tibial / ankle peak acceleration, and CT_norm represents the ground contact time; w1, w2, and w3 represent the weights corresponding to the impact rise slope, tibial / ankle peak acceleration, and ground contact time, respectively.

[0118] In some examples, the above-mentioned impact load index L can also be determined by combining the left and right symmetry parameters, as shown in the following formula (3): L=w1·AVLR_norm+w2·PTA_norm+w3·CT_norm+w4·ASYM_norm formula (3); Where ASYM_norm represents the left and right symmetry parameters, and w4 represents the weights corresponding to the left and right symmetry parameters.

[0119] In this embodiment, the weights corresponding to the peak tibial / ankle acceleration, the slope of the impact rise, and the ground contact time can be values ​​between 0 and 1. The weights can be set according to requirements, and are not limited here.

[0120] In some examples, neural network models can be used to determine the corresponding weights, so that the weights of the tibial / ankle peak acceleration, the slope of the impact rise, and the ground contact time can be matched with the movement habits and physical condition of the athlete.

[0121] For example, training data can be collected, which may include the actual impact load index (label value) of the moving object, as well as the peak tibial / ankle acceleration, impact rise slope, and ground contact time collected by sensors during the moving object's movement. Next, the peak tibial / ankle acceleration, impact rise slope, and ground contact time from the training data can be input into a neural network. The neural network predicts the corresponding impact load index (predicted value). Based on the difference between the predicted value and the label value, the parameters of the neural network are updated, thus obtaining the weight values ​​configured for the peak tibial / ankle acceleration, impact rise slope, and ground contact time in the neural network. These weight values ​​are then used as the weights for the moving object, thereby achieving individualized solutions for the moving object, improving computational accuracy, and avoiding inaccurate calculations caused by individual differences between moving objects.

[0122] In this embodiment, since the impact load index is determined based on the peak acceleration of the tibia / ankle, the slope of the impact rise, and the ground contact time, multiple parameters such as the acceleration of lower limb movement, vertical loading rate, and cadence can be comprehensively considered to determine the impact load index from aspects such as the speed and amplitude of lower limb movement during exercise, thereby improving the accuracy of the impact load index.

[0123] In one example of this embodiment, the peak acceleration of the tibia / ankle, the slope of the impact rise, and the ground contact time, and in some examples, the left-right symmetry parameter, can be normalized and then the impact load index can be solved according to the above formula (2) or formula (3).

[0124] For example, individual parameters of the moving object can be obtained, wherein the individual parameters include at least one of the moving object's weight and walking speed; then, based on the individual parameters, the tibial / ankle peak acceleration, impact rise slope and ground contact time can be normalized respectively. Next, the impact load index can be determined based on the tibial / ankle peak acceleration, impact rise slope, ground contact time, and their respective weights obtained after normalization.

[0125] In this embodiment, the individual parameter can be body weight, which can be used to characterize the baseline of impact load. The individual parameter can also be walking speed, which can be used to characterize walking speed.

[0126] In this embodiment, the peak tibial / ankle acceleration, impact rise slope, and ground contact time can be divided by individual parameters to achieve normalization of these parameters.

[0127] After normalization, the above parameters are normalized to values ​​between 0 and 1.

[0128] By adopting the implementation scheme of this embodiment, the normalization process can reduce the amount of calculation required for the impact load index in the later stage. On the other hand, since individual parameters can be used as the basis for normalization, the influence of individual characteristics such as weight and walking speed can be eliminated, making the calculation of the impact load index more accurate and objective in reflecting the movement of the moving object.

[0129] In some embodiments, as described above, the parameter set can be closely associated with both the moving object and the type of motion; that is, each moving object can have its own independent parameter set. In practice, as the moving object progresses through motion, the parameters in the parameter set can be updated, for example, updating the breathing-gait coupling stability threshold and the impact load threshold.

[0130] Please refer to Figure 3 As shown, Figure 3 This embodiment illustrates a flowchart of the steps for updating the parameter set, as shown below. Figure 3 As shown, the following steps may be included: Step S501: Acquire historical motion data collected within a preset time period; The historical motion data includes the cadence data and respiratory rhythm data of the moving object in its natural state, and the respiratory rhythm data includes respiratory rate; Step S502: Based on gait frequency data and respiratory rhythm data under natural conditions, determine the distribution of baseline gait frequency, baseline respiratory rate, and baseline respiratory-gait coupling stability. Step S503: Based on the baseline cadence, baseline respiratory rate, baseline respiratory-gait coupling stability distribution, and a preset forgetting factor, update at least some parameter thresholds; the forgetting factor is used to adjust the weight of the influence of historical motion data on the parameter thresholds; Among them, several parameter thresholds include: cadence threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0131] In this embodiment, historical motion data can be motion data collected before the current motion of the moving object. Furthermore, it can be data collected when the moving object is in its natural state, that is, data collected when it is not in motion.

[0132] Historical exercise data may include cadence data and respiratory rhythm data. Respiratory rhythm data includes respiratory rate, and cadence data includes cadence.

[0133] In this embodiment, the baseline cadence can be determined based on cadence data, and the baseline respiratory rate can be determined based on respiratory rhythm data. The baseline cadence can be the average cadence across multiple historical time periods in the historical exercise data; the baseline respiratory rate can be the average respiratory rate across multiple historical time periods in the historical exercise data.

[0134] In this embodiment, the distribution of basic breathing-gait coupling stability can be determined based on gait frequency data and respiratory rhythm data. That is, the calculation process of basic breathing-gait coupling stability can refer to the calculation process of the above formula (1), which will not be elaborated here.

[0135] In this embodiment, at least some parameter thresholds can be updated based on the baseline cadence, baseline respiratory rate, baseline respiratory-gait coupling stability distribution, and a preset forgetting factor.

[0136] In this embodiment, the parameter thresholds in the parameter set may include, in addition to the breathing-gait coupling stability threshold and the impact load threshold, at least one of the following: gait frequency threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0137] For example, the breathing-gait coupling stability threshold can be updated based on the basic breathing-gait coupling stability distribution and a preset forgetting factor. For instance, a new breathing-gait coupling stability threshold can be determined based on the basic breathing-gait coupling stability distribution, and this new threshold can be updated in the parameter set. When calculating the new breathing-gait coupling stability threshold, the basic breathing-gait coupling stability can be increased with a certain weight, and the basic breathing-gait coupling stability distribution can be transformed into a motion scenario to obtain the breathing-gait coupling stability distribution that the moving object should have during motion. This breathing-gait coupling stability distribution can then be used as the new breathing-gait coupling stability threshold.

[0138] The breathing-gait coupling stability threshold includes the target interval [C_low, C_high], or the target value and the minimum stability threshold.

[0139] For example, the parameter set may also include cadence threshold ranges, inhalation / exhalation step ratio and inhalation / exhalation duration ratio ranges, gait symmetry thresholds and ground contact time thresholds, etc.

[0140] Specifically, the step frequency threshold range in the parameter set can be updated based on the step frequency threshold range. The step frequency threshold range includes the target step frequency lower limit F_min and upper limit F_max, as well as the maximum step size ΔF_max for each rhythm adjustment.

[0141] Specifically, the parameters for the inspiratory-to-expiratory step ratio and the inspiratory-to-expiratory duration ratio can be updated based on the baseline respiratory rate. The inspiratory-to-expiratory step ratio can be 2:2, 2:3, or 3:2. This ratio can be understood as the number of steps taken during inhalation versus exhalation, reflecting the correlation between respiration and gait. The inspiratory-to-expiratory duration ratio range refers to the ratio between the duration of inhalation and the duration of exhalation, reflecting the respiratory rhythm.

[0142] Specifically, the baseline ranges for the inspiratory-to-expiratory step ratio and the inspiratory-to-expiratory duration ratio can be determined based on the baseline respiratory rate. That is, when initially constructing the parameter set, the ranges for these ratios can be determined based on the baseline respiratory rate over a historical period. Subsequently, these ranges can be periodically updated based on the new baseline respiratory rate.

[0143] Specifically, the gait symmetry threshold and ground contact time threshold in the parameter set can be updated based on the baseline gait frequency. In this embodiment, the gait symmetry threshold A_th and the ground contact time threshold CT_th are used to identify the compensation and fatigue stages and trigger a "stop / recovery / technical adjustment" prompt.

[0144] In some examples, the parameter thresholds in the parameter set can also include constraint thresholds, such as blood oxygen saturation and body surface temperature, to prevent exercise from exceeding limits and causing health risks. Thus, when constructing blood oxygen saturation and body surface temperature, historical exercise data can also include the blood oxygen saturation and body surface temperature of the exercise object under natural conditions.

[0145] Using the technical solution of this embodiment, as the exercise progresses, the physical qualities of the exerciser change. As a result, their natural cadence, breathing rate, and other parameters will change. Although this change is slight, it can still be collected by the sensor. Therefore, the parameters in the parameter set can be updated periodically as the exerciser progresses, so that the parameter thresholds in the parameter set can be adapted to the current physical state of the exerciser, thereby providing accurate exercise guidance for the exerciser.

[0146] In some embodiments, the parameter thresholds in the parameter set on which the current motion depends can be updated according to the motion scene in which the moving object is located, so that the parameter thresholds in the parameter set can match the motion scene in which the moving object is currently located.

[0147] Among them, the sports scene includes the sports environment and the sports equipment worn by the sports object.

[0148] For example, scene parameters corresponding to the motion type can be obtained based on the motion type, and adjustment weights corresponding to multiple parameter thresholds in the parameter set can be determined based on the scene parameters; and the multiple parameter thresholds can be updated based on the adjustment weights.

[0149] In this example, different types of exercise correspond to different scenario parameters; among them, scenario parameters include at least one of the following: load weight, type of equipment worn, altitude, slope, and training stage.

[0150] Among the scene parameters corresponding to the type of exercise, the load weight, type of equipment worn, altitude, slope, and training stage are all related to the current exercise scene. In practice, the load weight, type of equipment worn, altitude, and slope can be used as scene parameters to participate in the updating of multiple parameter thresholds.

[0151] In this embodiment, the multiple parameter thresholds also include at least one of the following: cadence threshold range, inhalation / exhalation step ratio and inhalation / exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0152] Among them, the adjustment weights corresponding to the cadence threshold range, the inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, the gait symmetry threshold and the ground contact time threshold can be calculated based on the load weight, the type of equipment worn, the altitude, and the slope.

[0153] For example, the greater the load, the lower the cadence threshold range and the higher the inhalation-exhalation duration ratio range need to be. The adjustment weight corresponding to the cadence threshold range is larger, while the adjustment weight corresponding to the inhalation-exhalation duration ratio range is smaller.

[0154] In some examples, load weight, equipment type, altitude, and slope can be normalized to the same space to calculate the corresponding adjustment weights. For instance, load weight, equipment type, altitude, and slope can all be normalized to values ​​of 0-1, thereby aligning multimodal scene parameters within the same space and facilitating the calculation of adjustment weights.

[0155] By employing the technical solution of this embodiment, the parameter thresholds in the current parameter set can be adjusted according to scene parameters, making the parameter thresholds in the current parameter set compatible with the movement scene. This allows for targeted monitoring of movement in different movement scenarios, more accurately preventing the risk of sports injuries. For example, in high-altitude areas, by adjusting the parameter thresholds, altitude sickness that may occur during movement can be detected in a timely manner. Combined with the aforementioned output movement prompts, health risks to the movement can be avoided promptly.

[0156] For example, in sports scenarios with steep inclines, the impact load index threshold can be adjusted, such as by lowering the impact load index threshold, to monitor the risk of knee injury in athletes during mountain climbing or downhill sports. Combined with the aforementioned sports prompts, the risk of knee injury can be avoided.

[0157] In some embodiments, in the types of exercise in rehabilitation training, the parameter thresholds in the parameter set can be updated in stages based on the exercise data of the exercise subject, thereby automatically customizing a rehabilitation exercise plan for the exercise subject.

[0158] For example, please refer to Figure 4 As shown, Figure 4 This illustrates another parameter threshold update step, such as... Figure 4 As shown, the following steps may be included: Step S601: Based on multiple parameter thresholds, output standard rhythm guidance information to the moving object; The standard rhythm guidance information includes gait rhythm information and respiratory rhythm information, which instruct the exercise subject to exercise according to the gait rhythm information and respiratory rhythm information; Step S602: Based on the motion data of the moving object and the standard rhythm guidance information, determine the motion difference between the motion performed by the moving object and the standard motion corresponding to the standard rhythm guidance information; Step S603: Adjust the threshold values ​​of at least some parameters based on motion differences.

[0159] In this embodiment, during the initial stage, after constructing a parameter set based on historical motion data and obtaining multiple parameter thresholds, standard rhythmic guidance information is output to the moving object based on these thresholds. This allows the moving object to perform movements according to the parameter thresholds in the initial stage, ensuring that the movement is controlled by these thresholds. This strategy can be applied to medical rehabilitation exercises.

[0160] For example, for a moving object, the target step frequency range, left-right symmetry threshold A_th, impact load threshold L_th, breathing mode target R_br, and breathing-gait coupling stability target C_target can be issued.

[0161] Among them, the target gait frequency range, left-right symmetry threshold A_th, impact load threshold L_th, breathing mode target R_br, and breathing-gait coupling stability target C_target can be used to generate standard rhythm guidance information. The standard rhythm guidance information includes gait frequency rhythm information and breathing rhythm information to instruct the moving object to move according to the gait frequency rhythm information and breathing rhythm information.

[0162] The standard rhythm guidance information may include beat sounds, vibration beats, and voice prompts, with the voice prompts used to guide the user's breathing rhythm.

[0163] Next, motion data can be collected when the moving object moves according to the standard rhythmic guidance information. It's understandable that even if the moving object moves according to the standard rhythmic guidance information, this motion is not standard and differs from the standard rhythmic guidance information. Therefore, the motion difference between the moving object's motion and the standard motion corresponding to the standard rhythmic guidance information can be determined.

[0164] Specifically, based on respiratory data, the intervals of the inspiratory-to-expiratory step ratio and the inspiratory-to-expiratory duration ratio can be determined. Then, the difference between the intervals of the inspiratory-to-expiratory step ratio and the inspiratory-to-expiratory duration ratio and the intervals of the inspiratory-to-expiratory step ratio and the inspiratory-to-expiratory duration ratio in the parameter set can be compared, and this difference can be taken as the respiratory difference.

[0165] Furthermore, the respiratory-gait coupling stability can be determined based on lower limb movement data and respiratory data, and the difference between it and the respiratory-gait coupling stability threshold in the parameter set can be used as the respiratory-gait coupling stability difference.

[0166] Furthermore, based on lower limb movement data, gait frequency, gait symmetry parameters, and ground contact time can be determined, and these can be compared with the gait frequency threshold range, gait symmetry threshold, and ground contact time threshold in the parameter set to obtain gait differences.

[0167] In some examples, the impact load index can also be calculated based on the lower limb movement data according to formula (2) or (3), and the impact load index can be compared with the impact load threshold in the parameter set to obtain the load difference.

[0168] Thus, movement differences can include the aforementioned differences in breathing, breathing-gait coupling stability, and gait, and in some cases, differences in load.

[0169] Based on the aforementioned differences in movement, it can be determined whether the subject is capable of moving according to the standard rhythmic guidance information. This involves assessing the subject's current physical condition and its adaptation to and compliance with the standard rhythmic guidance information. Compliance includes: the percentage of effective training time within the target time interval, the number and duration of exceeding the threshold, and the recovery time after movement correction.

[0170] Based on the comparison results, it can be determined whether to advance to the next stage.

[0171] If the difference is large, it indicates that the exercise cannot keep up with the standard rhythm guidance information. In this case, the parameter threshold can be adjusted appropriately to reduce the exercise intensity.

[0172] If the difference is small, it indicates that the exercise subject can keep up with the standard rhythmic guidance information. In this case, the current parameter threshold can be maintained for a period of time before updating the parameter threshold to increase the exercise intensity and adapt to the next stage of rehabilitation training.

[0173] When updating the parameter thresholds within the parameter set based on motion differences, it is possible to update some parameter thresholds or all parameter thresholds; no limitation is made here.

[0174] In some embodiments, when outputting motion cues, the corresponding motion cues can be based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold. For example, different comparison results can output different motion cues, which may refer to different information categories.

[0175] For example, a first motion prompt can be output based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, as well as the gait frequency in the lower limb motion data; wherein the first motion prompt includes an inhalation / exhalation prompt synchronized with the gait frequency beat; Furthermore, based on the relationship between the impact load index and the impact load threshold, a second motion prompt is output, which includes correction prompts for any of the stride length, stride frequency, and landing mode.

[0176] In this example, when the respiratory-gait coupling stability is less than the respiratory-gait coupling stability threshold, i.e., when the respiratory-gait coupling stability is low and breathing is disordered, the first motion cue information output can incorporate gait frequency generation from lower limb motion data. For example, it can generate inhalation / exhalation cues synchronized with the gait frequency beat, gradually locking the respiratory rhythm to the ideal inspiratory-expiratory ratio.

[0177] In this example, the difference R_load between the impact load index L and the impact load threshold L_th is obtained by comparing the two. When the difference R_load reaches a preset level and meets the output conditions, a second motion prompt is output. The second motion prompt may include correction prompts for any of stride length, stride frequency, and landing mode. For example, the second motion prompt may be a motion correction prompt such as reducing stride length / increasing stride frequency / adjusting landing mode.

[0178] This system allows for pre-setting multiple difference levels for the impact load index, with different difference levels corresponding to different secondary motion prompt messages. For example, it can include setting low, medium, and high levels, where low level indicates slight load overload, medium level indicates significant load overload, and high level indicates excessive load overload. When the difference R_loa falls within the corresponding difference level, the corresponding secondary motion prompt message can be output.

[0179] For example, at a low level, the impact load index L is higher than the impact load threshold. Based on the difference R_load between the two, the rhythm adjustment amount, such as the cadence adjustment amount and stride adjustment amount, can be calculated, and the cadence beat and inhalation / exhalation prompts synchronized with the ground contact event can be output to reduce the impact load index L.

[0180] In the medium level, if the impact load index L or ground contact time CT continues to rise, it is preferable to first increase the step frequency in small steps to reduce the step size and impact. In the high level, when continuous over-impact is detected and A_th exceeds the limit, a deceleration or road segment change prompt is output.

[0181] As mentioned above, when the difference R_loa reaches the preset level and the output conditions are met, the second motion prompt information is output. The output conditions are met because the motion data is stable and accurate during transmission.

[0182] In other words, before outputting motion prompts, the transmission quality of the motion data and the quality of the transmitted motion data itself can be checked. The former can be represented by channel quality, and the latter by data quality. When both channel quality and data quality meet the standards, the corresponding motion prompts can be output.

[0183] In some embodiments, transmission quality and data quality can be detected when acquiring motion data. Specifically, during the reception of motion data, at least two of the following can be detected: quality flags, packet loss rate, signal-to-noise ratio estimation, and artifacts corresponding to the motion data; and a quality score can be generated based on at least two of the following: quality flags, packet loss rate, signal-to-noise ratio estimation, and artifacts. Among them, the quality score is used to characterize at least one of the acquisition quality of motion data, the transmission quality of motion data, and the data quality of motion data. The quality flag is carried in the motion data and is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference.

[0184] In this embodiment, the quality flag bit Q can be added to the motion data after the wearable sensor collects the motion data, based on whether the device shakes, is loosely worn, or is subject to environmental interference during the collection process.

[0185] In some feasible examples, a quality flag can be added to each motion data frame by a wireless communication module connected to the wearable sensor. In addition, the wireless communication module can also add a sensor channel identifier, sampling frequency information, a monotonically increasing sequence number, and a timestamp to each motion data frame.

[0186] Among them, the sensor channel identifier represents the source of the data; The sampling frequency information can be used to determine the sliding time window mentioned above.

[0187] The monotonically increasing sequence number can be used to represent each motion data frame so that it can be retransmitted based on the monotonically increasing sequence number in the event of packet loss.

[0188] The timestamp is used to mark the acquisition time of the motion data frame in order to align the respiratory data and lower limb motion data.

[0189] The packet loss rate can be calculated by the receiving party of the motion data. Specifically, the packet loss rate can be calculated based on the monotonically increasing sequence number carried in the motion data.

[0190] Among them, the signal-to-noise ratio estimation is used to estimate the quality of the transmission channel between the wireless communication module and the sensor, which can be added to the motion data frame by the wireless communication module.

[0191] Among them, artifacts are the artifact detection results obtained by detecting artifacts in motion data. They reflect the interference or distortion introduced by non-target factors in the collected motion data.

[0192] In this embodiment, a quality score can be generated based on at least two of the following: quality flag, packet loss rate, signal-to-noise ratio estimation, and artifacts.

[0193] For example, a quality score can be generated based on the quality flag and packet loss rate, or a quality score can be generated based on the quality flag, packet loss rate, and signal-to-noise ratio.

[0194] In some examples, the quality score Q can be generated according to the following formula (4), based on the packet loss rate, signal-to-noise ratio estimate, and artifacts, and their respective weights: Q = w1·SNR_norm + w2·(1 PLR) + w3·(1 A_artifact); Where SNR_norm is the normalized signal-to-noise ratio, PLR is the packet loss rate, A_artifact is the artifact percentage, w1, w2, and w3 are weighting coefficients, and the value of Q ranges from [0,1].

[0195] This format is for illustrative purposes only; the specific weights and calculation methods can be adjusted according to the application scenario.

[0196] In this embodiment, when outputting motion prompt information, the motion prompt information can be output based on the relationship between mass fraction, breathing-gait coupling stability and breathing-gait coupling stability threshold, and the relationship between impact load index and impact load threshold.

[0197] For example, when the quality score is greater than a preset quality score, motion prompt information can be output as described above, such as outputting the first rhythm information and the second motion prompt information. When the quality score is lower than a preset quality score, a "signal quality insufficient / please adjust wearing" prompt can be output, and this data segment can be recorded for traceability.

[0198] As mentioned earlier, respiratory data and lower limb movement data need to be aligned in time to improve the accuracy of multimodal data. In some embodiments, data collected by various sensors can be aligned to a unified time reference, and quality issues such as packet loss, noise, saturation, and artifacts can be marked and quantified.

[0199] For example, a unified timeline can be established for each sensor. If multiple sensors use independent clocks, the clock offset and drift parameters of each channel can be estimated through periodic synchronization beacons or handshake messages, and the timestamps can be compensated for drift. Then, the multimodal data can be resampled or interpolated according to the compensated timestamps so that signals such as breathing and gait are aligned within the same time window.

[0200] This ensures that the data collected by the sensor is time-aligned, thereby avoiding calculation errors, improving the accuracy of calculating breathing-gait coupling stability and shock load index, as well as the accuracy of determining parameter thresholds.

[0201] In one embodiment, each data frame can be stored in a local ring buffer, and local caching can be performed when the wireless link (i.e., the transmission channel mentioned above) is disconnected. When the wireless link is restored, the missing motion data frames can be resent according to the sequence number to achieve end-to-end data integrity and timing consistency.

[0202] In some embodiments, motion data of multiple different moving objects can be statistically analyzed to present the motion differences between the different moving objects.

[0203] For example, in group training such as military and police marching with loads, obstacle crossing, and intermittent sprinting, the training intensity, fatigue level, impact load index L, breathing-gait coupling stability C, and mass fraction Q of multiple personnel can be analyzed, such as by analyzing the differences between different trainees using heat maps or charts.

[0204] In practice, when any trainee's risk level reaches the difference level mentioned above and the quality score Q meets the output conditions, a graded warning can be triggered: prompting the instructor to implement interventions such as "reducing intensity / adjusting rhythm / replenishing / suspending training" for the trainee, and recording the intervention timestamp for subsequent review.

[0205] Below, we will provide scenario-based examples of the sports training and risk assessment methods disclosed herein, using several sports types as examples.

[0206] In different scenarios, the parameter set library 800 may include, but is not limited to, the following examples: (a) Running / marathon sports scenario: First, the parameter thresholds in the parameter set include: setting the target cadence range based on F_pref, for example, F_min=F_pref, F_max=F_pref×(1+α), where α is 0.03~0.10; and providing the R_br set and the breathing-gait coupling stability threshold C_target range to improve rhythm stability while maintaining pace. When the impact load index L or ground contact time CT determined based on motion data continues to rise, it is preferable to first increase the step frequency with small steps to reduce the stride length and impact.

[0207] (b) Sports scenarios involving mountain climbing / uphill walking: First, the parameter thresholds in the parameter set are combined with slope and altitude information to set a more conservative ΔF_max (step adjustment amount), and tend to prolong exhalation or increase the inspiratory-to-expiratory step ratio to stabilize the respiratory rhythm. When the respiratory-gait coupling stability C decreases significantly and the left-right symmetry deteriorates, a rhythmic prompt to reduce intensity or allow for segmented rest is output.

[0208] Meanwhile, in this scenario, when the impact load index L increases and the step frequency is low, the beat frequency is increased by taking small steps; when the breathing-gait coupling stability C is low and breathing is disordered, an inhalation / exhalation prompt synchronized with the step frequency beat is output, so that the breathing rhythm is gradually locked to the target inhalation-exhalation ratio.

[0209] (c) Downhill / Technical running scenarios: Parameter set: Set the impact load threshold L_th to be more stringent, and prioritize "increasing step frequency, decreasing step length, and reducing peak ground impact" as the corrective strategy; When the continuous over-impact load threshold L is detected to rise and the gait symmetry threshold A_th exceeds the limit, a prompt to decelerate or change road segment is output.

[0210] (d) Sports scenarios involving heavy loads and high-intensity training for military and police personnel: Parameter set: Introduce a load factor K_load (determined by the input load weight, type of equipment worn, or training subject) to weight and adjust the impact load threshold L_th, ground contact time threshold CT_th, and recovery rule; The training management terminal 710 displays the risk level, quality score Q, and thermal / hypoxia risk gating status for multiple people, and issues graded warnings to trainees who reach the alert level.

[0211] (e) Exercise scenarios for medical and rehabilitation training: Parameter set: Parameter thresholds issued by the medical rehabilitation terminal, including target gait frequency range, gait symmetry threshold A_th, impact load threshold L_th, respiratory mode target set R_br and respiratory-gait coupling stability threshold C_target, etc. The respiratory mode target set R_br includes the inspiratory-exhalation step ratio set (such as 2:2, 2:3, 3:2, etc.) and / or inspiratory-exhalation duration ratio range.

[0212] Different safety thresholds and rhythms are set according to the rehabilitation stage (early / mid / intensive); the system records compliance and completion, that is, it statistically analyzes the differences in exercise according to standard exercise prompts, and adjusts the parameter thresholds according to the differences in exercise.

[0213] Therefore, the output of motion prompts can vary depending on the type of motion.

[0214] Based on the same inventive concept, this disclosure also provides a sports training and risk assessment system, please refer to... Figure 5 As shown, Figure 5 A schematic diagram of the system architecture is shown, such as... Figure 5 As shown, the evaluation system includes a data acquisition module 100, a communication module 200, a data processing and intelligent analysis module 500, and a training feedback and risk alert module 600. The acquisition module includes multiple sensors, which are used to acquire motion data of the moving object. The communication module is connected to the acquisition module and configured to send the motion data to the data processing and intelligent analysis module, add a sensor channel identifier, sampling frequency, sequence number, timestamp, and quality flag to each frame of motion data, and when a disconnection from the data processing and intelligent analysis module is detected, store the motion data to be sent in a local ring buffer, and when the connection with the data processing and intelligent analysis module is restored, resend the missing motion data according to the sequence number and timestamp; wherein, the quality flag is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference; The data processing and intelligent analysis module is configured to execute the aforementioned sports training and risk assessment method; The training feedback and risk warning module is configured to send the exercise warning information to the terminal held by the exercise object.

[0215] The exercise prompt information includes at least one of voice prompt information, text prompt information, beat sound and vibration beat, and the exercise prompt information is used to prompt the user to adjust the cadence and / or breathing rhythm.

[0216] The content executed by each module in the sports training and risk assessment system in this embodiment can be referred to the description in the above embodiment of the sports training and risk assessment method, and will not be repeated here.

[0217] like Figure 5As shown, the communication module can be a wireless communication module, which can send the motion data collected by the acquisition module to the data processing terminal 300 and / or the cloud server 400. The data processing and intelligent analysis module can be located in the data processing terminal 300 and / or the cloud server 400.

[0218] like Figure 6 As shown, the acquisition module may include various types of wearable sensors, such as inertial measurement units (120a, 120b) and a respiratory signal acquisition unit 110 for acquiring chest and abdominal respiratory movements, as described in the risk assessment method embodiment.

[0219] The respiratory signal acquisition unit 110 is used to acquire the above-mentioned respiratory data, including respiratory sound data.

[0220] like Figure 5 As shown, the system may also include a parameter library 800, which stores parameter sets corresponding to each moving object under different motion types. The parameter library may be located in the cloud 400.

[0221] like Figure 5 As shown, the system may also include a remote monitoring terminal 700 connected to the data processing, training feedback and risk warning module. The remote monitoring terminal 700 includes a training management terminal 710 and / or the aforementioned medical rehabilitation terminal 720. The training management terminal 710 is used to display and provide graded warnings for the risk levels of multiple trainees in groups.

[0222] The device held by the person exercising can be a mobile phone, a smart bracelet, a smart watch, or other similar device.

[0223] The data processing and intelligent analysis module may include a first determining module, used to determine the respiratory phase within the sliding time window based on the respiratory data within the sliding time window; Based on the lower limb movement data within the sliding time window, determine the gait phase within the sliding time window; Based on the respiratory data and lower limb movement data within the sliding time window, parameter K is determined, wherein parameter K represents the integer ratio between breathing and gait. Based on the parameter K, the breathing phase, and the gait phase, determine the phase difference between the breathing phase and the gait phase within the sliding time window; The breathing-gait coupling stability is determined based on the phase difference corresponding to each of the sliding time windows.

[0224] The data processing and intelligent analysis module may include a second determining module, used to obtain the peak tibial / ankle acceleration and impact rise slope of the object's foot at the moment of ground contact from the lower limb motion data; and to determine the impact load index based on the peak tibial / ankle acceleration, the impact rise slope, the ground contact time, and their respective weights.

[0225] In some examples, the second determining module may also acquire individual parameters of the moving object, including at least one of the moving object's weight and walking speed; based on the individual parameters, normalize the peak tibial / ankle acceleration, the impact rise slope, and the ground contact time; and determine the impact load index based on the peak tibial / ankle acceleration, impact rise slope, ground contact time, and their respective weights obtained after the normalization.

[0226] The data processing and intelligent analysis module may include a first parameter update module, which performs the following steps: Acquire historical motion data collected within a preset time period. The historical motion data includes gait frequency data and respiratory rhythm data of the moving object in a natural state, wherein the respiratory rhythm data includes respiratory rate. Based on the gait frequency data and respiratory rhythm data in a natural state, determine the distribution of baseline gait frequency, baseline respiratory rate, and baseline respiratory-gait coupling stability. Based on the distribution of baseline gait frequency, baseline respiratory rate, baseline respiratory-gait coupling stability, and a preset forgetting factor, update at least some of the parameter thresholds. The forgetting factor is used to adjust the influence weight of the historical motion data on the parameter thresholds. Among them, the multiple parameter thresholds also include at least one of the following: cadence threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0227] The data processing and intelligent analysis module may include a second parameter update module, which performs the following steps: Based on the type of exercise, scene parameters corresponding to the type of exercise are obtained, and different types of exercise correspond to different scene parameters; wherein, the scene parameters include at least one of the following: load weight, type of equipment worn, altitude, slope, and training stage; Based on the scenario parameters, determine the adjustment weights corresponding to the multiple parameter thresholds in the parameter set; Based on the adjusted weights, the threshold values ​​of multiple parameters are updated.

[0228] The training feedback and risk warning module can be used to output first movement warning information based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the gait frequency in the lower limb movement data; wherein the first movement warning information includes inhalation / exhalation warnings synchronized with the gait frequency beat; and output second movement warning information based on the relationship between the impact load index and the impact load threshold, wherein the second movement warning information includes correction warnings for any of stride length, gait frequency, and landing mode.

[0229] The communication module is further configured to detect at least two of the following during the reception of the motion data: a quality flag bit, packet loss rate, signal-to-noise ratio estimation, and artifacts. The quality flag bit is carried in the motion data and is used to indicate states such as saturation, severe shaking, loose fitting, or environmental interference. A quality score is generated based on at least two of the quality flag bit, packet loss rate, signal-to-noise ratio estimation, and artifacts. The quality score is used to characterize at least one of the acquisition quality of the motion data, the transmission quality of the motion data, and the data quality of the motion data.

[0230] Accordingly, the training feedback and risk warning module described above can be used to output the motion warning information based on the relationship between the quality score, the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold.

[0231] The exercise type includes rehabilitation exercise type, and the multiple parameter thresholds also include: cadence threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

[0232] The system may also include a remote monitoring terminal for receiving the assessment results to enable group monitoring, tiered early warning, and / or rehabilitation follow-up.

[0233] The system may also include a medical rehabilitation terminal for performing the following steps: Based on multiple parameter thresholds, standard rhythm guidance information is output to the moving object; the standard rhythm guidance information includes gait frequency rhythm information and respiratory rhythm information, to instruct the moving object to move according to the gait frequency rhythm information and respiratory rhythm information; Based on the motion data of the moving object and the standard rhythm guidance information, determine the motion difference between the motion performed by the moving object and the standard motion corresponding to the standard rhythm guidance information; Based on the motion differences, at least some of the parameter thresholds are adjusted.

[0234] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0235] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0236] The array substrate, display panel, and liquid crystal 40 projector provided in this disclosure have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this disclosure. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this disclosure. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this disclosure. Therefore, the content of this specification should not be construed as a limitation of this disclosure.

[0237] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0238] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0239] The terms "an embodiment," "embodiment," or "one or more embodiments" as used herein mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of this disclosure. Furthermore, please note that the examples of the phrase "in one embodiment" do not necessarily all refer to the same embodiment.

[0240] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this disclosure may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0241] In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This disclosure can be implemented by means of hardware comprising a plurality of different elements and by means of a suitably programmed computer. In a unit claim enumerating a plurality of means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0242] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit them. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A method for sports training and risk assessment, characterized in that, The method includes: Real-time acquisition of motion data collected from a moving object, the motion data including respiratory data and lower limb motion data, the respiratory data including at least one of respiratory rate, respiratory amplitude and / or inspiratory-to-expiratory ratio, the lower limb motion data including at least one of cadence, ground contact time, swing time and left-right symmetry parameters; Obtain a parameter set corresponding to the current movement type of the moving object. The movement type is used to characterize the current movement scenario of the moving object. Different movement types correspond to different parameter sets. The parameter set includes multiple parameter thresholds. The multiple parameter thresholds include at least a breathing-gait coupling stability threshold corresponding to the breathing data and an impact load threshold corresponding to the lower limb movement data. Based on the respiratory data and the lower limb movement data, the respiratory-gait coupling stability of the exercise object is determined, and based on the lower limb movement data, the impact load index of the exercise object is determined; wherein, the respiratory-gait coupling stability is used to characterize the relative relationship between breathing and gait; Based on the respiratory sound data, the respiratory sound features of the moving object are extracted, and the respiratory sound risk features are determined; Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, the relationship between the impact load index and the impact load threshold, and the relationship between the breathing sound risk characteristics and the corresponding parameter threshold, motion prompt information is output for the moving object; The exercise prompt information includes at least one of voice prompt information, text prompt information, beat sound and vibration beat, and the exercise prompt information is used to prompt the user to adjust the cadence and / or breathing rhythm.

2. The method according to claim 1, characterized in that, The extracted breath sound features include at least one of the following: respiratory cycle sound energy distribution, dominant frequency band energy, frequency band ratio, respiratory phase transition features, continuity features, abnormal crackles, dry rales, wet rales, wheezing, and / or respiratory recovery time; the determination of breath sound risk features includes: Based on the extracted breath sound features, the following breath sound risk features were determined: The exercise subject's ventilation load status, airway patency status, respiratory recovery status, and / or probability of abnormal respiratory events are at least one of the following:

3. The method for sports training and risk assessment according to claim 1, characterized in that, The exercise prompt information includes at least one of rhythm prompt information, load adjustment prompt information, rehabilitation guidance information and / or risk warning information; The load adjustment prompts and / or rehabilitation guidance information include at least one of the following: adjusting cadence, adjusting breathing rhythm, reducing training intensity, shortening training duration, extending recovery time, switching to recovery training mode, switching to rehabilitation guidance mode, and prompting to pause training.

4. The method for sports training and risk assessment according to claim 3, characterized in that, The method further includes: Based on the combined changing trends of the breathing-gait coupling stability, the impact load index, and the respiratory sound risk characteristics, the level of excessive fatigue risk and / or the level of sports injury risk are determined. When the risk level of excessive fatigue and / or the risk level of sports injury exceed a preset threshold, the load adjustment prompt information and / or the risk warning information are output.

5. The method for sports training and risk assessment according to claim 1, characterized in that, The determination of the respiratory-gait coupling stability of the moving object based on the respiratory data and the lower limb movement data includes: Based on the respiratory data within the sliding time window, determine the respiratory phase within the sliding time window; Based on the lower limb movement data within the sliding time window, determine the gait phase within the sliding time window; Based on the respiratory data and lower limb movement data within the sliding time window, parameter K is determined, wherein parameter K represents the integer ratio between breathing and gait. Based on the parameter K, the breathing phase, and the gait phase, determine the phase difference between the breathing phase and the gait phase within the sliding time window; The breathing-gait coupling stability is determined based on the phase difference corresponding to each of the sliding time windows.

6. The method for sports training and risk assessment according to claim 1, characterized in that, The step of determining the impact load index of the exercise object based on the lower limb movement data includes: From the lower limb motion data, obtain the peak tibial / ankle acceleration and impact rise slope of the moving object at the moment of foot contact with the ground; The impact load index is determined based on the peak tibial / ankle acceleration, the slope of the impact rise, the ground contact time, and their respective weights.

7. The method for sports training and risk assessment according to claim 6, characterized in that, The determination of the impact load index based on the peak tibial / ankle acceleration, the impact rise slope, and the ground contact time, and their respective weights, includes: Obtain individual parameters of the moving object, wherein the individual parameters include at least one of the moving object's weight and walking speed; Based on the individual parameters, the peak tibial / ankle acceleration, the impact rise slope, and the ground contact time were normalized respectively. The impact load index is determined based on the tibial / ankle peak acceleration, impact rise slope, ground contact time, and their respective weights obtained after the normalization process.

8. The method for sports training and risk assessment according to claim 1, characterized in that, The method further includes: Acquire historical motion data collected within a preset time period. The historical motion data includes the cadence data and respiratory rhythm data of the moving object in a natural state. The respiratory rhythm data includes respiratory rate. Based on the gait frequency and respiratory rhythm data under natural conditions, the distributions of baseline gait frequency, baseline respiratory rate, and baseline respiratory-gait coupling stability are determined. Based on the baseline cadence, baseline respiratory rate, baseline respiratory-gait coupling stability distribution, and a preset forgetting factor, at least some of the parameter thresholds are updated; the forgetting factor is used to adjust the influence weight of the historical motion data on the parameter thresholds. Among them, the multiple parameter thresholds also include at least one of the following: cadence threshold range, inhalation-exhalation step ratio and inhalation-exhalation duration ratio range, gait symmetry threshold and ground contact time threshold.

9. The method for sports training and risk assessment according to claim 1, characterized in that, The method further includes: During the process of receiving the motion data, at least two of the following are detected: quality flag bit, packet loss rate, signal-to-noise ratio estimate, and artifacts corresponding to the motion data; wherein, the quality flag bit is carried in the motion data and is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference. A quality score is generated based on at least two of the quality flag, packet loss rate, signal-to-noise ratio estimation, and artifacts; the quality score is used to characterize at least one of the acquisition quality of the motion data, the transmission quality of the motion data, and the data quality of the motion data. Based on the relationship between the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, motion prompts are output for the moving object, including: Based on the relationship between the mass fraction, the breathing-gait coupling stability and the breathing-gait coupling stability threshold, and the relationship between the impact load index and the impact load threshold, the motion prompt information is output.

10. The method according to any one of claims 1 to 8, characterized in that, The method further includes: Based on the type of exercise, scene parameters corresponding to the type of exercise are obtained, and different types of exercise correspond to different scene parameters; wherein, the scene parameters include at least one of the following: load weight, type of equipment worn, altitude, slope, and training stage; Based on the scenario parameters, determine the adjustment weights corresponding to the multiple parameter thresholds in the parameter set; Based on the adjusted weights, the threshold values ​​of multiple parameters are updated.

11. The method for sports training and risk assessment according to claim 1, characterized in that, The exercise type includes rehabilitation exercise types, and the multiple parameter thresholds also include: cadence threshold range, inhalation / exhalation step ratio and inhalation / exhalation duration ratio range, gait symmetry threshold and ground contact time threshold; the method further includes: Based on multiple parameter thresholds, standard rhythm guidance information is output to the moving object; the standard rhythm guidance information includes gait frequency rhythm information and respiratory rhythm information, to instruct the moving object to move according to the gait frequency rhythm information and respiratory rhythm information; Based on the motion data of the moving object and the standard rhythm guidance information, determine the motion difference between the motion performed by the moving object and the standard motion corresponding to the standard rhythm guidance information; Based on the motion differences, at least some of the parameter thresholds are adjusted.

12. A sports training and risk assessment system, characterized in that, The evaluation system includes a data acquisition module, a communication module, a data processing and intelligent analysis module, and a training feedback and risk warning module. The acquisition module includes multiple sensors, which are used to acquire motion data of the moving object. The communication module is connected to the acquisition module and configured to send the motion data to the data processing and intelligent analysis module, add a sensor channel identifier, sampling frequency, sequence number, timestamp, and quality flag to each frame of motion data, and when a disconnection from the data processing and intelligent analysis module is detected, store the motion data to be sent in a local ring buffer, and when the connection with the data processing and intelligent analysis module is restored, resend the missing motion data according to the sequence number and timestamp; wherein, the quality flag is used to indicate states such as saturation, severe shaking, loose wearing, or environmental interference; The data processing and intelligent analysis module is configured to execute the sports training and risk assessment method according to any one of claims 1-11; The training feedback and risk warning module is configured to send the exercise warning information to the terminal held by the exercise object.

Images