Tolerance evaluation method and device, computer readable storage medium, terminal

Abstract

A kind of endurance evaluation method and device, computer readable storage medium, terminal, method includes: determining the acceleration signal of the multiple to be evaluated movement of evaluation object in the first movement duration of pre-set;Based on the first interval duration and first acceleration threshold, determine multiple target peaks and target troughs;For each target trough, determine the previous and the next target peak adjacent to the current target trough;If the time interval of the two target peaks determined is less than or equal to the second interval duration, then determine the start time of the current to be evaluated movement based on the first target peak, and determine the end time of the current to be evaluated movement based on the second target peak;According to the start time and end time of each time, determine endurance index;Wherein, the first interval duration is less than or equal to the second interval duration.The above scheme helps to improve the efficiency of endurance evaluation, improve the accuracy and reliability of evaluation result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of exercise endurance evaluation, and in particular to an endurance evaluation method and device, a computer readable storage medium, and a terminal. BACKGROUND

[0002] The exercise endurance of human muscles can reflect the muscle control ability of the human body to some extent, and plays a very important role in people's daily work and life, as well as various sports. For example, the good or bad of the lower limb exercise endurance of the human body can directly affect the daily walking, running, stair climbing, etc. Therefore, evaluating the lower limb exercise endurance of the human body can help to avoid accidental injuries caused by insufficient functions. In addition, by evaluating the lower limb exercise endurance of athletes before and after training, the training effect can be understood in time, and the training scheme can be adjusted in time.

[0003] In the prior art, the commonly used exercise endurance evaluation method often uses an exercise endurance evaluation scale and is based on the subjective report of a patient to evaluate. This method has the problems of low evaluation sensitivity and large individual differences in the perception of question expression, thereby resulting in insufficient accuracy and reliability of the evaluation result. Another commonly used evaluation method is to perform a walking test within a certain time period. This method requires a specific site, and usually also requires the presence of relevant professionals, and has high testing cost, low convenience and low efficiency. SUMMARY

[0004] The technical problem solved by the embodiments of the present application is how to improve the accuracy and reliability of the evaluation result on the basis of improving the efficiency of endurance evaluation.

[0005] To solve the above technical problem, the embodiments of the present application provide an endurance evaluation method, comprising the following steps: determining an acceleration signal of an evaluation object performing a plurality of to-be-evaluated exercises within a preset first exercise duration; determining a plurality of target wave crests and a plurality of target wave troughs from the acceleration signal based on a preset first interval duration and a first acceleration threshold; for each target wave trough, determining a previous target wave crest and a next target wave crest adjacent to the current target wave trough, and recording them as a first target wave crest and a second target wave crest respectively; if the time interval between the first target wave crest and the second target wave crest is less than or equal to a preset second interval duration, determining the starting time of the current to-be-evaluated exercise based on the first target wave crest, and determining the ending time of the current to-be-evaluated exercise based on the second target wave crest; determining an endurance index of the evaluation object performing the to-be-evaluated exercise according to the starting time and the ending time of each to-be-evaluated exercise; wherein the first interval duration is less than or equal to the second interval duration.

[0006] Optionally, the determining the plurality of target peaks and the plurality of target troughs from the acceleration signal based on the preset first interval duration and the first acceleration threshold comprises: selecting, from the peaks of the acceleration signal, peaks that satisfy that an interval duration between each two adjacent peaks is greater than or equal to the first interval duration and an acceleration value is greater than the first acceleration threshold, as the target peaks; and determining, between each two adjacent target peaks, a trough with a minimum acceleration value and less than the first acceleration threshold, as the target trough.

[0007] Optionally, the first acceleration threshold is obtained by performing a weighted operation on a mean value and a standard deviation of the acceleration signal with a preset weight ratio.

[0008] Optionally, the determining the start time of the current to-be-evaluated movement based on the first target peak comprises: taking a time point of a trough adjacent to the first target peak, or a time point of an acceleration value less than or equal to a mean value of the acceleration signal and before the first target peak, as the start time of the current to-be-evaluated movement.

[0009] Optionally, the determining the end time of the current to-be-evaluated movement based on the second target peak comprises: taking a time point of a trough adjacent to the second target peak, or a time point of an acceleration value less than or equal to a mean value of the acceleration signal and after the second target peak, as the end time of the current to-be-evaluated movement.

[0010] Optionally, the endurance index of the evaluation object performing the to-be-evaluated movement comprises one or more of: a mean movement duration, a median movement duration, a variability parameter, a single longest movement duration, a single shortest movement duration, a movement duration skewness, a movement duration kurtosis, a movement duration interquartile range, a movement duration median absolute error, a movement duration root mean square, a movement duration standard deviation, a movement duration variance, a movement duration empirical distribution function percentile, a movement duration empirical distribution function slope of each to-be-evaluated movement performed by the evaluation object within the first movement duration.

[0011] Optionally, the endurance index of the evaluation object performing the to-be-evaluated movement is the variability parameter of each to-be-evaluated movement performed by the evaluation object within the first movement duration; and the determining the endurance index of the evaluation object performing the to-be-evaluated movement based on the start time and the end time of each to-be-evaluated movement comprises: determining an actual movement duration of each to-be-evaluated movement based on the start time and the end time of each to-be-evaluated movement; determining a mean movement duration and a movement duration standard deviation of each to-be-evaluated movement based on the actual movement duration of each to-be-evaluated movement; and performing a division on the mean movement duration and the movement duration standard deviation of each to-be-evaluated movement to obtain the variability parameter of each to-be-evaluated movement performed by the evaluation object.

[0012] Optionally, the determining the acceleration signal of the evaluation object performing the plurality of to-be-evaluated motions within the preset first motion duration comprises: collecting triaxial acceleration signals of the evaluation object performing the plurality of to-be-evaluated motions within the first motion duration; and performing a modulus operation on the triaxial acceleration signals to determine the acceleration signal.

[0013] Optionally, after the acceleration signal is determined, the method further comprises: performing noise reduction processing and smoothing processing on the acceleration signal.

[0014] Optionally, after the endurance index of the evaluation object performing the to-be-evaluated motion is determined, the method further comprises: inputting the endurance index into a pre-trained lower limb motion endurance prediction model to determine a lower limb motion endurance grade of the evaluation object; wherein the pre-trained lower limb motion endurance prediction model is obtained by training a machine learning model using endurance indexes of a plurality of sample objects performing the to-be-evaluated motion as training sample data.

[0015] Optionally, the training sample data of the pre-trained lower limb motion endurance prediction model further comprises first annotation data, the first annotation data comprising walking distances of the plurality of sample objects within a preset second motion duration; and when the lower limb motion endurance grade of the evaluation object is determined, a corresponding walking distance is also determined.

[0016] Optionally, the training sample data of the pre-trained motion endurance prediction model further comprises second annotation data, the second annotation data comprising cardiopulmonary exercise test indexes of the plurality of sample objects; and when the motion endurance grade of the evaluation object is determined, a corresponding cardiopulmonary exercise test index is also determined.

[0017] Optionally, the endurance index is a lower limb motion endurance index; and the to-be-evaluated motion is selected from a sit-to-stand motion, a deep squat motion, a squat jump motion, and a vertical jump motion.

[0018] The embodiment of the present application also provides a kind of endurance evaluation device, comprising: acceleration signal determination module, for determining the acceleration signal of the multiple evaluation objects in the first movement duration of preset for carrying out multiple to be evaluated movement;Wave crest and trough determination module, based on the first interval duration of preset and first acceleration threshold, from the acceleration signal, determine multiple target wave crest and multiple target wave trough;Target wave crest determination module, for each target wave trough, determine the previous target wave crest and the next target wave crest adjacent to current target wave trough, respectively recorded as first target wave crest and second target wave crest;Start time and end time determination module, for if the time interval of the first target wave crest and the second target wave crest is less than or equal to the second interval duration of preset, based on the first target wave crest, determine the start time of current to be evaluated movement, and based on the second target wave crest, determine the end time of current to be evaluated movement;Endurance index determination module, for determining the endurance index of the evaluation object for carrying out the to be evaluated movement according to the start time and end time of each to be evaluated movement;Wherein, the first interval duration is less than or equal to the second interval duration.

[0019] The embodiment of the present application also provides a kind of computer readable storage medium, which stores computer program, the computer program is run on processor, executes the steps of the above-mentioned endurance evaluation method.

[0020] The embodiment of the present application also provides a kind of terminal, comprising memory and processor, the memory is stored with the computer program capable of running on the processor, the processor runs the computer program, executes the steps of the above-mentioned endurance evaluation method.

[0021] Compared with prior art, the technical scheme of the embodiment of the present application has the following beneficial effects:

[0022] The embodiment of the present application provides a kind of endurance evaluation method, determines the acceleration signal of the multiple evaluation objects in the first movement duration of preset for carrying out multiple to be evaluated movement;Based on the first interval duration of preset and first acceleration threshold, from the acceleration signal, determine multiple target wave crest and multiple target wave trough;For each target wave trough, determine the previous target wave crest and the next target wave crest adjacent to current target wave trough, respectively recorded as first target wave crest and second target wave crest;If the time interval of the first target wave crest and the second target wave crest is less than or equal to the second interval duration of preset, based on the first target wave crest, determine the start time of current to be evaluated movement, and based on the second target wave crest, determine the end time of current to be evaluated movement;According to the start time and end time of each to be evaluated movement, determine the endurance index of the evaluation object for carrying out the to be evaluated movement;Wherein, the first interval duration is less than or equal to the second interval duration.

[0023] In the embodiments of the present application, by setting the preset first interval duration and the first acceleration threshold, there is an opportunity to eliminate the peaks in the acceleration signal that have an inappropriate interval duration or acceleration value (invalid peaks), thereby helping to screen out a plurality of target peaks and a plurality of target troughs that can accurately reflect the actual acceleration and deceleration state during the movement. Further, if the time interval between the first target peak and the second target peak is too long (less than or equal to the preset second interval duration), it means that the evaluation object is likely to have not effectively performed the to-be-evaluated movement (for example, distracted or paused for other reasons) during this time interval. Based on this, the present embodiment can further eliminate the target peaks with an excessively long interval duration from the plurality of determined target peaks, based on the elimination of invalid peaks. Thereby, it helps to improve the accuracy of the start time and end time of each to-be-evaluated movement, and further improve the accuracy of the endurance evaluation result.

[0024] Further, the determination of the start time of the current to-be-evaluated movement based on the first target peak includes: taking the time point of a trough adjacent to the first target peak, or the time point of the last acceleration value less than or equal to the average value of the acceleration signal before the first target peak, as the start time of the current to-be-evaluated movement. In actual application, since there is usually a certain non-movement duration (also referred to as interference duration) between adjacent movements (such as the intermediate process from the end of the previous sit-stand movement to the start of the current sit-stand movement), for example, a stationary phase between adjacent movements, by using the present embodiment, the interference duration before the start of the current movement can be removed to obtain effective "dehydration" data, thereby improving the accuracy and reliability of the subsequently determined endurance index.

[0025] Further, the determination of the acceleration signal of the evaluation object performing a plurality of to-be-evaluated movements within a preset first movement duration includes: collecting three-axis acceleration signals of the evaluation object performing the plurality of to-be-evaluated movements within the first movement duration; and performing a modulus operation on the three-axis acceleration signals to determine the acceleration signal. In the embodiments of the present application, by performing a modulus operation, the acceleration vectors in multiple directions in the three-axis acceleration signals are converted into an acceleration scalar with a single direction, which can remove the influence of the directionality of each to-be-evaluated movement and focus on the main movement direction of the to-be-evaluated movement (for example, the vertical direction in the sit-stand movement), thereby further improving the accuracy of the endurance evaluation.

[0026] Further, the training sample data of the pre-trained lower limb exercise endurance prediction model further comprises labeled data, the labeled data comprising walking distances of the plurality of sample objects in a preset second exercise duration; when determining the lower limb exercise endurance level of the evaluation object, the corresponding walking distance is determined. In the embodiment of the present application, on the one hand, by converting the exercise endurance indicators which may contain multiple data into exercise endurance levels and walking distances in the second exercise duration, the intuitive feeling of the evaluation result can be enhanced, and the user experience can be improved; on the other hand, in a specific application scenario, such as in the walking competitive training of athletes, by obtaining the walking distances in different training stages, the training effect of each stage can be more intuitively and quickly analyzed and monitored, and the training scheme can be further improved in a targeted manner. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a flowchart of a kind of endurance evaluation method in the embodiment of the present application;

[0028] Figure 2 is Figure 1 a flowchart of a specific implementation of step S12 in

[0029] Figure 3 is a schematic diagram of determining target wave crest from acceleration signal in the embodiment of the present application;

[0030] Figure 4 is a schematic diagram of determining target wave trough from acceleration signal in the embodiment of the present application;

[0031] Figure 5 is a schematic diagram of determining start time and end time of single to-be-evaluated exercise from acceleration signal in the embodiment of the present application;

[0032] Figure 6 is Figure 1 a flowchart of a specific implementation of step S15 in

[0033] Figure 7 is a structural schematic diagram of a kind of endurance evaluation device in the embodiment of the present application. DETAILED DESCRIPTION

[0034] As described in the background, it is of great significance to evaluate the muscle exercise endurance of human body.

[0035] In the prior art, commonly used exercise endurance evaluation methods often use exercise endurance evaluation scales and are based on subjective reports of patients for evaluation. Such methods have low evaluation sensitivity and large individual differences in perception of problem expression, resulting in insufficient accuracy and reliability of evaluation results. Another commonly used evaluation method is to perform a walking test for a certain length of time. This method requires a specific site and is usually performed with the presence of relevant professionals, and has high testing costs, low convenience and low efficiency.

[0036] To solve the above technical problems, an embodiment of the present application provides an endurance evaluation method, which specifically comprises: determining acceleration signals of an evaluation object performing a plurality of to-be-evaluated exercises within a preset first exercise duration; determining a plurality of target peaks and a plurality of target troughs from the acceleration signals based on a preset first interval duration and a first acceleration threshold; for each target trough, determining a previous target peak and a subsequent target peak adjacent to the current target trough, and recording the previous target peak and the subsequent target peak as a first target peak and a second target peak, respectively; if a time interval between the first target peak and the second target peak is less than or equal to a preset second interval duration, determining a start time of the current to-be-evaluated exercise based on the first target peak and determining an end time of the current to-be-evaluated exercise based on the second target peak; determining an endurance index of the evaluation object performing the to-be-evaluated exercises according to the start time and the end time of each to-be-evaluated exercise; and wherein the first interval duration is less than or equal to the second interval duration.

[0037] From the above, in the embodiment of the present application, by the preset first interval duration and the first acceleration threshold, there is an opportunity to eliminate peaks (invalid peaks) in the acceleration signal that have an inappropriate interval duration or an inappropriate acceleration value, thereby helping to screen a plurality of target peaks and a plurality of target troughs that can accurately reflect the actual acceleration and deceleration states during the exercise. Further, if the time interval between the first target peak and the second target peak is too long (less than or equal to the preset second interval duration), it means that the evaluation object is likely to not perform effective to-be-evaluated exercises (e.g., distracted or other reasons for pausing) within this interval duration. Based on this, on the basis of having eliminated invalid peaks, by the preset second interval duration, target peaks with excessively long interval durations can be further eliminated from the determined plurality of target peaks. Thus, it helps to improve the accuracy of the determined start time and end time of each to-be-evaluated exercise, and further improves the accuracy of the endurance evaluation result.

[0038] To make the above-mentioned purposes, features and benefits of the present application more obvious and easy to understand, specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0039] Reference Figure 1 , Figure 1This is a flowchart of an endurance assessment method according to an embodiment of the present invention. The method may include steps S11 to S14:

[0040] Step S11: Determine the acceleration signals of the evaluation object during multiple motions to be evaluated within a preset first motion duration;

[0041] Step S12: Based on a preset first interval duration and a first acceleration threshold, determine multiple target peaks and multiple target troughs from the acceleration signal;

[0042] Step S13: For each target valley, determine the previous target peak and the next target peak adjacent to the current target valley, and denot them as the first target peak and the second target peak, respectively;

[0043] Step S14: If the time interval between the first target peak and the second target peak is less than or equal to the preset second interval duration, then determine the start time of the current motion to be evaluated based on the first target peak, and determine the end time of the current motion to be evaluated based on the second target peak.

[0044] Step S15: Determine the endurance index of the evaluation object for the evaluation exercise based on the start and end times of each exercise to be evaluated.

[0045] Wherein, the duration of the first interval is less than or equal to the duration of the second interval.

[0046] In the specific implementation of step S11, the assessment subjects may include athletes, fitness enthusiasts, people with muscle-related diseases or at risk of developing muscle-related diseases, children whose muscles are in the development stage, and other subjects with muscle endurance assessment needs, such as healthy adults.

[0047] The motion to be evaluated can be any motion with two opposite directions of motion. In one direction (referred to as the first direction of motion), there is a process of acceleration followed by deceleration until stopping. In the other direction (referred to as the second direction of motion), there is also a process of acceleration followed by deceleration until stopping. In both the first and second directions of motion, there are consecutive maximum accelerations and maximum decelerations.

[0048] In a non-limiting sense, the exercise to be evaluated can be selected from: sit-stand exercises, squat exercises, squat jumps, and vertical jumps, etc. In the examples of exercises to be evaluated listed above, the body parts used are mainly the lower limbs; therefore, the corresponding endurance index can specifically refer to the lower limb exercise endurance index, or the muscle endurance index of the lower limbs. The lower limbs can refer to the part of the human body below the abdomen, for example, including the buttocks, thighs, knees, shins, and feet.

[0049] In specific implementations, the first movement duration can be properly set according to the age of the evaluation subject, the muscle health status, the purpose of the endurance evaluation, and the like. It can be understood that, in most application scenarios, the first movement duration should not be too long, otherwise the movement intensity is too large, which is prone to risks such as excessive fatigue, muscle or bone injury, and the like; the first movement duration should not be too short, otherwise sufficient evaluation data cannot be obtained, which leads to a decrease in the accuracy and reliability of the evaluation result.

[0050] In some embodiments, an appropriate duration in [30s, 90s] can be selected as the first movement duration, for example, the first movement duration is set to 60s.

[0051] Further, the step S11 can include: collecting a three-axis acceleration signal (usually represented as a continuous signal) of the evaluation subject performing the plurality of to-be-evaluated movements within the first movement duration; and performing a modulus operation on the three-axis acceleration signal to determine the acceleration signal.

[0052] Specifically, the sampling rate of collecting the three-axis acceleration signal can be properly set in combination with the actual operation data amount and operation efficiency requirements, and the like, for example, it can be set to 20Hz.

[0053] In the embodiments of the present application, by performing the modulus operation on the three-axis acceleration signal, the acceleration vector with multiple directions can be converted into the acceleration scalar with a single direction. In this way, the influence of the directionality of each to-be-evaluated movement can be effectively removed, and the main movement direction of the to-be-evaluated movement (for example, the vertical direction in the sit-to-stand movement) can be focused on, thereby further improving the accuracy of the endurance evaluation.

[0054] Further, after the acceleration signal is determined, the method further includes: performing noise reduction processing and smoothing processing on the acceleration signal.

[0055] Specifically, one or more of a Finite Impulse Response (FIR) low-pass filtering algorithm, an Infinite Impulse Response (IIR) low-pass filtering algorithm, a moving average filtering algorithm, and the like, or other appropriate filtering algorithms can be used for noise reduction processing and smoothing processing. When the FIR low-pass filtering algorithm is used for noise reduction processing, the cutoff frequency can be set to 8Hz; when the moving average filtering algorithm is used for smoothing processing, the window width can be set to 0.35 seconds.

[0056] In the embodiment of the present application, by performing the noise reduction processing and the smoothing processing on the acceleration signal, the noise data in the acceleration signal can be filtered out, the quality of the acceleration signal for subsequent data processing and analysis can be improved, and the accuracy and reliability of the evaluation result can be improved.

[0057] In the implementation of step S12, the first interval duration is used to indicate a minimum interval duration that should be met between the determined adjacent two target wave crests. Specifically, it can be indicated that, during a single to-be-evaluated movement of the evaluation object, a minimum interval duration should be met between a moment at which the maximum acceleration in the first movement direction occurs and a moment at which the maximum deceleration in the second movement direction occurs.

[0058] Taking sit-to-stand movement as an example, for a single sit-to-stand, the first interval duration can specifically indicate a minimum interval duration that should be met between a moment at which the maximum acceleration in the standing-up process occurs and a moment at which the maximum deceleration in the sitting-down process occurs.

[0059] It can be understood that, in the acceleration signal, if the interval duration between adjacent two wave crests is too short, it can mean that the two wave crests can be noise data generated in the sampling, or the corresponding single movement is unqualified (for example, in a certain sit-to-stand, the person stands up immediately and then sits down, and the sit-to-stand is not complete and effective). Therefore, by the first interval duration, such invalid wave crests can be removed.

[0060] In the implementation, the specific value of the first interval duration can be set in combination with actual application, for example, can be set in combination with the age of the evaluation object, the health status of the movement part (for example, lower limbs), historical related data of endurance evaluation, and the like. Non-limitingly, an appropriate value in [0.3 seconds, 1.3 seconds] can be selected as the first interval duration.

[0061] The first acceleration threshold can be used to indicate a minimum acceleration value that should be met by each target wave crest. Specifically, it can be indicated that, during a single to-be-evaluated movement of the evaluation object, a minimum value that should be met by the maximum acceleration in the first movement direction, and a maximum value that should be met by the maximum deceleration in the second movement direction.

[0062] In the implementation, the first acceleration threshold can be obtained by performing weighted operation on the mean value and the standard deviation of the acceleration signal by using a preset weight ratio.

[0063] The average value can represent an average level of acceleration values at each time in the acceleration signal, and the standard deviation can represent fluctuation of the acceleration values at each time around the average value. Therefore, in the embodiment of the present application, the weighted operation result of both is used as the first acceleration threshold, which helps to eliminate too small peaks (such peaks can be noise data, or peaks generated by invalid motion, etc.) from each peak in the acceleration signal, and improves the accuracy of the determined target peak.

[0064] Non-limitingly, the preset weight ratio can be selected from [4 / 5:5 / 4], for example, the weight ratio can be set to 5 / 4 (i.e. 1:0.8).

[0065] Referring to Figure 2 , Figure 2 is Figure 1 a flow chart of one specific implementation of step S12. The step S12 can specifically include steps S21 to S22.

[0066] In step S21, peaks satisfying that the interval duration of each adjacent two peaks is greater than or equal to the first interval duration, and the acceleration value is greater than the first acceleration threshold, are selected from the peaks in the acceleration signal as the target peaks.

[0067] In step S22, a valley with the minimum acceleration value and less than the first acceleration threshold is determined between each adjacent two target peaks as the target valley.

[0068] Further, before the step S21, the acceleration signal can be subjected to peak and valley detection to detect a plurality of to-be-screened peaks and a plurality of to-be-screened valleys. As to the peak and valley detection in the continuous signal, the existing conventional method can be used, which will not be described here.

[0069] After detecting the plurality of to-be-screened peaks and the plurality of to-be-screened valleys, the above steps S21 and S22 are executed, which can eliminate peaks (invalid peaks) with an interval duration less than the first interval duration or an acceleration value less than or equal to the first acceleration threshold from the to-be-screened peaks, and eliminate valleys (invalid valleys) with an acceleration value greater than the first acceleration threshold from the to-be-screened valleys. Thereby, the plurality of target peaks and the plurality of target valleys which can accurately reflect the actual acceleration and deceleration states in the motion process are screened out.

[0070] Referring to Figure 3 and Figure 4 , Figure 3 is a schematic diagram of determining target peaks from an acceleration signal in the embodiment of the present application, Figure 4is a schematic diagram of determining a target trough from an acceleration signal in an embodiment of the present application.

[0071] Without limitation, Figure 3 , Figure 4 The acceleration signal shown can be part of an acceleration signal of the evaluation object performing multiple sit-stand motions within the first motion duration, which can be obtained by performing a modulus operation on the collected original three-axis acceleration, and then performing noise reduction and smoothing processing. Wherein, the abscissa can represent time t (or sampling time), and the ordinate represents the actual acceleration value a (or can also represent the superposition value of the actual acceleration value a and the gravitational acceleration g). Figure 3 and Figure 4 The horizontal dashed line in each coordinate of and represents the first acceleration threshold.

[0072] In Figure 3 , the position marked by each small dot is the position of a plurality of target peaks determined from the acceleration signal.

[0073] In Figure 4 , the position of each small dot is the position of the trough with the minimum acceleration value between each adjacent two target peaks, which contains a plurality of troughs corresponding to non-motion time (such as stationary phase) between adjacent two sit-stand, i.e. invalid troughs.

[0074] In Figure 4 , each small dot is the position of a plurality of target troughs obtained by removing the troughs with an acceleration value greater than the first acceleration threshold from the left side of the troughs determined in the lower coordinate. The selected target troughs can more accurately reflect the maximum deceleration during the standing-up process in each sit-stand.

[0075] In the implementation of step S13, for each target trough, the previous target peak and the next target peak adjacent to the current target trough are determined, which are respectively denoted as the first target peak and the second target peak.

[0076] Taking sit-stand motion as an example, the time point to which the current target trough belongs can refer to the time point to which the maximum deceleration in the standing-up process of the current sit-stand belongs; the time point to which the previous target peak adjacent to the current target trough belongs can refer to the time point to which the maximum acceleration in the standing-up process of the current sit-stand belongs; and the time point to which the next target peak adjacent to the current target trough belongs can refer to the maximum deceleration in the sitting-down process of the current sit-stand.

[0077] In the implementation of step S14, for a single to-be-evaluated movement (e.g., a single sit-stand), if the time interval between the first target peak and the second target peak is too long (greater than the preset second interval length), it means that the evaluation subject is likely to not effectively perform the to-be-evaluated movement (e.g., distracted or paused for other reasons) in this interval length, and therefore the sampling data in this interval length should be discarded.

[0078] The second interval length is greater than or equal to the first interval length. In the implementation, the specific value of the second interval length can be set in combination with actual application. Without limitation, an appropriate value in [2.5s, 4s] can be selected as the second interval length. For example, the second interval length can be set to 3s.

[0079] In the embodiment of the present application, on the basis of discarding invalid peaks through the above step S12, through the preset second interval length, the target peak with an excessively long interval length can be further discarded from the determined multiple target peaks. Thereby, it is helpful to improve the accuracy of the start time and the end time of each to-be-evaluated movement determined subsequently, and further improve the accuracy of the endurance evaluation result.

[0080] Further, in the step S14, determining the start time of the current to-be-evaluated movement based on the first target peak includes: taking the time point of the previous trough adjacent to the first target peak, or the time point of the last acceleration value less than or equal to the average value of the acceleration signal before the first target peak, as the start time of the current to-be-evaluated movement.

[0081] Further, in the step S15, determining the end time of the current to-be-evaluated movement based on the second target peak includes: taking the time point of the next trough adjacent to the second target peak, or the time point of the first acceleration value less than or equal to the average value of the acceleration signal after the second target peak, as the end time of the current to-be-evaluated movement.

[0082] In the implementation, the combination of the start time and the end time of the current to-be-evaluated movement can be in the following four cases:

[0083] Case one: the start time of the current to-be-evaluated movement is the time point of the previous trough adjacent to the first target peak, and the end time of the current to-be-evaluated movement is the time point of the next trough adjacent to the second target peak;

[0084] Case two: the starting time of the current motion to be evaluated is the time point of the previous trough adjacent to the first target peak, and the ending time of the current motion to be evaluated is the time point of the first acceleration value less than or equal to the average value of the acceleration signal after the second target peak;

[0085] Case three: the starting time of the current motion to be evaluated is the time point of the last acceleration value less than or equal to the average value of the acceleration signal before the first target peak, and the ending time of the current motion to be evaluated is the time point of the next trough adjacent to the second target peak;

[0086] Case four: the starting time of the current motion to be evaluated is the time point of the last acceleration value less than or equal to the average value of the acceleration signal before the first target peak, and the ending time of the current motion to be evaluated is the time point of the first acceleration value less than or equal to the average value of the acceleration signal after the second target peak.

[0087] Referring to Figure 5 , Figure 5 is a schematic diagram of determining the starting time and the ending time of a single motion to be evaluated from an acceleration signal in an embodiment of the present application. Non-limitingly, Figure 5 The acceleration signal shown corresponds to a sit-to-stand motion.

[0088] wherein the horizontal coordinate can represent time t (or sampling time), and the vertical coordinate represents actual acceleration value a (or can also represent the superposition value of actual acceleration value and gravitational acceleration g), Figure 5 The horizontal dotted line in the coordinate shown represents the average value of the acceleration signal.

[0089] wherein in the current sit-to-stand, the current target trough is A1, the previous target peak adjacent to A1 is B1 (i.e., the first target peak), and the next target peak adjacent to A1 is C1 (i.e., the second target peak). In the previous adjacent sit-to-stand, the target trough is A2, the previous target peak adjacent to A2 is B2, and the next target peak adjacent to A2 is C2. In the next adjacent sit-to-stand, the target trough is A3, the previous target peak adjacent to A3 is B3, and the next target peak adjacent to A3 is C3.

[0090] Specifically, the time point t1 of A1 is the time point of the maximum deceleration in the process of standing up in the current sit-to-stand; the time point t2 of B1 is the time point of the maximum acceleration in the process of standing up in the current sit-to-stand; and the time point t3 of C1 is the time point of the maximum deceleration in the process of standing up in the current sit-to-stand.

[0091] Similarly, the time t4 to which A2 belongs is the time at which the maximum deceleration in the standing-up process in the previous sitting-stand motion belongs; the time t5 to which B2 belongs is the time at which the maximum acceleration in the standing-up process in the previous sitting-stand motion belongs; and the time t6 to which C2 belongs is the time at which the maximum deceleration in the standing-up process in the previous sitting-stand motion belongs.

[0092] Similarly, the time t7 to which A3 belongs is the time at which the maximum deceleration in the standing-up process in the subsequent sitting-stand motion belongs; the time t8 to which B3 belongs is the time at which the maximum acceleration in the standing-up process in the subsequent sitting-stand motion belongs; and the time t9 to which C3 belongs is the time at which the maximum deceleration in the standing-up process in the subsequent sitting-stand motion belongs.

[0093] Based on the motion principle of the sitting-stand motion itself and Figure 5 Based on the change trend of the acceleration signal shown in FIG. 6, it can be understood that the starting time of the current sitting-stand motion should be located between t6 and t2, and the time interval includes the non-motion time (for example, the stationary phase) between the end of the previous sitting-stand motion and the start of the current sitting-stand motion; and the ending time of the current sitting-stand motion should be located between t3 and t8, and the time interval includes the non-motion time (for example, the stationary phase) between the end of the current sitting-stand motion and the start of the subsequent sitting-stand motion.

[0094] Specifically, the scheme for determining the starting time of the current sitting-stand motion can include the following process: detecting forward from the first target peak B1, when a trough D1 is detected, the time to which the detected trough D1 belongs is taken as the starting time of the current sitting-stand motion, or when an acceleration value less than or equal to the average value of the acceleration signal is detected, the time to which the detected acceleration value belongs is taken as the starting time of the current sitting-stand motion. (For example, Figure 5 In the waveform diagram shown in FIG. 6, the time T1 to which the trough D1 belongs can be taken as the starting time of the current sitting-stand motion.

[0095] Correspondingly, the scheme for determining the ending time of the current sitting-stand motion can include the following process: detecting backward from the second target peak C1, when a trough is detected, the time to which the detected trough belongs is taken as the starting time of the current sitting-stand motion, or when an acceleration value less than or equal to the average value of the acceleration signal is detected (corresponding to the position D2 in FIG. 6), the time to which D2 belongs is taken as the ending time of the current sitting-stand motion. (For example, Figure 5 In the waveform diagram shown in FIG. 6, the time T2 to which D2 belongs can be taken as the ending time of the current sitting-stand motion. Figure 5 In the waveform diagram shown in FIG. 6, the time T2 to which D2 belongs can be taken as the ending time of the current sitting-stand motion.

[0096] In actual applications, because there is usually a certain non-motion time length (which can also be referred to as an interference time length) between adjacent two motions, such as Figure 5In the illustrated embodiment, between t6 and t2, the time interval includes the non-movement time (e.g., the stationary phase) between the end of the previous sitting and the start of the current sitting; and between t3 and t8, the time interval includes the non-movement time (e.g., the stationary phase) between the end of the current sitting and the start of the next sitting. Therefore, in the embodiment of the present application, the start time and the end time of each of the to-be-evaluated movements are determined by using the above scheme, the interference time before the start of the current movement and after the end of the current movement can be removed, so as to obtain effective "dehydration" data, and improve the accuracy and reliability of the subsequent determined endurance index.

[0097] In the implementation of step S15, the endurance index of the evaluation object in the to-be-evaluated movement is determined according to the start time and the end time of each of the to-be-evaluated movements.

[0098] Without limitation, the endurance index of the evaluation object in the to-be-evaluated movement can include one or more of the following: the average movement time, the median movement time, the variability parameter, the single longest movement time, the single shortest movement time, the movement time skewness, the movement time kurtosis, the movement time interquartile range (IQR), the movement time mean absolute deviation, the movement time root mean square, the movement time standard deviation, the movement time variance, the movement time empirical distribution function percentile count, and the movement time empirical distribution function slope of the evaluation object in the to-be-evaluated movement within the first movement time.

[0099] In the embodiment of the present application, after the start time and the end time of each of the to-be-evaluated movements of the evaluation object are accurately determined, the movement time of each of the to-be-evaluated movements can be determined based on the difference between the end time and the start time of each of the to-be-evaluated movements; then, the movement time of each of the to-be-evaluated movements is used to perform different mathematical operations or function fitting, to obtain one or more of the above-mentioned endurance indexes, or other parameters as endurance indexes, so as to help to finely evaluate the movement endurance of the evaluation object from multiple dimensions.

[0100] Reference Figure 6 , Figure 6 is Figure 1A flow chart of one specific implementation of step S15. In this embodiment, the endurance index is a variability parameter of each of the to-be-evaluated movements performed by the evaluation subject in the first movement duration. Step S15 can specifically include steps S61-S63.

[0101] In step S61, the actual movement duration of each of the to-be-evaluated movements is determined according to the start time and the end time of each of the to-be-evaluated movements.

[0102] Specifically, for each of the to-be-evaluated movements, the actual movement duration of the to-be-evaluated movement performed by the evaluation subject can be determined according to the difference between the end time and the start time of the to-be-evaluated movement. The actual movement duration is the duration obtained after removing the non-movement duration (or interference duration) between adjacent movements, e.g., the duration corresponding to the stationary phase, which is needed to represent the actual movement process of each of the to-be-evaluated movements.

[0103] In step S62, the average movement duration and the movement duration standard deviation of each of the to-be-evaluated movements are determined according to the actual movement duration of each of the to-be-evaluated movements.

[0104] In step S63, the average movement duration and the movement duration standard deviation of each of the to-be-evaluated movements are divided to obtain the variability parameter of each of the to-be-evaluated movements performed by the evaluation subject.

[0105] Specifically, the variability parameter can be determined using the following formula:

[0106]

[0107] wherein c v is used to indicate the variability parameter, δ is used to indicate the average movement duration of each of the to-be-evaluated movements, and μ is used to indicate the movement duration standard deviation of each of the to-be-evaluated movements.

[0108] The variability parameter can represent the discrete degree of the movement duration of each of the to-be-evaluated movements performed by the evaluation subject and is one of the important parameters of endurance evaluation. The greater the value of the variability parameter, the greater the discrete degree, the more unstable the movement duration of each of the to-be-evaluated movements, which means that the muscle endurance or control ability of the evaluation part of the evaluation subject is worse. Conversely, the smaller the value of the variability parameter, the smaller the discrete degree, the more stable the movement duration of each of the to-be-evaluated movements, which means that the muscle endurance or control ability of the evaluation part of the evaluation subject is better.

[0109] Further, after determining the endurance index of the evaluation object performing the to-be-evaluated exercise, the method further comprises: inputting the endurance index into a pre-trained exercise endurance prediction model to determine an exercise endurance level of the evaluation object; wherein the pre-trained exercise endurance prediction model is obtained by training a machine learning model using endurance indexes of a plurality of sample objects performing the to-be-evaluated exercise as training sample data.

[0110] In a specific embodiment, the training sample data of the pre-trained exercise endurance prediction model further comprises first annotation data, and the first annotation data comprises walking distances of the plurality of sample objects in a preset second exercise duration; when determining the exercise endurance level of the evaluation object, a corresponding walking distance is determined.

[0111] In a specific embodiment, the second exercise duration can be 6 minutes. For example, walking distances of a plurality of sample objects in a 6-minute walking test (6MWT) are used as the annotation data. The 6-minute walking distance is an important index in the field of exercise endurance evaluation.

[0112] In another specific embodiment, the training sample data of the pre-trained exercise endurance prediction model further comprises second annotation data, and the second annotation data comprises cardiopulmonary exercise test indexes of the plurality of sample objects; when determining the exercise endurance level of the evaluation object, a corresponding cardiopulmonary exercise test index is determined.

[0113] Specifically, after a plurality of sample objects in a preset second exercise duration (i.e., the first annotation data) and / or cardiopulmonary exercise test (CPET) indexes of the plurality of sample objects (i.e., the second annotation data) are annotated, the annotated training sample data is input into the machine learning model for training to obtain the exercise endurance prediction model.

[0114] In some non-limiting embodiments, the machine learning model can be selected from a multiple linear regression model, a random forest model, and a Generalized Additive Model (GAM), a spline model, and other non-linear models. In the embodiments of the present application, on the one hand, by converting the exercise endurance indicators that can contain multiple data into corresponding exercise endurance grades, walking distances of the second exercise duration, or cardiopulmonary exercise test indicators, the intuitive feeling of the evaluation results can be enhanced, and the user experience can be improved. On the other hand, in specific application scenarios, such as in the walking competitive training of athletes, by obtaining the walking distances in different training stages, the training effects of each stage can be more intuitively and quickly analyzed and monitored, and the training scheme can be further improved in a targeted manner. Or, in the exercise endurance evaluation scenario of the elderly population, not only the intuitive exercise endurance grades can be obtained, but also the cardiopulmonary exercise test indicators of the evaluation object can be obtained intuitively, and the cardiopulmonary function can be analyzed and monitored in combination with other parameters.

[0115] Reference Figure 7 , Figure 7 is a structural schematic diagram of an endurance evaluation device in an embodiment of the present application. The device can include:

[0116] An acceleration signal determination module 71 is configured to determine acceleration signals of the evaluation object performing a plurality of to-be-evaluated exercises within a preset first exercise duration.

[0117] A peak-valley determination module 72 is configured to determine a plurality of target peaks and a plurality of target valleys from the acceleration signals based on a preset first interval duration and a first acceleration threshold.

[0118] A target peak determination module 73 is configured to, for each target valley, determine a previous target peak and a next target peak adjacent to the current target valley, and record the previous target peak and the next target peak as a first target peak and a second target peak, respectively.

[0119] An exercise time determination module 74 is configured to, if a time interval between the first target peak and the second target peak is less than or equal to a preset second interval duration, determine a start time of the current to-be-evaluated exercise based on the first target peak, and determine an end time of the current to-be-evaluated exercise based on the second target peak.

[0120] An endurance indicator determination module 75 is configured to determine an endurance indicator of the evaluation object performing the to-be-evaluated exercises according to the start time and the end time of each to-be-evaluated exercise.

[0121] The first interval duration is less than or equal to the second interval duration.

[0122] For the principle, specific implementation and beneficial effects of the endurance evaluation device, please refer to the foregoing and Figures 1 to 6 The related description about the endurance evaluation method is shown, and will not be repeated here.

[0123] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is run by a processor to execute the steps of the endurance evaluation method. Figures 1 to 6 The computer readable storage medium can include a non-volatile memory or a non-transitory memory, and can also include an optical disc, a mechanical hard disk, a solid state disk, etc.

[0124] Specifically, in the embodiment of the present application, the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or can also be any conventional processor, etc.

[0125] It should also be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, but not by way of limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchlink dynamic random access memory (SLDRAM) and direct rambus RAM (DR RAM).

[0126] The embodiments of the present application also provide a terminal, including a memory and a processor, the memory stores a computer program capable of running on the processor, and the processor executes the computer program to perform the steps of the endurance evaluation method shown above. Figures 1 to 6 The terminal can include, but is not limited to, a mobile phone, a computer, a tablet computer and other terminal devices, and can also be a server, a cloud platform and the like.

[0127] It should be understood that the term "and / or" herein only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein represents that the front and rear associated objects are in an "or" relationship.

[0128] The "multiple" appearing in the embodiments of the present application means two or more.

[0129] The first, second and the like appearing in the embodiments of the present application are only for indicating and distinguishing the described objects, and do not have sequence, and do not represent the special limitation of the number of devices in the embodiments of the present application, and cannot constitute any limitation on the embodiments of the present application.

[0130] It should be noted that the serial numbers of the steps in the embodiments do not represent the limitation of the execution sequence of the steps.

[0131] Although the present application is disclosed as above, the present application is not limited to this. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, and therefore the protection scope of the present application should be subject to the range defined by the claims.

Claims

1. A method of endurance assessment, characterized by, The method comprises the following steps: determining acceleration signals of the evaluation object performing a plurality of to-be-evaluated motions within a preset first motion duration; determining a plurality of target peaks and a plurality of target troughs from the acceleration signals based on a preset first interval duration and a first acceleration threshold, comprising: selecting, from the peaks of the acceleration signals, peaks that satisfy that the interval duration of every two adjacent peaks is greater than or equal to the first interval duration and the acceleration value is greater than the first acceleration threshold, as the target peaks; determining, between every two adjacent target peaks, a trough with the minimum acceleration value and less than the first acceleration threshold, as the target trough; for each target trough, determining a previous target peak and a subsequent target peak adjacent to the current target trough, and denoting the previous target peak and the subsequent target peak as a first target peak and a second target peak, respectively; if the time interval of the first target peak and the second target peak is less than or equal to a preset second interval duration, determining the starting time of the current to-be-evaluated motion based on the first target peak and determining the ending time of the current to-be-evaluated motion based on the second target peak; determining a tolerance index of the evaluation object performing the to-be-evaluated motion according to the starting time and the ending time of each to-be-evaluated motion; wherein the first interval duration is less than or equal to the second interval duration.

2. The method of claim 1, wherein, The first acceleration threshold is obtained by performing weighted operation on the average value and the standard deviation of the acceleration signals with a preset weight ratio.

3. The method of claim 1, wherein, The determination of the starting time of the current to-be-evaluated motion based on the first target peak comprises: taking the time point of the previous trough adjacent to the first target peak or the time point of the last acceleration value less than or equal to the average value of the acceleration signals before the first target peak as the starting time of the current to-be-evaluated motion.

4. The method of claim 1, wherein, The determination of the ending time of the current to-be-evaluated motion based on the second target peak comprises: taking the time point of the subsequent trough adjacent to the second target peak or the time point of the first acceleration value less than or equal to the average value of the acceleration signals after the second target peak as the ending time of the current to-be-evaluated motion.

5. The method of claim 1, wherein, The tolerance index of the evaluation object performing the to-be-evaluated motion comprises one or more of the following: the average motion duration, the median motion duration, the variability parameter, the single longest motion duration, the single shortest motion duration, the motion duration skewness, the motion duration kurtosis, the motion duration quartile range, the motion duration median absolute error, the motion duration root mean square, the motion duration standard deviation, the motion duration variance, the motion duration empirical distribution function percentile, and the motion duration empirical distribution function slope of each to-be-evaluated motion of the evaluation object within the first motion duration.

6. The method of claim 1, wherein, The tolerance index is the variability parameter of each to-be-evaluated motion of the evaluation object within the first motion duration. The determination of the tolerance index of the evaluation object performing the to-be-evaluated motion according to the starting time and the ending time of each to-be-evaluated motion comprises: determining the actual motion duration of each to-be-evaluated motion according to the starting time and the ending time of each to-be-evaluated motion; According to the actual movement duration of each time of the to-be-evaluated movement, the average movement duration and the movement duration standard deviation of each time of the to-be-evaluated movement are determined; The average movement duration and the movement duration standard deviation of each time of the to-be-evaluated movement are divided to obtain the variability parameter of the evaluation object performing each time of the to-be-evaluated movement.

7. The method of claim 1, wherein, The method for determining the acceleration signal of the evaluation object performing a plurality of to-be-evaluated movements within a preset first movement duration comprises the following steps: Collecting the three-axis acceleration signals of the evaluation object performing the plurality of to-be-evaluated movements within the first movement duration; Performing a modulus operation on the three-axis acceleration signals to determine the acceleration signal.

8. The method according to claim 1 or 7, characterized in that, After the acceleration signal is determined, the method further comprises the following steps: Performing noise reduction processing and smoothing processing on the acceleration signal.

9. The method of claim 1, wherein, After the endurance index of the evaluation object performing the to-be-evaluated movement is determined, the method further comprises the following steps: Inputting the endurance index into a pre-trained movement endurance prediction model to determine the movement endurance grade of the evaluation object; The pre-trained movement endurance prediction model is obtained by training a machine learning model using the endurance indexes of a plurality of sample objects performing the to-be-evaluated movement as training sample data.

10. The method of claim 9, wherein, The training sample data of the pre-trained movement endurance prediction model further comprises first annotation data, and the first annotation data comprises the walking distances of the plurality of sample objects within a preset second movement duration; When the movement endurance grade of the evaluation object is determined, the corresponding walking distance is also determined.

11. The method according to claim 9 or 10, characterized in that, The training sample data of the pre-trained movement endurance prediction model further comprises second annotation data, and the second annotation data comprises the cardiopulmonary exercise test indexes of the plurality of sample objects; When the movement endurance grade of the evaluation object is determined, the corresponding cardiopulmonary exercise test index is also determined.

12. The method of claim 1, wherein, The endurance index is a lower limb movement endurance index; The to-be-evaluated movement is selected from the following: sit-to-stand movement, deep squat movement, squat jump movement, and vertical jump movement.

13. A tolerance assessment device, characterized by Comprise: An acceleration signal determination module for determining the acceleration signal of the evaluation object performing a plurality of to-be-evaluated movements within a preset first movement duration; A wave peak and wave trough determination module for determining a plurality of target wave peaks and a plurality of target wave troughs from the acceleration signal based on a preset first interval duration and a first acceleration threshold, comprising: selecting wave peaks that satisfy the interval duration of each adjacent two wave peaks being greater than or equal to the first interval duration and the acceleration value being greater than the first acceleration threshold from the wave peaks of the acceleration signal as the target wave peaks; determining a wave trough with the minimum acceleration value and less than the first acceleration threshold between each adjacent two target wave peaks as the target wave trough; A target wave peak determination module for determining, for each target wave trough, a previous target wave peak and a subsequent target wave peak adjacent to the current target wave trough, respectively denoted as a first target wave peak and a second target wave peak; a start time and end time determination module, configured to determine a start time of a current motion to be evaluated based on the first target peak if a time interval between the first target peak and the second target peak is less than or equal to a preset second interval duration, and determine an end time of the current motion to be evaluated based on the second target peak; a tolerance index determination module, configured to determine a tolerance index of the evaluation object in the motion to be evaluated according to the start time and the end time of each motion to be evaluated; wherein the first interval duration is less than or equal to the second interval duration.

14. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is run by the processor to execute the steps of the tolerance evaluation method in any one of claims 1 to 12.

15. A terminal comprising a memory and a processor, said memory having stored thereon a computer program capable of running on said processor, characterized in that, The processor runs the computer program to execute the steps of the tolerance evaluation method in any one of claims 1 to 12.

Images