Method for deducing maximum heart rate based on motion data

Abstract

The invention relates to a method for deducing the maximum heart rate based on exercise data. The method comprises the following steps that the exercise data of participants is acquired, collected and subpackaged; calculating a motion ability value through a relation between time and speed; the maximum heart rate value range is deduced through the relation between time and the heart rate; accurate calculation of the maximum heart rate is completed through the matching relation of the exercise ability values. According to the method, a topological graph relation is formed by constructing an intermediate index, namely an exercise ability value RQ, and heart rate and speed indexes collected by an exercise record, so that inference of the maximum heart rate is completed; compared with a maximum heart rate test in a laboratory environment, the method has the advantages that the testee can predict the maximum heart rate only through a common sports watch capable of collecting the heart rate and the speed through the algorithm, the cost is relatively low, the error between the calculated result and the maximum heart rate test result of the testee is within +-2bpm, and the accuracy is high.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a method for inferring maximum heart rate based on motion data. Background Art

[0002] As a commonly used concept in the field of sports science, maximum heart rate refers to the highest number of times the heart can beat per minute when the human body is performing extreme exercise, usually measured in beats per minute (bpm). Maximum heart rate has a wide range of applications in sports monitoring. During exercise, heart rate is closely related to exercise intensity. Based on the maximum heart rate, exercise intensity intervals can be accurately divided. For example, when the exercise heart rate is between 60% - 70% of the maximum heart rate, it belongs to low-intensity exercise, suitable for daily fitness, rehabilitation training, etc., which can effectively improve endurance, enhance cardiopulmonary function and is not easily fatigued; when the exercise heart rate is between 70% - 85%, it is medium-intensity exercise, which can efficiently improve aerobic metabolism ability and promote fat burning; when the exercise heart rate reaches 85% - 95%, it is high-intensity exercise, mainly developing anaerobic metabolism ability, improving speed and strength, but the duration should not be too long, otherwise it is easy to cause fatigue and injury risks. Reasonably controlling exercise intensity and avoiding overtraining or under-training, maximum heart rate plays an indispensable guiding role.

[0003] Currently, there are various scientific methods for measuring maximum heart rate. Among them, the common one is the incremental exercise test. The subject exercises on a treadmill or a stationary bike and gradually increases the exercise intensity according to a preset program. At the same time, a professional electrocardiogram monitoring device is used to record the electrocardiogram changes in real time. As the exercise intensity continues to increase, when the subject reaches the state of exhaustion, the heart rate monitored at this time is the maximum heart rate; another method is to use a cardiopulmonary function tester to accurately measure indicators such as oxygen uptake and carbon dioxide output of the subject during exercise, and combine with heart rate changes to calculate the maximum heart rate. Although these laboratory methods have high accuracy, they require professional equipment and personnel to operate, with high costs, and have certain requirements for the physical conditions of the subjects, making it difficult to popularize on a large scale.

[0004] At the level of the general public, with the popularization of wearable devices, the heart rate during exercise can be collected in real time through the sensors of these devices. However, to measure the maximum heart rate through wearable devices, the subject needs to perform a specific paradigm of exercise and reach the state of physical exhaustion. It is difficult for the general population without exercise habits to complete it alone and there is a certain health risk; at the same time, because the time that the maximum heart rate can be maintained is very short (within 10 seconds), the data transmission accuracy of wearable devices in a short time is also limited by the hardware itself. For these two reasons, it is very difficult for the general population to directly measure the maximum heart rate through wearable devices, and there is a need to develop a new data inference method to calculate the maximum heart rate. Summary of the Invention

[0005] In order to overcome the disadvantages and deficiencies existing in the prior art, the purpose of the present invention is to provide a method for inferring the maximum heart rate based on motion data. The present invention completes the inference of the maximum heart rate by constructing an intermediate index, namely the running quotient (RQ), to form a topological graph relationship with the heart rate and speed indexes collected from the motion records.

[0006] The purpose of the present invention is achieved by the following technical solutions: A method for inferring the maximum heart rate based on motion data, comprising the following steps:

[0007] S1. Obtain the collection and packaging of the motion data of the participant;

[0008] S2. Calculate the motion ability value through the relationship between time and speed;

[0009] S3. Infer the value range of the maximum heart rate through the relationship between time and heart rate;

[0010] S4. Complete the exact calculation of the maximum heart rate through the matching relationship of the motion ability value.

[0011] The above method adopted by the present invention is based on the theory of sports science, uses the user's historical training data, and infers the maximum heart rate in the way of mathematical modeling. The subject does not need to connect high-precision equipment in the laboratory environment, nor does it need to perform exhaustive exercise to make the heart load reach the critical point. As long as enough medium- and high-intensity training data is accumulated, a reasonable estimated value of the maximum heart rate can be obtained to guide their daily sports training.

[0012] In addition, the theory of sports science believes that the maximum intensity that the human body can maintain during exercise gradually decreases as the exercise time prolongs; speed and heart rate are common objective indicators to measure exercise intensity. For people with higher exercise ability, their performance in the above values can be summarized in the following two aspects: within the same time, they can maintain a higher-intensity exercise output (absolute value, speed); under the same-intensity exercise output, the internal load of the body is relatively low (relative value, percentage of maximum heart rate).

[0013] Based on the above idea, the present invention completes the inference of the maximum heart rate by constructing an intermediate index, namely the running quotient (RQ), to form a topological graph relationship with the heart rate and speed indexes collected from the motion records. Compared with the maximum heart rate test in the laboratory environment, through the algorithm of the present invention, the subject only needs an ordinary sports watch that can collect heart rate and speed to perform the maximum heart rate prediction, and the cost is relatively low; the error between the result calculated by the present invention and the result of the subject's maximum heart rate test is within ±2 bpm, which can meet the needs of the sports population and has high accuracy; the present invention takes parameters through modeling with the user's actual historical data, and has better stability compared with the single direct measurement result.

[0014] Preferably, in step S1, the acquisition of the participant's motion data collection and packaging: after the historical motion data of the subject is input, through the computer system, according to the preset duration standards T1, T 2 ...T n is packaged to obtain the input data set S:

[0015] S T1 :s 11 ,st 12 ,st 1j ...st 1m ;

[0016] S T2 :st 21 ,st 22 ,st 2j ...st 2m ;

[0017] S Ti :st i1 ,st i2 ,st ij ...st im ;

[0018] S Tn :st n1 ,st n2 ,st nj ...st nm ; where st ij represents the jth motion record with a duration of Ti, and each record contains the average speed and the lowest heart rate during this period; V represents the speed information in the data set S, and vt ij represents the speed information in the record st ij ; H represents the heart rate information in the data set S, and ht ij represents the heart rate information in the record st ij ; at the same time, according to other relevant information of the motion record, a weight value W is assigned, and wt ij represents the weight in the record st ij : the farther the motion record is from the calculation time, the lower its weight value; when the motion record is marked as a test / match by the user, its weight will be increased.

[0019] Preferably, in step S2, the calculation of the motion ability value is carried out through the relationship between time and speed: for the speed information V in the data set, the outliers are excluded by statistical methods, and the maximum value set V of each duration is taken max : vmaxt 1 ,vmaxt 2 ...vmaxt n ; Since the motion ability RQ, duration Ti , T i The maximum speed vmaxt that can be maintained within a certain time i There is the following mathematical relationship among the three:

[0020]

[0021] Among them, Dc is a constant, and vc is a linear equation of RQ; input V max , T, and W arrays, and the weighted numerical distribution RQ of the motor ability can be obtained vmax .

[0022] Preferably, in step S3, the inference of the maximum heart rate value range is carried out through the relationship between time and heart rate: for the heart rate information H in the dataset, the outliers are excluded through statistics, and the maximum value set H of each duration is obtained by the method max : hmaxt 1 , hmaxt 2 ...hmaxt n ; Let hmaxt i % be the percentage of the maximum heart rate that the human body can maintain within time T i to the actual maximum heart rate; in this scheme, from the massive data, the hmaxt% corresponding to several fixed time periods (180 / 300 / 600 / 1200 / 1800 seconds) are respectively subject to the following normal distributions:

[0023] hmaxt 180 % ~ N(0.964, 0.008)

[0024] hmaxt 300 % ~ N(0.956, 0.011)

[0025] hmaxt 600 % ~ N(0.944, 0.010)

[0026] hmaxt 1200 % ~ N(0.928, 0.017)

[0027] hmaxt 1800 % ~ N(0.919, 0.019)

[0028] Therefore, input H max , T, and the reasonable estimation range of the maximum heart rate can be obtained, and its upper and lower bounds (2.5% quantile and 97.5% quantile) are respectively denoted as hmax lower , hmax upper .

[0029] Preferably, in step S4, the exact calculation of the maximum heart rate is completed through the matching relationship of the exercise ability value: First, according to the estimated maximum heart rate range in step S3, an array Hmax: hmax is generated lower , hmax lower +1, hmax lower +2...hmax upper ;

[0030] Since there is a general mathematical relationship among RQ, V, and H% = H / Hmax, given a specific maximum heart rate value Hmax k , the corresponding exercise ability inference RQ can be obtained k ; Let RQ k and the RQ obtained in step S2 vma are both estimates of the subject's exercise ability. The relative entropy KL(RQ vmax ||RQ k ) represents the difference between the two numerical distributions:

[0031]

[0032] Therefore, there is a maximum heart rate value that minimizes KL(RQ vmax ||RQ k ), and this value is the reasonable estimate of the maximum heart rate.

[0033] The beneficial effects of the present invention are as follows:

[0034] 1. Compared with the maximum heart rate test in a laboratory environment, through the algorithm of the present invention, the subject only needs an ordinary sports watch that can collect heart rate and speed to perform maximum heart rate prediction, and the test cost is low;

[0035] 2. The error between the result calculated by the present invention and the result of the subject's maximum heart rate test is within ±2 bpm, which can meet the needs of the sports population and has high accuracy;

[0036] 3. The present invention models and takes parameters through the actual historical data of the user, and has better stability compared with the single direct measurement result. Description of the Drawings

[0037] Figure 1 is the numerical relationship topology diagram of the present invention;

[0038] Figure 2 is the numerical relationship diagram of exercise ability, time, and maximum speed in the present invention;

[0039] Figure 3 is the numerical relationship diagram of exercise ability, maximum heart rate%, and speed in the present invention;

[0040] Figure 4It is a schematic diagram of the algorithm flow of the present invention. Specific Embodiments

[0041] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with embodiments and the appended Figures 1-4 drawings. The content mentioned in the embodiments does not limit the present invention.

[0042] It should be noted that when an element is referred to as "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element.

[0043] When an element is referred to as "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element.

[0044] It should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present application.

[0045] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features.

[0046] In the description of the present application, "a plurality" means two or more, unless otherwise specifically defined.

[0047] See Figures 1-4 , a method for inferring the maximum heart rate based on motion data, comprising the following steps:

[0048] S1. Obtain the collection and packaging of the motion data of the participant;

[0049] S2. Calculate the motion ability value through the relationship between time and speed;

[0050] S3. Infer the value range of the maximum heart rate through the relationship between time and heart rate;

[0051] S4. Complete the exact calculation of the maximum heart rate through the matching relationship of the motion ability value.

[0052] The above method adopted by the present invention is based on the theory of sports science. By using the historical training data of users, the maximum heart rate is inferred in the way of mathematical modeling. The subject does not need to connect high-precision equipment in the laboratory environment, nor does he need to perform exhaustive exercise to make the heart load reach the critical point. As long as enough medium- and high-intensity training data is accumulated, a reasonable maximum heart rate estimate can be obtained to guide his daily sports training.

[0053] The present invention completes the inference of the maximum heart rate by constructing an intermediate index, that is, the exercise ability value RQ, to form a topological graph relationship with the heart rate and speed indexes collected by the exercise record. Compared with the maximum heart rate test in the laboratory environment, through the algorithm of the present invention, the subject only needs an ordinary sports watch that can collect heart rate and speed to perform the maximum heart rate prediction, and the cost is relatively low; the result calculated by the present invention has an error within ±2 bpm compared with the result of the subject's maximum heart rate test, which can meet the needs of the sports population and has high accuracy; the present invention models and takes parameters through the actual historical data of users, and has better stability compared with the single direct measurement result.

[0054] In step S1, collect and package the exercise data of the participant: After the historical exercise data of the subject is input, through the computer system, according to the preset duration standards T1, T 2 ...T n perform packaging to obtain the input data set S:

[0055] S T1 :s 11 ,st 12 ,st 1j ...st 1m ;

[0056] S T2 :st 21 ,st 22 ,st 2j ...st 2m ;

[0057] S Ti :st i1 ,st i2 ,st ij ...st im ;

[0058] S Tn :st n1 ,st n2 ,st nj ...st nm ; where st ij represents the jth exercise record with a duration of Ti, and each record contains the average speed and the lowest heart rate during this period; V represents the speed information in the data set S, vtij Represents the speed information in record st ij ; H represents the heart rate information in dataset S, ht ij Represents the heart rate information in record st ij ; Meanwhile, according to other relevant information of the exercise record, a weight value W, wt ij is assigned to represent the weight in record st ij : The older the exercise record is from the calculation time, the lower its weight value; when the exercise record is marked as a test / match by the user, its weight will be increased.

[0059] In step S2, the exercise ability value is calculated through the relationship between time and speed: for the speed information V in the dataset, outliers are excluded through statistical methods, and the maximum value set V max of each duration is taken: vmaxt 1 , vmaxt 2 ... vmaxt n ; Since there is the following mathematical relationship among the exercise ability RQ, duration T i , T i and the maximum speed vmaxt that can be maintained within the time i :

[0060]

[0061] where Dc is a constant and vc is a linear equation of the first order of RQ; by inputting the three arrays of V max , T, and W, the weighted numerical distribution RQ of the exercise ability can be obtained vmax .

[0062] In step S3, the inference of the maximum heart rate value range is carried out through the relationship between time and heart rate: for the heart rate information H in the dataset, outliers are excluded through statistics, and the maximum value set H max of each duration is taken: hmaxt 1 , hmaxt 2 ... hmaxt n ; Let hmaxt i % be the percentage of the maximum heart rate that the human body can maintain within the time of T i to the actual maximum heart rate; in this scheme, from the massive data, the hmaxt% corresponding to several fixed time periods (180 / 300 / 600 / 1200 / 1800 seconds) respectively follow the following normal distributions:

[0063] hmaxt 180 % ~ N(0.964, 0.008)

[0064] hmaxt 300% to N(0.956, 0.011)

[0065] hmaxt 600 % to N(0.944, 0.010)

[0066] hmaxt 1200 % to N(0.928, 0.017)

[0067] hmaxt 1800 % to N(0.919, 0.019)

[0068] Therefore, for the input H max , T can obtain a reasonable estimated range of the maximum heart rate, and its upper and lower bounds (2.5% percentile and 97.5% percentile) are respectively denoted as hmax lower , hmax upper .

[0069] In step S4, the exact calculation of the maximum heart rate is completed through the matching relationship of the exercise ability value: First, according to the estimated range of the maximum heart rate in step S3, an array Hmax is generated: hmax lower , hmax lower + 1, hmax lower + 2... hmax upper ;

[0070] Since there is a general mathematical relationship among RQ, V, and H% = H / Hmax, given a specific maximum heart rate value Hmax k , the corresponding exercise ability inference RQ k can be obtained; comparing RQ k with the RQ obtained in step S2 vma , both are estimates of the subject's exercise ability. Using the relative entropy KL(RQ vmax || RQ k ) to represent the difference between the two numerical distributions:

[0071]

[0072] Therefore, there exists a maximum heart rate value that minimizes KL(RQ vmax || RQ k ), and this value is the reasonable estimate of the maximum heart rate.

[0073] The above embodiments are preferred implementation schemes of the present invention. In addition, the present invention can also be implemented in other ways. Any obvious replacement without departing from the concept of the present invention is within the protection scope of the present invention.

Claims

1. A method for inferring maximum heart rate based on exercise data, characterized in that: The steps include: S1. Obtaining and packaging the participants’ exercise data; S2, calculating the athletic ability value through the relationship between time and speed; S3, inferring the maximum heart rate value range through the relationship between time and heart rate; S4. The maximum heart rate is accurately calculated through the matching relationship of the athletic ability values.

2. The method for inferring maximum heart rate based on exercise data according to claim 1, characterized in that: In step S1, the exercise data of the participants is collected and packaged: after the historical exercise data of the subjects is input, the computer system calculates the exercise data according to the preset time standards T1, T2...T n Perform subpackaging to obtain the input data set S: S T1 :s 11 ,st 12 ,st 1j ...st 1m ; S T2 :st 21 ,st 22 ,st 2j ...st 2m ; S Ti :st i1 ,st i2 ,st ij ...st im ; S Tn :st n1 ,st n2 ,st nj ...st nm ; Among them, st ij represents the jth exercise record of duration Ti, each record contains the average speed and minimum heart rate during this period; V represents the speed information in the data set S, vt ij Indicates record st ij H represents the heart rate information in the dataset S, ht ij Indicates record st ij At the same time, according to other relevant information of the exercise record, a weight value W, wt ij Indicates record st ij The weight in .

3. The method for inferring maximum heart rate based on sports data according to claim 1, characterized in that: In step S2, the athletic ability value is calculated based on the relationship between time and speed: for the speed information V in the data set, outliers are excluded by statistical methods, and the maximum value set V of each duration is taken. max :vmaxt1,vmaxt2...vmaxt n ; Due to the athletic ability RQ, duration T i 、T i The maximum speed that can be maintained within a certain time vmaxt i There is the following mathematical relationship between the three: Where Dc is a constant, vc is a linear equation of RQ; input V max , T, W three arrays, we can get the weighted numerical distribution RQ of athletic ability vmax .

4. The method for inferring maximum heart rate based on exercise data according to claim 1, characterized in that: In step S3, the maximum heart rate range is inferred by the relationship between time and heart rate: for the speed information V in the data set, outliers are excluded by statistics to obtain the maximum value set H of each duration max :hmaxt1,hmaxt2...hmaxt n ; Due to the duration T i 、T i The maximum heart rate percentage that can be maintained within a certain period of time hmax%t i , there is a general mathematical relationship, and hmax%t i =hmaxt i / hmax, hmax is the subject’s maximum heart rate, input H max , T can get a reasonable estimation range of the maximum heart rate, and its upper and lower bounds are recorded as hmax lower , hmax upper .

5. The method for inferring maximum heart rate based on exercise data according to claim 1, characterized in that: In step S4, the maximum heart rate is calculated accurately through the matching relationship of the athletic ability value: first, according to the maximum heart rate range estimation in step S3, an array Hmax is generated: hmax lower , hmax lower +1,hmax lower +2...hmax upper ; Since there is a general mathematical relationship between RQ, V, and H% = H / Hmax, given a specific maximum heart rate value Hmax k , we can get the corresponding sports ability inference RQ k ; RQ k and RQ obtained in step S2 vma are all estimates of the subjects' athletic ability, using relative entropy KL (RQ vmax ||RQ k ) represents the difference between two numerical distributions: Therefore, there is a maximum heart rate value such that KL(RQ vmax ||RQ k ) is the smallest, which is a reasonable estimate of the maximum heart rate.

Images