CN116407088A - Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle - Google Patents
Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle Download PDFInfo
- Publication number
- CN116407088A CN116407088A CN202310187037.9A CN202310187037A CN116407088A CN 116407088 A CN116407088 A CN 116407088A CN 202310187037 A CN202310187037 A CN 202310187037A CN 116407088 A CN116407088 A CN 116407088A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- user
- data set
- exercise
- power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 81
- 238000012549 training Methods 0.000 title claims abstract description 46
- 230000004044 response Effects 0.000 claims abstract description 116
- 238000009532 heart rate measurement Methods 0.000 claims abstract description 71
- 230000000284 resting effect Effects 0.000 claims description 13
- SAZUGELZHZOXHB-UHFFFAOYSA-N acecarbromal Chemical compound CCC(Br)(CC)C(=O)NC(=O)NC(C)=O SAZUGELZHZOXHB-UHFFFAOYSA-N 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 description 17
- 230000008569 process Effects 0.000 description 10
- 238000003066 decision tree Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000000386 athletic effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/0255—Recording instruments specially adapted therefor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Cardiology (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Psychiatry (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Physical Education & Sports Medicine (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The embodiment of the invention provides a training method of a exercise heart rate prediction model based on a power vehicle and a heart rate prediction method, wherein the method comprises the following steps: predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; and obtaining a trained target exercise heart rate prediction model according to the exercise heart rate characteristic information of the user and the gradient lifting tree algorithm. The method provided by the embodiment of the invention realizes accurate prediction of the heart rate of the user.
Description
Technical Field
The invention relates to the technical field of physiological signal monitoring, in particular to a training method of an exercise heart rate prediction model based on a power vehicle and a heart rate prediction method.
Background
The exercise heart rate prediction can effectively improve exercise efficiency, prevent exercise injury and evaluate exercise capacity, and is helpful for more reasonably designing exercise plans. Has more important positive significance for pregnant women, slow patients and the elderly.
In the related art, the heart rate prediction of the user is performed based on experience, resulting in lower accuracy of the heart rate prediction. Thus, how to accurately predict the heart rate of a user is a technical problem that needs to be solved by those skilled in the art.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides a training method of a exercise heart rate prediction model and a heart rate prediction method based on a power vehicle.
Specifically, the embodiment of the invention provides the following technical scheme:
in a first aspect, an embodiment of the present invention provides a exercise heart rate prediction model training method based on a power vehicle, including:
predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
Determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period;
obtaining a trained target exercise heart rate prediction model according to the user exercise heart rate characteristic information and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used for predicting the exercise heart rate of the user.
Further, the exercise heart rate first prediction model is determined based on:
determining a user-stabilized heart rate of the power vehicle at a first power and a user-stabilized heart rate of the power vehicle at a second power;
and linearly estimating the user stable heart rate under the first power and the user stable heart rate under the second power to obtain the first prediction model of the exercise heart rate.
Further, the heart rate response fitting curve is determined based on the following:
determining a user heart rate response interval after the power of the power vehicle in the target time period changes according to the user heart rate data at each moment in the heart rate measurement data set and the power of the power vehicle in the target time period; the heart rate response interval represents a time period interval in which the standard deviation of the heart rate of a user is greater than a threshold value under the condition that the power of the power vehicle is changed;
And fitting the heart rate data of the user in the heart rate measurement data set corresponding to the heart rate response interval of the user to obtain the heart rate response fitting curve.
Further, the heart rate drift fit curve is determined based on the following:
updating the user heart rate data in the first heart rate prediction data set according to the heart rate response fitting curve in the heart rate response interval to obtain a second heart rate prediction data set;
and fitting the second heart rate prediction data set and the heart rate measurement data set to obtain a heart rate drift fitting curve.
Further, the updating the user heart rate in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set includes:
in the heart rate response interval, updating the heart rate data of the user in the first heart rate prediction data set according to a heart rate response fitting curve to obtain a second heart rate prediction data set;
and updating the user heart rate data in the second heart rate prediction data set according to the heart rate drift fitting curve to obtain the target heart rate prediction data set.
Further, the determining user exercise heart rate characteristic information from the target heart rate prediction dataset and the heart rate measurement dataset comprises at least one of:
obtaining a reserve heart rate percentage corresponding to the target heart rate prediction data set according to the target heart rate prediction data set and the resting heart rate;
obtaining a reserve heart rate percentage corresponding to the heart rate measurement data set according to the heart rate measurement data set and the resting heart rate;
determining the quantity of user motion according to the heart rate measurement data set and the motion time corresponding to the target time period;
and taking the reserve heart rate percentage corresponding to the target heart rate prediction data set, the reserve heart rate percentage corresponding to the heart rate measurement data set, the user motion quantity and the heart rate drift slope corresponding to the heart rate drift fitting curve as the user motion heart rate characteristic information.
In a second aspect, an embodiment of the present invention further provides a exercise heart rate prediction method based on a power vehicle, including:
acquiring the power of a power vehicle in a first time period;
predicting the heart rate of the user under each power according to the power of the power vehicle and the first prediction model of the exercise heart rate in the first time period to obtain a heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
And inputting the heart rate prediction data set into a target exercise heart rate prediction model to obtain heart rate information of the user in the first time period, wherein the target exercise heart rate prediction model is trained based on the exercise heart rate prediction model training method based on the power vehicle according to the first aspect.
In a third aspect, an embodiment of the present invention further provides a exercise heart rate prediction model training device based on a power vehicle, including:
the first processing module is used for predicting the heart rate of the user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in the target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
the second processing module is used for updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
The determining module is used for determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period;
the training module is used for obtaining a trained target exercise heart rate prediction model according to the user exercise heart rate characteristic information and the gradient lifting tree algorithm; the target exercise heart rate prediction model is used for predicting the exercise heart rate of the user.
In a fourth aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the power-vehicle-based exercise heart rate prediction model training method according to the first aspect or the power-vehicle-based exercise heart rate prediction method according to the second aspect when the program is executed.
In a fifth aspect, embodiments of the present invention also provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the power car based exercise heart rate prediction model training method as described in the first aspect or the power car based exercise heart rate prediction method as described in the second aspect.
In a sixth aspect, embodiments of the present invention further provide a computer program product comprising a computer program which, when executed by a processor, implements the power cart based exercise heart rate prediction model training method as described in the first aspect or the power cart based exercise heart rate prediction method as described in the second aspect.
According to the exercise heart rate prediction model training method and the heart rate prediction method based on the power vehicle, the first heart rate prediction data set, namely the linear relation between the power of the power vehicle and the exercise heart rate of the user, is obtained according to the power of the power vehicle and the exercise heart rate first prediction model in the target time period, so that the exercise heart rate of the user can be predicted simply and rapidly; updating the heart rate of the user in the first heart rate prediction data set through the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set, and overcoming heart rate response errors and heart rate drift errors, so that the predicted target heart rate prediction data set is more accurate; finally, establishing an association relation between the user heart rate and the user exercise heart rate characteristic information, and characterizing the exercise heart rate characteristic information of the user through the user heart rate, so that more and richer user exercise characteristic information can be obtained conveniently, and further, the gradient lifting tree algorithm is trained based on more and richer user exercise characteristic information, so that the target exercise heart rate prediction model obtained after training can be used for predicting heart rate more accurately, and more accurate heart rate prediction results can be obtained.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a training method of a power vehicle-based exercise heart rate prediction model according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a exercise heart rate prediction model training device based on a power vehicle according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The method provided by the embodiment of the invention can be applied to a physiological signal monitoring scene, and accurate prediction of the heart rate of the user is realized.
In the related art, the heart rate prediction of the user is performed based on experience, resulting in lower accuracy of the heart rate prediction. Thus, how to accurately predict the heart rate of a user is a technical problem that needs to be solved by those skilled in the art.
According to the exercise heart rate prediction model training method based on the power vehicle, a first heart rate prediction data set, namely, the linear relation between the power of the power vehicle and the exercise heart rate of a user is obtained according to the power of the power vehicle and the exercise heart rate first prediction model in a target time period, so that the exercise heart rate of the user can be predicted simply and quickly; updating the heart rate of the user in the first heart rate prediction data set through the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set, and overcoming heart rate response errors and heart rate drift errors, so that the predicted target heart rate prediction data set is more accurate; finally, establishing an association relation between the user heart rate and the user exercise heart rate characteristic information, and characterizing the exercise heart rate characteristic information of the user through the user heart rate, so that more and richer user exercise characteristic information can be obtained conveniently, and further, the gradient lifting tree algorithm is trained based on more and richer user exercise characteristic information, so that the target exercise heart rate prediction model obtained after training can be used for predicting heart rate more accurately, and more accurate heart rate prediction results can be obtained.
The following describes the technical scheme of the present invention in detail with reference to fig. 1 to 3. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.
Fig. 1 is a flowchart of an embodiment of a training method for a power vehicle-based exercise heart rate prediction model according to an embodiment of the present invention. As shown in fig. 1, the method provided in this embodiment includes:
specifically, in the prior art, the heart rate prediction of the user is performed based on experience, resulting in lower accuracy of the heart rate prediction.
In order to solve the above problems, in the embodiment of the present invention, a first heart rate prediction data set is obtained by predicting a user heart rate at each power according to a first prediction model of power and exercise heart rate of a power vehicle in a target time period; wherein the exercise heart rate first prediction model is used for predicting the heart rate of the user based on the power of the power car. Optionally, the exercise heart rate first prediction model is determined based on linear estimation of the heart rate of the user when the power of the power car is w2 and the heart rate of the user when the power of the power car is w 1:
HR target ＝(HR w2 -HR w1 )×(W target -W 1 )÷(W 2 -W 1 )+HR w1 ；
Wherein HR is w2 Representing the steady heart rate, HR, of the user at a power w2 of the powered vehicle w1 Representing the steady heart rate of the user at a power of W1 for a powered vehicle, optionally at W 1 After the heart rate is stable, using a heart rate meter to record the average heart rate of 30 seconds, namely, the stable heart rate HR of the user when the power of the power vehicle is w1 w1 ；W target Representing any power of the power vehicle within the target time period, HR target The power of a user corresponding to the power vehicle under any power in a target time period is represented; optionally, under the condition that the power of the power vehicle changes for multiple times in the target time period, for example, the power of the power vehicle in the first sub-time period in the target time period is power 1, the stable heart rate of the user when the power of the power vehicle is power 1 can be obtained according to the first prediction model, and the power of the power vehicle in the second sub-time period in the target time period is power 2, the stable heart rate of the user when the power of the power vehicle is power 2 can be obtained according to the first prediction model; optionally, since the exercise heart rate first prediction model is determined based on linear estimation of the heart rate of the user when the power of the power vehicle is w2 and the heart rate of the user when the power of the power vehicle is w1, the heart rates of the users under the same power determined based on the exercise heart rate first prediction model are the same, and since the power of the power vehicle is the same in the same sub-time period, the heart rates of the users in the same sub-time period are the same; in this way, a stable heart rate of the user in each sub-period of the target period of time can be obtained, and a user heart rate-time data set, i.e. a first heart rate prediction data set HR, can be established target {t i ,HRT i }。
For example, in the target time period, the first sub-time period is 0-5 minutes, the corresponding power is 50W, and the heart of the user under the power is obtained according to the first prediction model of the exercise heart rateThe rate is X1; the second sub-time period is 5-10 minutes, the corresponding power is 100W, the heart rate of the user under the power is X2 according to the first prediction model of the exercise heart rate, and if the heart rate time data set HR of the user is target If the statistical granularity of the statistics is 1 second, a first heart rate prediction data set HR corresponding to the target time period targrt {t i ,HRT i }＝{1,X1；2,X1；………；300,X1；301,X2；301,X2；………；600,X2}。
102, updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
specifically, after predicting the heart rate of the user under each power according to the power of the power vehicle and the first prediction model of the exercise heart rate in the target time period to obtain a first heart rate prediction data set, the heart rate of the user is determined by linear estimation based on the heart rate of the user when the power of the power vehicle is w2 and the heart rate of the user when the power of the power vehicle is w1, so that the heart rates of the users under the same power determined based on the first prediction model of the exercise heart rate are the same; however, in the actual exercise process of the user, heart rate response exists in heart rate change of the user relative to power change of the power vehicle, namely after the power of the power vehicle is changed, the heart rate of the user is not directly hopped from a first stable heart rate before the power change to a second stable heart rate after the power change, but after the power of the power vehicle is changed, heart rate response exists, and the heart rate of the user is gradually adjusted from the first stable heart rate before the power change to the second stable heart rate after the power change after a heart rate response time period; the heart rate response time period is a time period corresponding to the time period when the heart rate of the user is gradually adjusted from the first stable heart rate before the power change to the second stable heart rate after the power change. In the process of carrying out heart rate prediction based on the exercise heart rate first prediction model, the influence of heart rate response on the accuracy of heart rate prediction of a user is not considered, so that the first heart rate prediction data set predicted based on the exercise heart rate first prediction model has errors in the heart rate response time period compared with the actual heart rate of the user.
For example, the power of the power car is adjusted from 80W to 100W, the stable heart rate of the user at the power car 80W is M1, and the stable heart rate of the user at the power car 100W is M2, but after the power of the power car is adjusted from 80W to 100W, the heart rate is not adjusted from M1 to M2 immediately along with the change of the power, but the heart rate is stabilized to M2 after the heart rate response time period gradually passes. I.e. the power of the power cart may be a jump procedure, while the heart rate of the user is a gradual procedure.
On the other hand, the heart rate drift phenomenon exists in the exercise process of the user, namely, when the user runs for a long time in the aerobic endurance zone, the same matching speed is maintained, but the heart rate slowly rises, namely, the heart rate drift is generated. In the process of carrying out heart rate prediction based on the first prediction model of the exercise heart rate, the influence of heart rate drift on the accuracy of heart rate prediction of the user is not considered, so that the first heart rate prediction data set predicted based on the first prediction model of the exercise heart rate has errors of heart rate drift compared with the actual heart rate of the user.
In order to overcome heart rate response errors and heart rate drift errors in a first heart rate prediction data set which is predicted based on a motion heart rate first prediction model, so that predicted heart rates of users are more accurate, in the embodiment of the invention, the heart rates of the users in the first heart rate prediction data set are updated according to a heart rate response fitting curve and a heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set, so that after the first heart rate prediction data set predicted by the exercise heart rate first prediction model is obtained, the target heart rate data set can be obtained based on the first heart rate prediction data set and the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set. The heart rate of the user in the first heart rate prediction data set is updated according to the heart rate response fitting curve and the heart rate drift fitting curve, so that the obtained target heart rate prediction data set overcomes heart rate response errors and heart rate drift errors, and the predicted target heart rate prediction data set is more accurate.
specifically, after updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set, determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set in the embodiment of the invention; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period through a heart rate meter; the exercise heart rate characteristic information of the user is determined through a target heart rate data set which has overcome heart rate response errors and heart rate drift errors and a heart rate measurement data set which reflects the actual exercise heart rate of the user and is measured by using a heart rate meter; optionally, the user athletic information may include a user's quantity of motion, a user's reserve heart rate percentage; the method comprises the steps of establishing an association relation between exercise heart rate characteristic information of a user and the heart rate of the user, characterizing the exercise heart rate characteristic information of the user through the heart rate of the user, isolating other factors affecting the quantity of motion of the user and the reserve heart rate percentage of the user, and accordingly, based on heart rate data of the user, simply and quickly characterizing and obtaining the exercise characteristic information of the user, and further, obtaining more and richer exercise characteristic information of the user based on the heart rate information of the user. Alternatively, the exercise heart rate characteristic information corresponding to the measurement data set may be obtained based on the heart rate measurement data set, or the exercise heart rate characteristic information corresponding to the target heart rate data set may be obtained based on the target heart rate data set.
104, obtaining a trained target exercise heart rate prediction model according to the exercise heart rate characteristic information of the user and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used to predict an exercise heart rate of the user.
Specifically, after the exercise heart rate characteristic information of the user is determined according to the target heart rate prediction data set and the heart rate measurement data set, namely the exercise heart rate characteristic information of the user is characterized by the heart rate of the user, a trained target exercise heart rate prediction model can be obtained according to the exercise heart rate characteristic information of the user and a gradient lifting tree algorithm; optionally, the gradient lifting tree algorithm in the embodiment of the invention may be a gradient lifting tree algorithm xgboost, which is a method for approaching discrete function values, and a decision tree model is constructed according to a given training data set, so that the decision tree model can correctly classify and predict an instance; decision tree learning typically includes 3 steps: feature selection, decision tree generation and decision tree pruning.
Optionally, after determining the exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set, that is, after selecting the characteristic value, the embodiment of the invention can obtain the trained target exercise heart rate prediction model based on the exercise heart rate characteristic information of the user and the gradient lifting tree algorithm; optionally, the method for generating the decision tree based on the selected feature information is a conventional method in the art, and in the embodiments of the present application, a key process is described, and specific details are not repeated. The exercise heart rate characteristic information of the user is input into a gradient lifting tree algorithm, and a trained target exercise heart rate prediction model can be obtained after training through a target loss function; the target heart rate data set overcoming the heart rate response error and the heart rate drift error is rectified and trained based on the target training function, so that a more accurate heart rate prediction result can be obtained based on the trained target exercise heart rate prediction model, and the heart rate prediction can be more accurately performed based on the target exercise heart rate prediction model.
According to the method, the first heart rate prediction data set is obtained according to the first prediction model of the power and the exercise heart rate of the power vehicle in the target time period, namely, the exercise heart rate of the user can be predicted simply and rapidly based on the linear relation between the power of the power vehicle and the exercise heart rate of the user; updating the heart rate of the user in the first heart rate prediction data set through the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set, and overcoming heart rate response errors and heart rate drift errors, so that the predicted target heart rate prediction data set is more accurate; finally, establishing an association relation between the user heart rate and the user exercise heart rate characteristic information, and characterizing the exercise heart rate characteristic information of the user through the user heart rate, so that more and richer user exercise characteristic information can be obtained conveniently, and further, the gradient lifting tree algorithm is trained based on more and richer user exercise characteristic information, so that the target exercise heart rate prediction model obtained after training can be used for predicting heart rate more accurately, and more accurate heart rate prediction results can be obtained.
In one embodiment, the heart rate response fitting curve is determined based on the following:
determining a user heart rate response interval after the power of the power vehicle changes in the target time period according to the user heart rate data at each moment in the heart rate measurement data set and the power of the power vehicle in the target time period; the heart rate response interval represents a time period interval in which the heart rate standard deviation of the user is greater than a threshold value under the condition that the power of the power vehicle is changed;
and fitting the heart rate data of the user in the heart rate measurement data set corresponding to the heart rate response interval of the user to obtain a heart rate response fitting curve.
Specifically, in the embodiment of the invention, the first heart rate prediction data set is obtained according to the first prediction model of the power and the exercise heart rate of the power vehicle in the target time period, namely, the exercise heart rate of the user can be simply and quickly predicted based on the linear relation between the power of the power vehicle and the exercise heart rate of the user; however, in the process of predicting the heart rate based on the exercise heart rate first prediction model, the influence of the heart rate response on the accuracy of heart rate prediction of the user is not considered, so that the first heart rate prediction data set predicted based on the exercise heart rate first prediction model has errors in the heart rate response time period compared with the actual heart rate of the user.
In order to overcome the heart rate response errors in the first heart rate prediction data set predicted based on the exercise heart rate first prediction model, the predicted heart rate of the user is more accurate. Optionally, the heart rate response fitting curve is determined based on: determining a user heart rate response interval after the power of the power vehicle changes in the target time period according to the user heart rate data at each moment in the heart rate measurement data set and the power of the power vehicle in the target time period; the heart rate response interval represents a time period interval in which the heart rate standard deviation of the user is greater than a threshold value under the condition that the power of the power vehicle is changed; fitting the heart rate data of the user in the heart rate measurement data set corresponding to the heart rate response interval of the user, so that a heart rate response fitting curve can be obtained; that is, determining whether the heart rate response interval of the user belongs to by determining the standard deviation of the heart rate change of the user at each moment compared with the heart rate change of the user at the previous moment; optionally, if the standard deviation of the change of the heart rate of the user at the moment is smaller than the first threshold value compared with the previous moment, the heart rate of the user is in a stable state, and the moment does not belong to the heart rate response interval; if the standard deviation of the change of the heart rate of the user at the moment is larger than or equal to the first threshold value compared with the change of the heart rate of the user at the previous moment, the heart rate of the user is not in a stable state, and the moment belongs to the heart rate response interval. Further, the measured heart rate of the user in the heart rate response interval is obtained, and the measured heart rate of the user in the heart rate response interval is fitted, so that a heart rate response fitting curve can be accurately obtained; optionally, fitting the measured heart rate of the user in the heart rate response interval by a least square method to obtain a heart rate response fitting curve; and then the heart rate of the user in the first heart rate prediction data set is updated through the heart rate response fitting curve, so that a heart rate prediction result overcoming the heart rate response error can be obtained, and the predicted heart rate of the user is more accurate.
According to the method, the user heart rate response interval after the power of the power vehicle is changed in the target time period is determined according to the user heart rate data at each moment in the heart rate measurement data set and the power of the power vehicle in the target time period, the user heart rate data in the heart rate measurement data set corresponding to the user heart rate response interval is fitted to obtain a heart rate response fitting curve, and further the user heart rate in the first heart rate prediction data set is updated through the heart rate response fitting curve, so that a heart rate prediction result overcoming heart rate response errors can be obtained, and the predicted user heart rate is more accurate.
In one embodiment, the heart rate drift fit curve is determined based on the following:
in the heart rate response interval, updating the heart rate data of the user in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set;
and fitting the second heart rate prediction data set and the heart rate measurement data set to obtain a heart rate drift fitting curve.
Specifically, in the embodiment of the invention, the first heart rate prediction data set is obtained according to the first prediction model of the power and the exercise heart rate of the power vehicle in the target time period, namely, the exercise heart rate of the user can be simply and quickly predicted based on the linear relation between the power of the power vehicle and the exercise heart rate of the user; however, in the process of predicting the heart rate based on the exercise heart rate first prediction model, the influence of the heart rate drift on the accuracy of the heart rate prediction of the user is not considered, so that the first heart rate prediction data set predicted based on the exercise heart rate first prediction model has an error of the heart rate drift compared with the real heart rate of the user.
In order to overcome the heart rate drift error in the first heart rate prediction data set predicted based on the motion heart rate first prediction model, the predicted heart rate of the user is more accurate. Optionally, the heart rate drift fit curve is determined based on: in the heart rate response interval, updating the heart rate data of the user in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set; fitting the second heart rate prediction data set and the heart rate measurement data set to obtain a heart rate drift fitting curve; namely, after updating the heart rate data of the user in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set overcoming the heart rate response error, fitting the second heart rate prediction data set overcoming the heart rate response error with a heart rate measurement data set obtained by measuring a heart rate meter to obtain a heart rate drift fitting curve; optionally, the second heart rate prediction data set and the heart rate measurement data set can be fitted through a least square method to obtain a heart rate drift fitting curve; and the predicted heart rate of the user is updated through the heart rate drift fitting curve, so that a heart rate prediction result which overcomes the heart rate drift error can be obtained, and the predicted heart rate of the user is more accurate.
The method of the above embodiment updates the user heart rate data in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set; fitting the second heart rate prediction data set and the heart rate measurement data set to obtain a heart rate drift fitting curve, and updating the predicted heart rate of the user through the heart rate drift fitting curve, so that a heart rate prediction result overcoming a heart rate drift error can be obtained, and the predicted heart rate of the user is more accurate.
In an embodiment, updating the heart rate of the user in the first heart rate prediction dataset according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain the target heart rate prediction dataset comprises:
in the heart rate response interval, updating the heart rate data of the user in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set;
and updating the user heart rate data in the second heart rate prediction data set according to the heart rate drift fitting curve to obtain a target heart rate prediction data set.
Specifically, in the embodiment of the invention, the first heart rate prediction data set is obtained according to the first prediction model of the power and the exercise heart rate of the power vehicle in the target time period, namely, the exercise heart rate of the user can be simply and quickly predicted based on the linear relation between the power of the power vehicle and the exercise heart rate of the user; however, in the process of carrying out heart rate prediction based on the exercise heart rate first prediction model, the influence of heart rate response and heart rate drift on the accuracy of heart rate prediction of the user is not considered, so that the first heart rate prediction data set predicted based on the exercise heart rate first prediction model has errors of heart rate response and heart rate drift compared with the actual heart rate of the user.
In order to overcome the heart rate response errors and heart rate drift errors in the first heart rate prediction data set predicted based on the motion heart rate first prediction model, the predicted heart rate of the user is more accurate. Optionally, updating the heart rate data of the user in the first heart rate prediction data set according to the heart rate response fitting curve to obtain a second heart rate prediction data set, so that the obtained second heart rate prediction data set overcomes heart rate response errors; and the heart rate data of the user in the second heart rate prediction data set can be updated through the heart rate drift fitting curve to obtain a target heart rate prediction data set, so that the obtained target heart rate prediction data set further overcomes the heart rate drift error on the basis of overcoming the heart rate response error, and the predicted target heart rate prediction data set is more accurate.
According to the method, the heart rate data of the user in the first heart rate prediction data set is updated according to the heart rate response fitting curve to obtain a second heart rate prediction data set, so that the obtained second heart rate prediction data set overcomes heart rate response errors; and updating the user heart rate data in the second heart rate prediction data set through a heart rate drift fitting curve to obtain a target heart rate prediction data set, so that the obtained target heart rate prediction data set further overcomes the heart rate drift error on the basis of overcoming the heart rate response error, and the predicted target heart rate prediction data set is more accurate.
In an embodiment, the user exercise heart rate characteristic information is determined from the target heart rate prediction dataset and the heart rate measurement dataset, comprising at least one of:
obtaining a reserve heart rate percentage corresponding to the target heart rate prediction data set according to the target heart rate prediction data set and the resting heart rate;
obtaining a reserve heart rate percentage corresponding to the heart rate measurement data set according to the heart rate measurement data set and the resting heart rate;
determining the quantity of user motion according to the heart rate measurement data set and the motion time corresponding to the target time period;
And taking the heart rate drift slope corresponding to the reserve heart rate percentage corresponding to the target heart rate prediction data set, the reserve heart rate percentage corresponding to the heart rate measurement data set, the user motion quantity and the heart rate drift fitting curve as the user motion heart rate characteristic information.
Specifically, in the embodiment of the invention, a trained target exercise heart rate prediction model is obtained according to user exercise heart rate characteristic information and a gradient lifting tree algorithm; optionally, inputting the exercise heart rate characteristic information of the user into a gradient lifting tree algorithm, and training through a target loss function to obtain a trained target exercise heart rate prediction model; optionally, exercise heart rate characteristic information corresponding to the measurement data set can be obtained based on the heart rate measurement data set, exercise heart rate characteristic information corresponding to the target heart rate data set can be obtained based on the target heart rate data set, and training is performed based on the exercise heart rate characteristic information obtained by the heart rate measurement data set, the exercise heart rate characteristic information corresponding to the target heart rate data set and the target loss function, so that a trained target exercise heart rate prediction model can be obtained. Optionally, the trained target exercise heart rate prediction model obtains the user exercise heart rate characteristic information through the heart rate measurement data set, and further rectifies and trains the target heart rate data set which overcomes the heart rate response error and the heart rate drift error based on the dimension of the user exercise heart rate characteristic information, so that the trained target exercise heart rate prediction model can obtain a more accurate heart rate prediction result, and the heart rate prediction can be performed more accurately based on the target exercise heart rate prediction model.
Optionally, determining the user exercise heart rate characteristic information from the target heart rate prediction dataset and the heart rate measurement dataset comprises:
obtaining a reserve heart rate percentage corresponding to the target heart rate prediction data set according to the target heart rate prediction data set and the resting heart rate; wherein, the reserve heart rate percentage refers to the ratio between the heart rate reached at the current exercise and the reserve heart rate; reserve heart rate is the difference between maximum heart rate and resting heart rate; alternatively, the reserve heart rate percentage corresponding to the target heart rate prediction dataset may be determined based on: % HRR target ＝(HR target -HR rest )÷(HR max -HR rest ) The method comprises the steps of carrying out a first treatment on the surface of the wherein%HRR target Representing a reserve heart rate percentage corresponding to a predicted user heart rate included in the target heart rate prediction dataset; HR (HR) target Representing a predicted user heart rate included in the target heart rate prediction dataset; HR (HR) rest Representing a resting heart rate of the user; HR (HR) max Representing the maximum heart rate of the user. Alternatively,% HRR may be used target Conversion to
Alternatively, based on the method, the reserve heart rate percentage corresponding to the heart rate measurement data set can be obtained according to the heart rate measurement data set and the resting heart rate; and the user exercise heart rate characteristic information obtained based on the heart rate measurement data set and the user exercise heart rate characteristic information obtained based on the target heart rate prediction data set are input into a gradient lifting tree algorithm for training, so that the user exercise heart rate characteristic information obtained based on the target heart rate prediction data set can be further corrected and trained based on the dimension of the characteristic information, and the trained target exercise heart rate prediction model can obtain a more accurate heart rate prediction result, so that the heart rate prediction can be performed more accurately based on the target exercise heart rate prediction model.
Optionally, determining a user motion amount according to the heart rate measurement data set and the motion time corresponding to the target time period; alternatively, the user's quantity of motion may be determined by the following formula:
wherein HR is j Representing the user measured heart rate at the j-th moment comprised by the user heart rate measurement dataset.
Optionally, in the embodiment of the present invention, the heart rate drift slope corresponding to the fitted drift curve obtained after fitting is also used as the characteristic information in the input gradient lifting tree algorithm; alternatively, the coefficient value in the fitted drift curve may be taken as the fitted drift curve slope; the motion quantity and the heart rate drift slope of the user are characterized through the heart rate of the user, so that more and richer user motion characteristic information can be obtained conveniently, the gradient lifting tree algorithm is trained based on more and richer user motion characteristic information, the target motion heart rate prediction model obtained after training can be used for predicting the heart rate more accurately, and more accurate heart rate prediction results can be obtained.
According to the method, the reserve heart rate percentage corresponding to the target heart rate prediction data set is obtained according to the target heart rate prediction data set and the resting heart rate; obtaining a reserve heart rate percentage corresponding to the heart rate measurement data set according to the heart rate measurement data set and the resting heart rate; determining the quantity of user motion according to the heart rate measurement data set and the motion time corresponding to the target time period; taking the reserve heart rate percentage corresponding to the target heart rate prediction data set, the reserve heart rate percentage corresponding to the heart rate measurement data set, the user motion quantity and the heart rate drift slope corresponding to the heart rate drift fitting curve as the user motion heart rate characteristic information; the user exercise heart rate characteristic information is input into a gradient lifting tree algorithm for training, so that correction and training are performed on the user exercise heart rate characteristic information obtained based on the target heart rate prediction data set based on the dimension of the characteristic information and the user exercise heart rate characteristic information obtained based on the heart rate measurement data set, and a more accurate heart rate prediction result can be obtained by the trained target exercise heart rate prediction model, and heart rate prediction can be performed more accurately based on the target exercise heart rate prediction model; further, through establishing an association relation between the user motion quantity characteristic information, the user heart rate drift slope characteristic information and the user heart rate, the motion quantity and the heart rate drift slope of the user are characterized through the user heart rate, so that more and richer user motion characteristic information can be obtained conveniently, and further, a gradient lifting tree algorithm is trained based on more and richer user motion characteristic information, so that a target motion heart rate prediction model obtained after training can be used for predicting the heart rate more accurately, and a more accurate heart rate prediction result can be obtained.
In an embodiment, the embodiment of the invention also discloses a exercise heart rate prediction method based on a power vehicle, which comprises the following steps:
acquiring the power of a power vehicle in a first time period;
predicting the heart rate of the user under each power according to a first power and exercise heart rate prediction model of the power vehicle in a first time period to obtain a heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
and inputting the heart rate prediction data set into a target exercise heart rate prediction model to obtain heart rate information of the user in a first time period, wherein the target exercise heart rate prediction model is trained based on the exercise heart rate prediction model training method based on the power vehicle.
Specifically, after a trained target exercise heart rate prediction model is obtained according to user exercise heart rate characteristic information and a gradient lifting tree algorithm, the power of a power vehicle in a first time period can be obtained; optionally, the power of the power car in the first time period may be the power of the power car in any time period, and further, the power of the power car in the first time period is input to the first prediction model of exercise heart rate, so that a predicted heart rate prediction dataset of the user can be obtained based on a linear relationship between the power of the power car and the heart rate of the user, and further, the heart rate response error and the heart rate offset error of the user heart rate in the heart rate prediction dataset are corrected according to the determined heart rate response fitting curve and the heart rate drift curve of the user, and then are input to the target exercise heart rate prediction model, and the target heart rate prediction model can accurately predict the heart rate of the user according to the characteristic information corresponding to the corrected heart rate prediction dataset, so that accurate and efficient prediction of the heart rate of the user is realized.
The exercise heart rate prediction model training method based on the power vehicle comprises the following specific steps of:
(1) The exercise carrier is used as the power exercise bicycle, the power exercise bicycle adopts a constant power mode, so that stable power output in the riding process of a user is ensured, and average power data with the width of 5 seconds is synchronously output; the user wears a heart rate meter, and the heart rate meter provides real-time heart rate data;
(2) Before exercise, collecting the age of the user, and calculating the maximum heart rate HR of the user max ＝207-0.7×age；
(3) The user is guided to rest for 5 minutes to collect the resting heart rate HR before the exercise begins rest ；
(4) During exercise, the first stage is to warm up at 25W for 5 min, and after heart rate is stabilized, 30 seconds of average heart rate is recorded as HR warmup In the second stage with W 1 Is moved for 5 minutes at the power of (2) and an average heart rate HR of 30 seconds is recorded after the heart rate has stabilized w1 In the third stage with W 2 Is moved for 5 minutes at the power of (2) and an average heart rate HR of 30 seconds is recorded after the heart rate has stabilized w2 In the fourth stage W 3 Is moved for 15 minutes; the course of motion records the heart rate-time dataset HR { t } i ,HR i }；
(5) According to step 4, wherein W 1 ＝25W×(120bmp-HR rest )÷(HR warmup -HR rest )，W 2 ＝25W×(135bmp-HR rest )÷(HR warmup -HR rest )，W 3 ＝(W 2 -W 1 )×(HR max -HR rest )÷(HR w2 -HR w1 )×55％+W 1 ；
(6) According to step 4, the method for calculating the target heart rate according to the exercise power of the user is that,
HR target ＝(HR w2 -HR w1 )×(W target -W 1 )÷(W 2 -W 1 )+HR w1 the method comprises the steps of carrying out a first treatment on the surface of the According to the preset power and time of each stage described in 4, calculating target heart rate-time data set HR according to the established power-time data set target {t i ,HRT i }；
According to the description 4, the heart rate standard deviation maximum value in the heart rate stabilizing section is used as a boundary threshold value of heart rate following change after exercise power change, the heart rate standard deviation is slid, and a heart rate response section after power change is judged;
(7) According to steps 4 and 5, for the heart rate-time dataset HR { t } i ,HR i Heart rate response interval the heart rate data of the heart rate response interval is fitted to the heart rate response interval by means of a first order least squares method, the target heart rate-time data set HR target {t i ,HRT i Replacing the heart rate response interval part with linear interpolation according to fitting data;
(8) According to steps 4 and 7, using HR { t } i ,HR i Sum HR of target {t i ,HRT i Conversion of data sets into
(9) According to 8, pair
First order least squares fitting z=min Σ (y) is performed i -ax i -b) 2 Obtaining an extremum to obtain a slope a value;
(10) According to steps 4 and 5, the HRR is calculated according to the reserve heart rate percentage formula% target ＝(HR target -HR rest )÷(HR max -HR rest ) For HR target {t i ,HRT i Conversion to percent reserve heart rate
(11) Data set
According to->
Performing linear interpolation;
(12) Selected characteristic rate drift slope a value and reserve heart rate percentage%hrr i Exercise amount
As characteristic information, further inputting characteristic values corresponding to the characteristic information into a gradient lifting tree algorithm, a trained target exercise heart rate prediction model can be obtained, and further prediction of the heart rate of the user can be accurately performed based on the trained target exercise heart rate prediction model.
(13) After the exercise heart rate characteristic information of the user is determined, namely after the characteristic value is selected, the exercise heart rate characteristic information is put into a CART regression tree model, and a regression tree is generated according to a least square regression tree generation algorithm; sequentially calculating standard deviation of the features, selecting an optimal segmentation variable j and segmentation points s, traversing j, scanning the segmentation points s for a fixed segmentation variable j, and selecting
Minimum (j, s); dividing the region by the selected pair (j, s) and determining the output value:
R 1 (j,s)＝{x|x (1) ≤s}，R 2 (j,s)＝{x|x (j) >s}
calling the sub-regions in sequence, and dividing the input space into M regions R after completion 1 、R 2 、R 3 、......、R M Generating a decision tree;
(14) The objective loss function is as follows:
wherein G is j For leaf node j comprising the sum of the first partial derivatives of the samples, H j For leaf node j, T is the number of leaf nodes, w j And the node weight values are lambda and gamma, and the regulating parameters are lambda and gamma.
The objective function of leaf node j is:
then each leaf node weight w j And the final Obj target value is reached at this time:
the trained target exercise heart rate prediction model can be obtained by inputting the characteristic value corresponding to the characteristic information into the gradient lifting tree algorithm, and further, the prediction of the heart rate of the user can be accurately performed based on the trained target exercise heart rate prediction model.
The motion heart rate prediction device based on the power vehicle provided by the invention is described below, and the motion heart rate prediction device based on the power vehicle described below and the motion heart rate prediction method based on the power vehicle described above can be correspondingly referred to each other.
Fig. 2 is a schematic structural diagram of the exercise heart rate prediction device based on the power vehicle. The exercise heart rate prediction device based on power car that this embodiment provided includes:
the first processing module 710 is configured to predict a heart rate of a user at each power according to a first prediction model of power and exercise heart rate of the power vehicle in a target time period, so as to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
the second processing module 720 is configured to update the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve, so as to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
a determining module 730 for determining exercise heart rate characteristic information of the user based on the target heart rate prediction dataset and the heart rate measurement dataset; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period;
The training module 740 is configured to obtain a trained target exercise heart rate prediction model according to the exercise heart rate feature information of the user and the gradient lifting tree algorithm; the target exercise heart rate prediction model is used to predict an exercise heart rate of the user.
The device of the embodiment of the present invention is configured to perform the method of any of the foregoing method embodiments, and its implementation principle and technical effects are similar, and are not described in detail herein.
Fig. 3 illustrates a physical schematic diagram of an electronic device, which may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform a power cart based exercise heart rate prediction model training method comprising: predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle; updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set; determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period; obtaining a trained target exercise heart rate prediction model according to the exercise heart rate characteristic information of the user and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used to predict an exercise heart rate of the user.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of power cart-based exercise heart rate prediction model training provided by the methods described above, the method comprising: predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle; updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set; determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period; obtaining a trained target exercise heart rate prediction model according to the exercise heart rate characteristic information of the user and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used to predict an exercise heart rate of the user.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the above provided exercise heart rate prediction model training method based on a power cart, the method comprising: predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle; updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set; determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period; obtaining a trained target exercise heart rate prediction model according to the exercise heart rate characteristic information of the user and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used to predict an exercise heart rate of the user.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The exercise heart rate prediction model training method based on the power vehicle is characterized by comprising the following steps of:
predicting the heart rate of a user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in a target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
Determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period;
obtaining a trained target exercise heart rate prediction model according to the user exercise heart rate characteristic information and a gradient lifting tree algorithm; the target exercise heart rate prediction model is used for predicting the exercise heart rate of the user.
2. The power car-based exercise heart rate prediction model training method of claim 1, wherein the exercise heart rate first prediction model is determined based on:
determining a user-stabilized heart rate of the power vehicle at a first power and a user-stabilized heart rate of the power vehicle at a second power;
and linearly estimating the user stable heart rate under the first power and the user stable heart rate under the second power to obtain the first prediction model of the exercise heart rate.
3. The power car-based exercise heart rate prediction model training method of claim 2, wherein the heart rate response fitting curve is determined based on:
determining a user heart rate response interval after the power of the power vehicle in the target time period changes according to the user heart rate data at each moment in the heart rate measurement data set and the power of the power vehicle in the target time period; the heart rate response interval represents a time period interval in which the standard deviation of the heart rate of a user is greater than a threshold value under the condition that the power of the power vehicle is changed;
And fitting the heart rate data of the user in the heart rate measurement data set corresponding to the heart rate response interval of the user to obtain the heart rate response fitting curve.
4. A power car based exercise heart rate prediction model training method as claimed in claim 3, wherein the heart rate drift fitting curve is determined based on:
updating the user heart rate data in the first heart rate prediction data set according to the heart rate response fitting curve in the heart rate response interval to obtain a second heart rate prediction data set;
and fitting the second heart rate prediction data set and the heart rate measurement data set to obtain a heart rate drift fitting curve.
5. The exercise heart rate prediction model training method based on the power cart of claim 4, wherein updating the user heart rate in the first heart rate prediction dataset according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction dataset comprises:
in the heart rate response interval, updating the heart rate data of the user in the first heart rate prediction data set according to a heart rate response fitting curve to obtain a second heart rate prediction data set;
And updating the user heart rate data in the second heart rate prediction data set according to the heart rate drift fitting curve to obtain the target heart rate prediction data set.
6. The power car-based exercise heart rate prediction model training method of claim 5, wherein the determining user exercise heart rate characteristic information from the target heart rate prediction dataset and heart rate measurement dataset comprises at least one of:
obtaining a reserve heart rate percentage corresponding to the target heart rate prediction data set according to the target heart rate prediction data set and the resting heart rate;
obtaining a reserve heart rate percentage corresponding to the heart rate measurement data set according to the heart rate measurement data set and the resting heart rate;
determining the quantity of user motion according to the heart rate measurement data set and the motion time corresponding to the target time period;
and taking the reserve heart rate percentage corresponding to the target heart rate prediction data set, the reserve heart rate percentage corresponding to the heart rate measurement data set, the user motion quantity and the heart rate drift slope corresponding to the heart rate drift fitting curve as the user motion heart rate characteristic information.
7. A power car-based exercise heart rate prediction method, comprising:
Acquiring the power of a power vehicle in a first time period;
predicting the heart rate of the user under each power according to the power of the power vehicle and the first prediction model of the exercise heart rate in the first time period to obtain a heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
inputting the heart rate prediction data set into a target exercise heart rate prediction model to obtain heart rate information of the user in the first time period, wherein the target exercise heart rate prediction model is trained based on the exercise heart rate prediction model training method based on the power vehicle according to any one of claims 1-6.
8. Exercise heart rate prediction model trainer based on power car, characterized by comprising:
the first processing module is used for predicting the heart rate of the user under each power according to a first prediction model of the power and the exercise heart rate of the power vehicle in the target time period to obtain a first heart rate prediction data set; the exercise heart rate first prediction model is used for predicting the heart rate of a user based on the power of the power vehicle;
the second processing module is used for updating the heart rate of the user in the first heart rate prediction data set according to the heart rate response fitting curve and the heart rate drift fitting curve to obtain a target heart rate prediction data set; the heart rate response fitting curve and the heart rate drift fitting curve are used for representing the mapping relation between the first heart rate prediction data set and the target heart rate prediction data set;
The determining module is used for determining exercise heart rate characteristic information of the user according to the target heart rate prediction data set and the heart rate measurement data set; the heart rate measurement data set is obtained by measuring the heart rate of the user at each moment in the target period;
the training module is used for obtaining a trained target exercise heart rate prediction model according to the user exercise heart rate characteristic information and the gradient lifting tree algorithm; the target exercise heart rate prediction model is used for predicting the exercise heart rate of the user.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the power car based exercise heart rate prediction model training method of any one of claims 1 to 6 or the power car based exercise heart rate prediction method of claim 7 when the program is executed.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the power car based exercise heart rate prediction model training method of any one of claims 1 to 6 or the power car based exercise heart rate prediction method of claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187037.9A CN116407088B (en) | 2023-03-01 | 2023-03-01 | Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187037.9A CN116407088B (en) | 2023-03-01 | 2023-03-01 | Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116407088A true CN116407088A (en) | 2023-07-11 |
| CN116407088B CN116407088B (en) | 2024-01-09 |
Family
ID=87057347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310187037.9A Active CN116407088B (en) | 2023-03-01 | 2023-03-01 | Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116407088B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030149453A1 (en) * | 2002-02-07 | 2003-08-07 | Kroll Mark W. | System and method for evaluating risk of mortality due to congestive heart failure using physiologic sensors |
| US20040133115A1 (en) * | 2002-11-01 | 2004-07-08 | Hamilton Emily F. | Method and apparatus for identifying heart rate feature events |
| US20160346614A1 (en) * | 2015-05-29 | 2016-12-01 | Nike, Inc. | Calculating energy expenditure from athletic movement attributes |
| CN107291656A (en) * | 2016-04-06 | 2017-10-24 | 奥美之路(北京)健康科技股份有限公司 | The exercise test scheme and computational methods of a kind of cardio-pulmonary function |
| US20190039674A1 (en) * | 2016-02-04 | 2019-02-07 | Bluebrake S.R.L. | System For Assisting In Driving A Bicycle By Sending A Haptic Feedback To A Cyclist |
| CN112023343A (en) * | 2020-08-24 | 2020-12-04 | 北京动亮健康科技有限公司 | Method for adjusting torque of power vehicle, exercise training method and power vehicle |
| CN112384984A (en) * | 2018-07-12 | 2021-02-19 | 苹果公司 | Wearable computer with exercise machine connectivity to improve activity monitoring using calorie expenditure models |
| WO2021042590A1 (en) * | 2019-09-06 | 2021-03-11 | 江苏华康信息技术有限公司 | Electrocardiogram signal quantitative analysis method based on extreme value energy decomposition |
| US20210205661A1 (en) * | 2020-01-02 | 2021-07-08 | bOMDIC, Inc. | Exercise guiding method based on the different fitness performance levels |
| CN113974641A (en) * | 2021-11-03 | 2022-01-28 | 毕胜普生物科技有限公司 | Control method and system of load motion device |
| WO2022179860A1 (en) * | 2021-02-25 | 2022-09-01 | BitifEye Digital Test Solutions GmbH | Apparatus for controlling a training device |
-
2023
- 2023-03-01 CN CN202310187037.9A patent/CN116407088B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030149453A1 (en) * | 2002-02-07 | 2003-08-07 | Kroll Mark W. | System and method for evaluating risk of mortality due to congestive heart failure using physiologic sensors |
| US20040133115A1 (en) * | 2002-11-01 | 2004-07-08 | Hamilton Emily F. | Method and apparatus for identifying heart rate feature events |
| US20160346614A1 (en) * | 2015-05-29 | 2016-12-01 | Nike, Inc. | Calculating energy expenditure from athletic movement attributes |
| US20190039674A1 (en) * | 2016-02-04 | 2019-02-07 | Bluebrake S.R.L. | System For Assisting In Driving A Bicycle By Sending A Haptic Feedback To A Cyclist |
| CN107291656A (en) * | 2016-04-06 | 2017-10-24 | 奥美之路(北京)健康科技股份有限公司 | The exercise test scheme and computational methods of a kind of cardio-pulmonary function |
| CN112384984A (en) * | 2018-07-12 | 2021-02-19 | 苹果公司 | Wearable computer with exercise machine connectivity to improve activity monitoring using calorie expenditure models |
| WO2021042590A1 (en) * | 2019-09-06 | 2021-03-11 | 江苏华康信息技术有限公司 | Electrocardiogram signal quantitative analysis method based on extreme value energy decomposition |
| US20210205661A1 (en) * | 2020-01-02 | 2021-07-08 | bOMDIC, Inc. | Exercise guiding method based on the different fitness performance levels |
| CN112023343A (en) * | 2020-08-24 | 2020-12-04 | 北京动亮健康科技有限公司 | Method for adjusting torque of power vehicle, exercise training method and power vehicle |
| WO2022179860A1 (en) * | 2021-02-25 | 2022-09-01 | BitifEye Digital Test Solutions GmbH | Apparatus for controlling a training device |
| CN113974641A (en) * | 2021-11-03 | 2022-01-28 | 毕胜普生物科技有限公司 | Control method and system of load motion device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116407088B (en) | 2024-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112668128B (en) | Method and device for selecting terminal equipment nodes in federal learning system | |
| CN111091199B (en) | Federal learning method, device and storage medium based on differential privacy | |
| CN111832627B (en) | Image classification model training method, classification method and system for suppressing label noise | |
| CN108364195B (en) | User retention probability prediction method and device, prediction server and storage medium | |
| WO2015103964A1 (en) | Method， apparatus， and device for determining target user | |
| CN108052862B (en) | Age estimation method and device | |
| CN110413754B (en) | Conversational (in) reward evaluation and conversational methods, media, apparatuses, and computing devices | |
| CN107832259A (en) | A kind of load forecasting method based on time series and Kalman filtering | |
| WO2020260656A1 (en) | Pruning and/or quantizing machine learning predictors | |
| CN111814963B (en) | Image recognition method based on deep neural network model parameter modulation | |
| CN112365033B (en) | Wind power interval prediction method, system and storage medium | |
| CN116992779B (en) | Simulation method and system of photovoltaic energy storage system based on digital twin model | |
| JP7315007B2 (en) | LEARNING DEVICE, LEARNING METHOD AND LEARNING PROGRAM | |
| CN109102509B (en) | Segmentation model training method and device and computer readable storage medium | |
| CN116407088B (en) | Exercise heart rate prediction model training method and heart rate prediction method based on power vehicle | |
| CN114840375A (en) | Aging performance testing method and system for semiconductor storage product | |
| CN113035348A (en) | Diabetes diagnosis method based on GRU feature fusion | |
| CN112651500A (en) | Method for generating quantization model and terminal | |
| CN115035304A (en) | Image description generation method and system based on course learning | |
| CN115983320A (en) | Federal learning model parameter quantification method based on deep reinforcement learning | |
| CN116246126A (en) | Iterative unsupervised domain self-adaption method and device | |
| CN115219910A (en) | Analysis method and device for battery residue prediction error | |
| CN113435927B (en) | User willingness prediction method, device, equipment and storage medium | |
| CN114648178A (en) | Operation and maintenance strategy optimization method of electric energy metering device based on DDPG algorithm | |
| CN112132310A (en) | Power equipment state estimation method and device based on improved LSTM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |