CN112582067A - Age estimation model training and age estimation method and device based on big data - Google Patents
Age estimation model training and age estimation method and device based on big data Download PDFInfo
- Publication number
- CN112582067A CN112582067A CN202011521753.9A CN202011521753A CN112582067A CN 112582067 A CN112582067 A CN 112582067A CN 202011521753 A CN202011521753 A CN 202011521753A CN 112582067 A CN112582067 A CN 112582067A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- user
- age
- walking
- time period
- 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.)
- Withdrawn
Links
Images
Classifications
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The application provides an age estimation model training and age estimation method and device based on big data, and relates to the technical field of data processing, wherein the method comprises the following steps: acquiring dynamic data information samples of a user in a target time period through wearable equipment; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting a basic characteristic information sample and a multi-dimensional index characteristic sample of a user into a deep neural network for training to obtain a training age; and adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model. Therefore, the age estimation model is generated based on the dynamic data information collected by the wearable device and the basic information of the user, and the health age estimation efficiency and accuracy of the user are improved.
Description
Technical Field
The application relates to the technical field of data processing, in particular to an age estimation model training and age estimation method and device based on big data.
Background
Generally, "healthy age", also called "physiological age" or "biological age", generally refers to the level of physiological and functional responses when a living body reaches a certain chronological age, i.e. the degree of physiological and functional responses corresponding to a certain chronological age, which is measured from the medical and biological perspectives. It represents the degree of growth, maturation or aging of an organism, and is an age manifestation of a healthy condition. The general conceptual age is usually referred to as "natural age", which is also called "chronological age", "calendar age" or "chronological age", and is an age calculated purely from the passage of time according to the birth year and month of a person. It is characterized by that the life mileage and life duration of a person are not limited by life experience and life condition of the person, and can be increased with the lapse of time.
In the related art, the health age is estimated based on the maximum oxygen uptake, however, the maximum oxygen uptake refers to the content of oxygen which can be taken when the human body does exercises with the maximum intensity and continuously supports the following exercises without force, the direct measurement of the maximum oxygen uptake needs to allow the user to carry a related acquisition instrument to do strenuous exercise, the exhaled gas of the testee in the exhaustion state is analyzed, the direct measurement mode is complex, the user is required to reach the limit state of the body through the exercise, and the health age is estimated only by taking the maximum oxygen uptake as a quantitative index, and the method is single and comprehensive to a certain extent.
Disclosure of Invention
The present application aims to solve at least to some extent one of the above mentioned technical problems.
Therefore, the first purpose of the application is to provide an age estimation model training method based on big data, the technical problems of complexity, low efficiency and poor precision of a health age estimation mode in the prior art are solved, an age estimation model is generated through dynamic data information acquired based on wearable equipment, and the health age estimation efficiency and accuracy of a user are improved.
A second objective of the present application is to provide an age estimation method based on an age estimation model of big data.
A third objective of the present application is to provide an age estimation model training apparatus based on big data.
A fourth object of the present application is to provide an age estimation device based on an age estimation model of big data.
A fifth object of the present application is to propose a computer device.
A sixth object of the present application is to propose a non-transitory computer-readable storage medium.
In order to achieve the above object, a first embodiment of the present application provides an age estimation model training method based on big data, including: acquiring dynamic data information samples of a user in a target time period through wearable equipment; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting the basic characteristic information sample of the user and the multi-dimensional index characteristic sample into a deep neural network for training to obtain a training age; and adjusting the parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model.
In one embodiment of the present application, acquiring, by a wearable device, a dynamic data information sample of a user over a target time period includes at least two or more of the following: collecting the resting heart rate of the user in a non-sleep inactive state within the target time period; collecting the heart rate of the user in a sleep state in the target time period; collecting the heart rate of the user in the walking state in the target time period; acquiring the heart rate of the user in a running state within the target time period; collecting sleep time periods, deep sleep time periods and average sleep time periods of the user in the target time period; collecting a personal athletic performance index, pai, (personal Activity inteligence) of the user within the target time period; acquiring the total walking distance, the total walking time and the total walking steps of the user in the target time period; and acquiring the total running distance, the total running duration and the total running steps of the user in the target time period.
In an embodiment of the present application, the analyzing the dynamic data information sample to obtain a multi-dimensional index feature sample includes: calculating according to the resting heart rate of the user in the non-sleep inactive state to obtain the resting average heart rate, the resting high heart rate and the resting low heart rate; calculating according to the heart rate of the user in the sleep state to obtain the average sleep heart rate, the high-position sleep heart rate and the low-position sleep heart rate; calculating according to the heart rate of the user in the walking state, and acquiring the walking average heart rate, the walking high-level heart rate and the walking low-level heart rate; calculating according to the heart rate of the user in the running state, and acquiring the running average heart rate, the running high-level heart rate and the running low-level heart rate; calculating according to the sleep time period, the deep sleep time period and the average sleep time period of the user to obtain a core sleep time length and a deep sleep time length; generating the multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal motion index PAI, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps.
In an embodiment of the present application, the adjusting parameters of the deep neural network according to the training age and the real age through a preset loss metric formula to generate an age estimation model includes: decoding the training age to obtain a target dimension index characteristic, and calculating the target dimension index characteristic and a first loss measure of the multi-dimension index characteristic sample through a preset first loss measure formula; calculating a second loss metric of the training age and the real age through a preset second loss metric formula; and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
To achieve the above object, a second aspect of the present application provides an age estimation method based on an age estimation model of big data, including: acquiring dynamic data information of a user in a target time period through wearable equipment; analyzing the dynamic data to obtain multi-dimensional index characteristics; and inputting the basic characteristic information of the user and the multi-dimensional index characteristics into a trained age prediction model for processing to obtain the predicted age.
To achieve the above object, a third aspect of the present application provides an age estimation model training apparatus based on big data, including: the first acquisition module is used for acquiring dynamic data information samples of a user in a target time period through the wearable equipment; the analysis acquisition module is used for analyzing the dynamic data information sample to acquire a multi-dimensional index characteristic sample; the training acquisition module is used for inputting the basic characteristic information sample of the user and the multi-dimensional index characteristic sample into a deep neural network for training to acquire a training age; and the generation module is used for adjusting the parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model.
In an embodiment of the present application, the first acquisition module is specifically configured to at least two or more of the following combinations: collecting the resting heart rate of the user in a non-sleep inactive state within the target time period; collecting the heart rate of the user in a sleep state in the target time period; collecting the heart rate of the user in the walking state in the target time period; acquiring the heart rate of the user in a running state within the target time period; collecting sleep time periods, deep sleep time periods and average sleep time periods of the user in the target time period; collecting a personal athletic performance index (PAI) of the user in the target time period; acquiring the total walking distance, the total walking time and the total walking steps of the user in the target time period; and acquiring the total running distance, the total running duration and the total running steps of the user in the target time period.
In an embodiment of the application, the analysis acquisition module is specifically configured to: calculating according to the resting heart rate of the user in the non-sleep inactive state to obtain the resting average heart rate, the resting high heart rate and the resting low heart rate; calculating according to the heart rate of the user in the sleep state to obtain the average sleep heart rate, the high-position sleep heart rate and the low-position sleep heart rate; calculating according to the heart rate of the user in the walking state, and acquiring the walking average heart rate, the walking high-level heart rate and the walking low-level heart rate; calculating according to the heart rate of the user in the running state, and acquiring the running average heart rate, the running high-level heart rate and the running low-level heart rate; calculating according to the sleep time period, the deep sleep time period and the average sleep time period of the user to obtain a core sleep time length and a deep sleep time length; generating the multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal motion index PAI, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps.
In an embodiment of the application, the generating module is configured to: decoding the training age to obtain a target dimension index characteristic, and calculating the target dimension index characteristic and a first loss measure of the multi-dimension index characteristic sample through a preset first loss measure formula; calculating a second loss metric of the training age and the real age through a preset second loss metric formula; and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
To achieve the above object, a fourth aspect of the present application provides an age estimation device based on an age estimation model of big data, including: the second acquisition module is used for acquiring dynamic data information of the user in a target time period through the wearable equipment; the first acquisition module is used for analyzing the dynamic data to acquire multi-dimensional index characteristics; and the second acquisition module is used for inputting the basic characteristic information of the user and the multi-dimensional index characteristics into a trained age prediction model for processing to acquire the predicted age.
To achieve the above object, a fifth embodiment of the present application provides a computer device, including: a memory, a processor and a computer program stored on the memory and executable on the processor, when executing the computer program, implementing the age estimation model training, age estimation method based on big data as described in the above embodiments.
In order to achieve the above object, a sixth aspect of the present application provides a non-transitory computer-readable storage medium, wherein instructions of the storage medium, when executed by a processor, enable the processor to implement the big data based age estimation model training and age estimation method as described in the above embodiments.
The technical scheme provided by the application at least has the following beneficial technical effects:
acquiring dynamic data information samples of a user in a target time period through wearable equipment; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting a basic characteristic information sample and a multi-dimensional index characteristic sample of a user into a deep neural network for training to obtain a training age; and adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model. Therefore, the age estimation model is generated based on the dynamic data information collected by the wearable device and the basic information of the user, and the health age estimation efficiency and accuracy of the user are improved.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flowchart of an age estimation model training method based on big data according to an embodiment of the present disclosure;
FIG. 2 is a schematic flowchart of another big data-based age estimation model training method according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of model training provided in an embodiment of the present application;
fig. 4 is a schematic flowchart of an age estimation method based on an age estimation model of big data according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an age estimation model training apparatus based on big data according to an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of an age estimation apparatus based on an age estimation model of big data according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
Age estimation model training based on big data, age estimation method and apparatus according to embodiments of the present application are described below with reference to the accompanying drawings. The execution main body of the age estimation model training and age estimation method based on big data can be any portable terminal device, the terminal device can be a mobile phone, a tablet personal computer, a personal digital assistant, wearable equipment and other hardware devices with various operating systems, and the wearable equipment can be an intelligent bracelet, an intelligent watch, intelligent glasses and the like.
Fig. 1 is a schematic flowchart of an age estimation model training method based on big data according to an embodiment of the present disclosure. As shown in fig. 1, the age estimation model training method based on big data includes:
In this application embodiment, can gather heart rate, sleep, activity, step number, motion pattern data etc. of user in the target time quantum as dynamic data information sample through wearable equipment, in order to improve the accuracy of model, the user of this application can be a plurality ofly.
In the embodiment of the present application, the target time period may be set according to an application scenario, such as a day.
In the embodiment of the present application, there are many ways to collect the dynamic data information sample of the user in the target time period through the wearable device, which are illustrated as follows.
A first example, a resting heart rate in a non-sleep inactive state, a heart rate in a sleep state, a heart rate in a walking state, and a heart rate in a running state of a user over a target time period are collected.
Second, a sleep period, a deep sleep period, an average sleep period, and a personal motor index PAI of a user within a target period are collected.
The related patent CN107077523A calculates the personal motor function index PAI from the age, sex, resting heart rate and real-time heart rate by equations (1) - (8).
Specifically, the method comprises the following steps:
V＝a2,1,3+a2,1,4(1-e-Z) (2)
HRth＝RHR+HRR×0.2 (6)
HRR＝MHR-RHR (7)
MHR＝a2,1,6-a2,1,7×age (8)
wherein HR (t) is real-time heart rate; t is the integration time; RHR represents the user's resting heart rate; age represents the age of the user; { a2,1,iI 1,2, …,7 is a set of coefficients that need to be statistically calibrated for different ethnic groups.
In a third example, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps of the user within the target time period are collected.
And 102, analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample.
In the embodiment of the application, different dynamic data information samples can be analyzed in different modes to obtain a multi-dimensional index feature sample, and the setting is selected according to an application scenario, which is exemplified as follows.
A first example, calculating according to the resting heart rate of the user in the non-sleep inactive state, obtaining the resting average heart rate, the resting high heart rate and the resting low heart rate, calculating according to the heart rate of the user in the sleep state, obtaining the sleeping average heart rate, the sleeping high heart rate and the sleeping low heart rate, calculating according to the heart rate of the user in the walking state, obtaining the walking average heart rate, the walking high heart rate and the walking low heart rate, calculating according to the heart rate of the user in the running state, obtaining the running average heart rate, the running high heart rate and the running low heart rate, calculating according to the sleep time periods, the deep sleep time periods and the average sleep time periods of the user, obtaining the core sleep time periods and the deep sleep time periods, calculating according to the resting average heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping average heart rate, The method comprises the steps of generating a multi-dimensional index characteristic sample by using sleep low heart rate, walking average heart rate, walking high heart rate, walking low heart rate, running average heart rate, running high heart rate, running low heart rate, personal motion function index PAI, total walking distance, total walking time, total walking step number, total running distance, total running time and total running step number.
In a second example, a multi-dimensional index feature sample is generated according to the personal motion function index PAI, the total walking distance, the total walking time, the total walking steps, the total running distance, the total running time and the total running steps.
It should be noted that all or all departments can be selected to generate corresponding dimensional index feature samples as needed, and it can be understood that the more dimensional index feature samples, the more accurate the training model.
And 103, inputting the basic characteristic information sample and the multi-dimensional index characteristic sample of the user into a deep neural network for training to obtain a training age.
In the embodiment of the present application, the basic information refers to basic characteristics that are stable for a long period of time and a medium period of time of the user, and includes: gender, BMI (Body Mass Index), natural age interval.
It should be noted that, in the input information of the machine learning framework of the present application, which uses the natural age as the real age, the real age of the user cannot be directly used, otherwise, the model may finally converge to have all the weights of the input items except the natural age as 0, and directly output the input real age. Thus, the present application defines the real age interval as input: the true age interval is floor (max (0, natural age-1) ÷ 12) × 12+6, where max is the max function and floor is the rounded down function.
And step 104, adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula, and generating an age estimation model.
In the embodiment of the present application, there are various ways to generate the age estimation model by adjusting parameters of the deep neural network according to the training age and the real age through a preset loss metric formula, which are described as follows.
The first example comprises the steps of decoding a training age to obtain a target dimension index characteristic, and calculating first loss measurement of the target dimension index characteristic and a multi-dimension index characteristic sample through a preset first loss measurement formula; calculating a second loss metric of the training age and the real age through a preset second loss metric formula; and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
In the second example, the age estimation model is generated directly according to the residual error between the training age and the real age through a preset loss measurement formula and according to the residual error and the parameter of the threshold value adjustment deep neural network.
In summary, according to the age estimation model training method based on big data, dynamic data information samples of a user in a target time period are collected through wearable equipment; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting a basic characteristic information sample and a multi-dimensional index characteristic sample of a user into a deep neural network for training to obtain a training age; and adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model. Therefore, the age estimation model is generated based on the dynamic data information collected by the wearable device and the basic information of the user, and the health age estimation efficiency and accuracy of the user are improved.
In order to make the above process more clear to those skilled in the art, the following description is made in detail with reference to fig. 2.
Fig. 2 is a schematic flowchart of another age estimation model training method based on big data according to an embodiment of the present disclosure. As shown in fig. 2, the age estimation model training method based on big data includes:
In the embodiment of the application, the resting average heart rate in the non-sleep and non-active state is obtained through average calculation, 99% of bits of the heart rate in the non-sleep and non-active state are resting high-position heart rate, and 1% of bits of the heart rate in the non-sleep and non-active state are resting low-position heart rate.
In the embodiment of the application, the average calculation is performed on the heart rate in the sleep state to obtain the sleep average heart rate, the 99% bit of the heart rate in the sleep state is the sleep high-position heart rate, and the 1% bit of the heart rate in the sleep state is the sleep low-position heart rate.
In the embodiment of the application, the heart rate in the walking state is averaged to obtain the walking average heart rate, 99% of the number of the heart rate in the walking state is the walking high-position heart rate, and 1% of the number of the heart rate in the walking state is the walking low-position heart rate.
In the embodiment of the application, the average calculation is carried out on the heart rate in the running state to obtain the running average heart rate, the 99% digit of the heart rate in the running state is the running high-level heart rate, and the 1% digit of the heart rate in the running state is the running low-level heart rate.
In the embodiment of the application, the overlapping duration of the sleep period and the average sleep period is the core sleep duration; the overlapping duration of the deep sleep time period and the average sleep time period is the deep sleep time period.
And step 206, acquiring the personal motion function index PA, the total walking distance, the total walking time length, the total walking step number, the total running distance, the total running time length and the total running step number of the user in the target time period.
And step 207, generating a multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal athletic performance index PAI, the total walking distance, the total walking time, the total walking step number, the total running distance, the total running time and the total running step number.
In the embodiment of the application, one or more multidimensional index feature samples can be selected according to application needs to perform subsequent model training, and it can be understood that the more the multidimensional index feature samples are, the more accurate the model is trained.
And step 208, inputting the basic characteristic information sample and the multi-dimensional index characteristic sample of the user into a deep neural network for training to obtain a training age.
In the embodiment of the present application, the basic information refers to basic characteristics that are stable for a long period of time and a medium period of time of the user, and includes: gender, BMI (Body Mass Index), natural age interval.
It should be noted that, in the input information of the machine learning framework of the present application, which uses the natural age as the real age, the real age of the user cannot be directly used, otherwise, the model may finally converge to have all the weights of the input items except the natural age as 0, and directly output the input real age. Thus, the present application defines the real age interval as input: the true age interval is floor (max (0, natural age-1) ÷ 12) × 12+6, where max is the max function and floor is the rounded down function.
And 209, decoding the training age to obtain a target dimension index characteristic, and calculating the first loss measurement of the target dimension index characteristic and the multi-dimension index characteristic sample through a preset first loss measurement formula.
In the embodiment of the application, dynamic data information acquired by means of wearable equipment and the like is combined with basic information of a user and mapped to the healthy age, in the training process of the age estimation model, the output result of the model is based on the real age of the user, and the residual error of the model and the real age of the user is used as the model loss to feed back the iterative machine model until convergence.
It should be noted that, the trained age estimation model can map the individual features to the population space, associate the population corresponding to the individuals with different dynamic data information and basic information, and quantify the individual health age by the real age of the corresponding population by using the basic assumption that most individuals in the population have health ages consistent with the real age.
Specifically, as shown in fig. 3, mapping the individual characteristics of the user, that is, the basic characteristic information sample and the multi-dimensional index characteristic sample, to the health age of the user, that is, the individual characteristics specifically include 24 dimensions, that is, the 3-dimensional basic characteristic information sample and the 21-dimensional index characteristic sample described above are spliced in series; the encoder and the decoder are symmetrical multi-layer (such as 5 layers of full connection + RELU (Rectified Linear Unit) deep neural network structures, wherein the input and the output of the encoder are respectively 24-dimensional and 1-dimensional, and the input and the output of the decoder are respectively 1-dimensional and 24-dimensional.
In the training process of the age estimation model, a preset loss measurement formula is adopted for model loss. Wherein the first loss metric is used to measure the reconstruction error of the 24-dimensional output of the decoder for 24-dimensional individual features, i.e. the loss metric 1-Eq(z|x)[logP(x|z)]The loss metric 2 ═ KL (q (z | x) | | p (z)).
Where x, z represent the input and output of the encoder, respectively, minimizing the loss metric 1, i.e. given an encoder output q (z | x), the larger the decoder output P (x | z), the better, which corresponds to minimizing the reconstruction error; minimizing the loss metric 2 to approximate a posteriori distribution approximation q (z | x) to a prior distribution p (z); that is to say that the code z generated by q (z | x) cannot be too far out of spectrum to be compared with a certain distribution, thereby limiting the intermediate code generation. Optionally, the present application sets p (z) to be the same as the prior distribution of true age.
In summary, according to the age estimation model training method based on big data, dynamic data information samples of a user in a target time period are collected through wearable equipment; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting a basic characteristic information sample and a multi-dimensional index characteristic sample of a user into a deep neural network for training to obtain a training age; and adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model. Therefore, the age estimation model is generated based on the dynamic data information collected by the wearable device and the basic information of the user, and the health age estimation efficiency and accuracy of the user are improved.
Fig. 4 is a flowchart illustrating an age estimation method based on an age estimation model of big data according to an embodiment of the present disclosure. As shown in fig. 1, the age estimation method of the age estimation model based on big data includes:
In the embodiment of the application, the heart rate, the sleep, the activity amount, the step number, the motion pattern data and the like of the user in the target time period can be collected by the wearable device as dynamic data information.
In the embodiment of the present application, the target time period may be set according to an application scenario, such as a day.
In the embodiment of the present application, there are many ways to collect the dynamic data information of the user in the target time period through the wearable device, which are illustrated as follows.
A first example, a resting heart rate in a non-sleep inactive state, a heart rate in a sleep state, a heart rate in a walking state, and a heart rate in a running state of a user over a target time period are collected.
Second, a sleep period, a deep sleep period, an average sleep period, and a personal motor index PAI of a user within a target period are collected.
In a third example, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps of the user within the target time period are collected.
And step 302, analyzing the dynamic data to obtain multi-dimensional index characteristics.
In the embodiment of the application, different dynamic data information may be analyzed in different ways to obtain multi-dimensional index features, and the setting is selected according to an application scenario, for example, as follows.
A first example, calculating according to the resting heart rate of the user in the non-sleep inactive state, obtaining the resting average heart rate, the resting high heart rate and the resting low heart rate, calculating according to the heart rate of the user in the sleep state, obtaining the sleeping average heart rate, the sleeping high heart rate and the sleeping low heart rate, calculating according to the heart rate of the user in the walking state, obtaining the walking average heart rate, the walking high heart rate and the walking low heart rate, calculating according to the heart rate of the user in the running state, obtaining the running average heart rate, the running high heart rate and the running low heart rate, calculating according to the sleep time periods, the deep sleep time periods and the average sleep time periods of the user, obtaining the core sleep time periods and the deep sleep time periods, calculating according to the resting average heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping average heart rate, The multi-dimensional index features are generated according to the sleep low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal motion function index PAI, the total walking distance, the total walking time, the total walking step number, the total running distance, the total running time and the total running step number.
In a second example, a multi-dimensional index feature is generated according to the personal motion function index PAI, the total walking distance, the total walking time, the total walking steps, the total running distance, the total running time and the total running steps.
It should be noted that all or the departments may be selected to generate corresponding dimension index features as needed, and it can be understood that the more dimension index features, the more accurate the age estimation is.
In the embodiment of the present application, the basic information refers to basic characteristics that are stable for a long period of time and a medium period of time of the user, and includes: gender, BMI (Body Mass Index), natural age interval.
It should be noted that, therefore, the present application defines the real age interval as an input: the true age interval is floor (max (0, natural age-1) ÷ 12) × 12+6, where max is the max function and floor is the rounded down function.
And step 303, inputting the basic feature information and the multi-dimensional index features of the user into the trained age prediction model for processing to obtain the predicted age.
In the embodiment of the application, the basic characteristic information and the multi-dimensional index characteristic sample are connected in series and spliced to be input into a trained age prediction model for processing, and the predicted age, namely the health age of the user, is output.
Therefore, dynamic data information of the user in a target time period is collected through the wearable device, dynamic data are analyzed, multi-dimensional index features are obtained, basic feature information and the multi-dimensional index features of the user are input into a trained age prediction model to be processed, and the predicted age is obtained. From this, carry out age estimation through age estimation model based on the good user's basic information of dynamic data information that wearable equipment gathered, improve user's healthy age estimation efficiency and accuracy.
In order to implement the above embodiments, the present application further provides an age estimation model training apparatus based on big data.
Fig. 5 is a schematic structural diagram of an age estimation model training apparatus based on big data according to an embodiment of the present application.
As shown in fig. 5, the age estimation model training apparatus based on big data includes: a first acquisition module 510, an analysis acquisition module 520, a training acquisition module 530, and a generation module 540. Wherein the content of the first and second substances,
a first collecting module 510, configured to collect, by the wearable device, dynamic data information samples of the user within a target time period.
An analysis obtaining module 520, configured to analyze the dynamic data information sample to obtain a multi-dimensional index feature sample.
A training obtaining module 530, configured to input the basic feature information sample of the user and the multidimensional index feature sample into a deep neural network for training, so as to obtain a training age.
And the generating module 540 is configured to adjust parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula, so as to generate an age estimation model.
In an embodiment of the present application, the first acquiring module 510 is specifically configured to: collecting the resting heart rate of the user in a non-sleep inactive state within the target time period; and/or acquiring the heart rate of the user in a sleep state in the target time period; and/or acquiring the heart rate of the user in the walking state in the target time period; and/or, acquiring the heart rate of the user in a running state in the target time period; and/or, collecting the sleep time period, the deep sleep time period and the average sleep time period of the user in the target time period; and/or, collecting the personal motor function index PAI of the user in the target time period; and/or collecting the total walking distance, the total walking time and the total walking steps of the user in the target time period; and/or collecting the total running distance, the total running duration and the total running steps of the user in the target time period.
In an embodiment of the present application, the analysis obtaining module 520 is specifically configured to: calculating according to the resting heart rate of the user in the non-sleep inactive state to obtain the resting average heart rate, the resting high heart rate and the resting low heart rate; calculating according to the heart rate of the user in the sleep state to obtain the average sleep heart rate, the high-position sleep heart rate and the low-position sleep heart rate; calculating according to the heart rate of the user in the walking state, and acquiring the walking average heart rate, the walking high-level heart rate and the walking low-level heart rate; calculating according to the heart rate of the user in the running state, and acquiring the running average heart rate, the running high-level heart rate and the running low-level heart rate; calculating according to the sleep time period, the deep sleep time period and the average sleep time period of the user to obtain the core sleep time length and the deep sleep time length; and generating a multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal athletic function index PAI, the total walking distance, the total walking time length, the total walking step number, the total running distance, the total running time length and the total running step number.
In an embodiment of the present application, the generating module 540 is configured to: decoding the training age to obtain a target dimension index characteristic, and calculating the target dimension index characteristic and a first loss measure of the multi-dimension index characteristic sample through a preset first loss measure formula; calculating a second loss metric of the training age and the real age through a preset second loss metric formula; and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
It should be noted that the foregoing explanation of the embodiment of the age estimation model training method based on big data is also applicable to the age estimation model training apparatus based on big data of this embodiment, and is not repeated here.
In summary, the age estimation model training device based on big data of the embodiment collects dynamic data information samples of the user in a target time period through the wearable device; analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample; inputting a basic characteristic information sample and a multi-dimensional index characteristic sample of a user into a deep neural network for training to obtain a training age; and adjusting parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model. Therefore, the age estimation model is generated based on the dynamic data information collected by the wearable device and the basic information of the user, and the health age estimation efficiency and accuracy of the user are improved.
In order to implement the above embodiments, the present application also provides an age estimation device based on an age estimation model of big data.
Fig. 6 is a schematic structural diagram of an age estimation apparatus based on an age estimation model of big data according to an embodiment of the present application.
As shown in fig. 6, the age estimation apparatus of the age estimation model based on big data includes: a second acquisition module 610, a first acquisition module 620, and a second acquisition module 630. Wherein the content of the first and second substances,
a second collecting module 610, configured to collect, by the wearable device, dynamic data information of the user in a target time period.
A first obtaining module 620, configured to analyze the dynamic data and obtain a multidimensional index feature.
A second obtaining module 630, configured to input the basic feature information of the user and the multidimensional index feature into a trained age prediction model for processing, so as to obtain a predicted age.
It should be noted that the above explanation of the embodiment of the age estimation method based on the age estimation model based on big data is also applicable to the age estimation device based on the age estimation model based on big data of this embodiment, and is not repeated here.
Therefore, dynamic data information of the user in a target time period is collected through the wearable device, dynamic data are analyzed, multi-dimensional index features are obtained, basic feature information and the multi-dimensional index features of the user are input into a trained age prediction model to be processed, and the predicted age is obtained. From this, carry out age estimation through age estimation model based on the good user's basic information of dynamic data information that wearable equipment gathered, improve user's healthy age estimation efficiency and accuracy.
In order to implement the foregoing embodiments, the present application further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the age estimation model training and age estimation method based on big data as described in the foregoing embodiments are implemented.
To achieve the above embodiments, the present application also proposes a non-transitory computer-readable storage medium, in which instructions are executed by a processor to enable execution of the age estimation model training and age estimation method based on big data described in the above embodiments.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (12)
1. An age estimation model training method based on big data is characterized by comprising the following steps:
acquiring dynamic data information samples of a user in a target time period through wearable equipment;
analyzing the dynamic data information sample to obtain a multi-dimensional index characteristic sample;
inputting the basic characteristic information sample of the user and the multi-dimensional index characteristic sample into a deep neural network for training to obtain a training age;
and adjusting the parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model.
2. The method of claim 1, wherein the collecting, by the wearable device, the dynamic data information sample of the user over the target time period comprises at least two or more of the following in combination:
collecting the resting heart rate of the user in a non-sleep inactive state within the target time period;
collecting the heart rate of the user in a sleep state in the target time period;
collecting the heart rate of the user in the walking state in the target time period;
acquiring the heart rate of the user in a running state within the target time period;
collecting sleep time periods, deep sleep time periods and average sleep time periods of the user in the target time period;
collecting a personal athletic performance index (PAI) of the user in the target time period;
acquiring the total walking distance, the total walking time and the total walking steps of the user in the target time period;
and acquiring the total running distance, the total running duration and the total running steps of the user in the target time period.
3. The method of claim 2, wherein analyzing the dynamic data information samples to obtain multi-dimensional metric feature samples comprises:
calculating according to the resting heart rate of the user in the non-sleep inactive state to obtain the resting average heart rate, the resting high heart rate and the resting low heart rate;
calculating according to the heart rate of the user in the sleep state to obtain the average sleep heart rate, the high-position sleep heart rate and the low-position sleep heart rate;
calculating according to the heart rate of the user in the walking state, and acquiring the walking average heart rate, the walking high-level heart rate and the walking low-level heart rate;
calculating according to the heart rate of the user in the running state, and acquiring the running average heart rate, the running high-level heart rate and the running low-level heart rate;
calculating according to the sleep time period, the deep sleep time period and the average sleep time period of the user to obtain a core sleep time length and a deep sleep time length;
generating the multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal motion index PAI, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps.
4. The method of claim 1, wherein the adjusting parameters of the deep neural network according to the training age and the real age through a preset loss metric formula to generate an age estimation model comprises:
decoding the training age to obtain a target dimension index characteristic, and calculating the target dimension index characteristic and a first loss measure of the multi-dimension index characteristic sample through a preset first loss measure formula;
calculating a second loss metric of the training age and the real age through a preset second loss metric formula;
and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
5. An age estimation method using the big data based age estimation model according to any one of claims 1 to 4, comprising:
acquiring dynamic data information of a user in a target time period through wearable equipment;
analyzing the dynamic data to obtain multi-dimensional index characteristics;
and inputting the basic characteristic information of the user and the multi-dimensional index characteristics into a trained age prediction model for processing to obtain the predicted age.
6. An age estimation model training device based on big data, comprising:
the first acquisition module is used for acquiring dynamic data information samples of a user in a target time period through the wearable equipment;
the analysis acquisition module is used for analyzing the dynamic data information sample to acquire a multi-dimensional index characteristic sample;
the training acquisition module is used for inputting the basic characteristic information sample of the user and the multi-dimensional index characteristic sample into a deep neural network for training to acquire a training age;
and the generation module is used for adjusting the parameters of the deep neural network according to the training age and the real age through a preset loss measurement formula to generate an age estimation model.
7. The apparatus of claim 6, wherein the first acquisition module is specifically configured to at least two or more of the following:
collecting the resting heart rate of the user in a non-sleep inactive state within the target time period;
collecting the heart rate of the user in a sleep state in the target time period;
collecting the heart rate of the user in the walking state in the target time period;
acquiring the heart rate of the user in a running state within the target time period;
collecting sleep time periods, deep sleep time periods and average sleep time periods of the user in the target time period;
collecting a personal athletic performance index (PAI) of the user in the target time period;
acquiring the total walking distance, the total walking time and the total walking steps of the user in the target time period;
and acquiring the total running distance, the total running duration and the total running steps of the user in the target time period.
8. The apparatus of claim 7, wherein the analysis acquisition module is specifically configured to:
calculating according to the resting heart rate of the user in the non-sleep inactive state to obtain the resting average heart rate, the resting high heart rate and the resting low heart rate;
calculating according to the heart rate of the user in the sleep state to obtain the average sleep heart rate, the high-position sleep heart rate and the low-position sleep heart rate;
calculating according to the heart rate of the user in the walking state, and acquiring the walking average heart rate, the walking high-level heart rate and the walking low-level heart rate;
calculating according to the heart rate of the user in the running state, and acquiring the running average heart rate, the running high-level heart rate and the running low-level heart rate;
calculating according to the sleep time period, the deep sleep time period and the average sleep time period of the user to obtain a core sleep time length and a deep sleep time length;
generating the multi-dimensional index feature sample according to one or more of the resting average heart rate, the resting high heart rate, the resting low heart rate, the sleeping average heart rate, the sleeping high heart rate, the sleeping low heart rate, the walking average heart rate, the walking high heart rate, the walking low heart rate, the running average heart rate, the running high heart rate, the running low heart rate, the personal motion index PAI, the total walking distance, the total walking duration, the total walking steps, the total running distance, the total running duration and the total running steps.
9. The apparatus of claim 6, wherein the generation module is to:
decoding the training age to obtain a target dimension index characteristic, and calculating the target dimension index characteristic and a first loss measure of the multi-dimension index characteristic sample through a preset first loss measure formula;
calculating a second loss metric of the training age and the real age through a preset second loss metric formula;
and adjusting parameters of the deep neural network, and generating an age estimation model when the first loss metric and the second loss metric meet a preset threshold condition.
10. An age estimation apparatus to which the big data based age estimation model according to any one of claims 6 to 8 is applied, comprising:
the second acquisition module is used for acquiring dynamic data information of the user in a target time period through the wearable equipment;
the first acquisition module is used for analyzing the dynamic data to acquire multi-dimensional index characteristics;
and the second acquisition module is used for inputting the basic characteristic information of the user and the multi-dimensional index characteristics into a trained age prediction model for processing to acquire the predicted age.
11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1-5 when executing the computer program.
12. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011521753.9A CN112582067A (en) | 2020-12-21 | 2020-12-21 | Age estimation model training and age estimation method and device based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011521753.9A CN112582067A (en) | 2020-12-21 | 2020-12-21 | Age estimation model training and age estimation method and device based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112582067A true CN112582067A (en) | 2021-03-30 |
Family
ID=75136442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011521753.9A Withdrawn CN112582067A (en) | 2020-12-21 | 2020-12-21 | Age estimation model training and age estimation method and device based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112582067A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114469663A (en) * | 2022-03-24 | 2022-05-13 | 河南护眼优品实业有限公司 | Intelligent training method and device for preventing and controlling eyesight improvement |
| CN116726470A (en) * | 2022-10-14 | 2023-09-12 | 荣耀终端有限公司 | Physical ability age estimation method and electronic equipment |
| WO2023201285A3 (en) * | 2022-04-14 | 2023-11-23 | Juvyou (Europe) Limited | Computer-implemented systems and methods for health data analysis and management |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105748043A (en) * | 2016-02-17 | 2016-07-13 | 安徽华米信息科技有限公司 | Sleep quality mentoring method and device and wearable device |
| CN108185996A (en) * | 2017-12-27 | 2018-06-22 | 中国科学院深圳先进技术研究院 | Arteries age appraising model construction method and device |
| CN108992047A (en) * | 2018-07-18 | 2018-12-14 | 上海斐讯数据通信技术有限公司 | A kind of detection method and terminal of resting heart rate |
| CN109009017A (en) * | 2018-07-03 | 2018-12-18 | 中南大学 | A kind of intelligent health monitoring system and its data processing method |
| CN109192315A (en) * | 2018-06-23 | 2019-01-11 | 重庆大学 | Synthesis age detection system based on weighting kernel regression and packaged type offset search |
| CN110674397A (en) * | 2019-08-30 | 2020-01-10 | 北京百度网讯科技有限公司 | Method, device, equipment and readable medium for training age point prediction model |
| CN110710960A (en) * | 2019-09-12 | 2020-01-21 | 东莞市康助医疗科技有限公司 | Arterial blood vessel age estimation method and device |
| CN111126207A (en) * | 2019-12-12 | 2020-05-08 | 北京金山云网络技术有限公司 | Training method and device of age estimation model and electronic equipment |
| CN111210914A (en) * | 2019-12-04 | 2020-05-29 | 北京沃东天骏信息技术有限公司 | Method and device for inquiry reminding based on wearable equipment |
| CN111260186A (en) * | 2020-01-08 | 2020-06-09 | 缤刻普达（北京）科技有限责任公司 | Exercise capacity evaluation method and system, body fat scale and mobile terminal |
| CN111544853A (en) * | 2020-05-13 | 2020-08-18 | 广东高驰运动科技有限公司 | Physical index evaluation method and equipment in running exercise |
-
2020
- 2020-12-21 CN CN202011521753.9A patent/CN112582067A/en not_active Withdrawn
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105748043A (en) * | 2016-02-17 | 2016-07-13 | 安徽华米信息科技有限公司 | Sleep quality mentoring method and device and wearable device |
| CN108185996A (en) * | 2017-12-27 | 2018-06-22 | 中国科学院深圳先进技术研究院 | Arteries age appraising model construction method and device |
| CN109192315A (en) * | 2018-06-23 | 2019-01-11 | 重庆大学 | Synthesis age detection system based on weighting kernel regression and packaged type offset search |
| CN109009017A (en) * | 2018-07-03 | 2018-12-18 | 中南大学 | A kind of intelligent health monitoring system and its data processing method |
| CN108992047A (en) * | 2018-07-18 | 2018-12-14 | 上海斐讯数据通信技术有限公司 | A kind of detection method and terminal of resting heart rate |
| CN110674397A (en) * | 2019-08-30 | 2020-01-10 | 北京百度网讯科技有限公司 | Method, device, equipment and readable medium for training age point prediction model |
| CN110710960A (en) * | 2019-09-12 | 2020-01-21 | 东莞市康助医疗科技有限公司 | Arterial blood vessel age estimation method and device |
| CN111210914A (en) * | 2019-12-04 | 2020-05-29 | 北京沃东天骏信息技术有限公司 | Method and device for inquiry reminding based on wearable equipment |
| CN111126207A (en) * | 2019-12-12 | 2020-05-08 | 北京金山云网络技术有限公司 | Training method and device of age estimation model and electronic equipment |
| CN111260186A (en) * | 2020-01-08 | 2020-06-09 | 缤刻普达（北京）科技有限责任公司 | Exercise capacity evaluation method and system, body fat scale and mobile terminal |
| CN111544853A (en) * | 2020-05-13 | 2020-08-18 | 广东高驰运动科技有限公司 | Physical index evaluation method and equipment in running exercise |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114469663A (en) * | 2022-03-24 | 2022-05-13 | 河南护眼优品实业有限公司 | Intelligent training method and device for preventing and controlling eyesight improvement |
| CN114469663B (en) * | 2022-03-24 | 2023-09-05 | 河南护眼优品实业有限公司 | Intelligent training method and device for preventing and controlling vision enhancement |
| WO2023201285A3 (en) * | 2022-04-14 | 2023-11-23 | Juvyou (Europe) Limited | Computer-implemented systems and methods for health data analysis and management |
| CN116726470A (en) * | 2022-10-14 | 2023-09-12 | 荣耀终端有限公司 | Physical ability age estimation method and electronic equipment |
| WO2024078222A1 (en) * | 2022-10-14 | 2024-04-18 | 荣耀终端有限公司 | Fitness age estimation method and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112582067A (en) | Age estimation model training and age estimation method and device based on big data | |
| CN109087706B (en) | Human health assessment method and system based on sleep big data | |
| US11915825B2 (en) | Systems and methods of analyte measurement analysis | |
| CN104602604B (en) | Biont information processing system, wearable device, server system and the control method of Biont information processing system | |
| CN111084618A (en) | Wearable multifunctional respiration cycle detection system and method | |
| KR20090024808A (en) | Assessing dementia and dementia-type disorders | |
| CN108601566B (en) | Mental stress evaluation method and device | |
| CN103027696A (en) | Human motion energy consumption instrument | |
| CN113133752B (en) | Psychological assessment method, system, device and medium based on heart rate variability analysis | |
| CN111508607B (en) | Obesity prediction system based on BP neural network | |
| CN115299887B (en) | Detection and quantification method and system for dynamic metabolic function | |
| CN109124655A (en) | State of mind analysis method, device, equipment, computer media and multifunctional chair | |
| CN112967780A (en) | Physical ability age prediction method, device, equipment and storage medium based on PAI | |
| CN1275179C (en) | Computer evaluating method for human body sub-health status | |
| CN110353692B (en) | System and method for evaluating physical ability, fatigue and recovery ability based on biological signals | |
| Altini et al. | Personalizing energy expenditure estimation using a cardiorespiratory fitness predicate | |
| CN113040738B (en) | Blood pressure detecting device | |
| CN112472052A (en) | Weight prediction method, device and equipment based on personal motor function index (PAI) | |
| CN112967802A (en) | Linear fusion model training and age estimation method and device for estimating physiological age | |
| Roh et al. | A study on the quality of life improvement in fixed IoT environments: Utilizing active aging biomarkers and big data | |
| CN111276249B (en) | Human behavior activity degree measuring method and system based on human behavior information entropy | |
| CN116019429A (en) | Health monitoring method, device, equipment and storage medium based on physiological index | |
| CN110192872B (en) | Pressure evaluation calibration method and device and storage medium | |
| EP4360559A1 (en) | A training method of a computer-implemented machine learning model for obtaining a resting energy expenditure (ree) parameter of a subject | |
| CN113194811A (en) | Method, device and system for evaluating recovery state of hospital patient and storage medium |
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 | ||
| CB02 | Change of applicant information |
Address after: 230051 No. 01, 5th floor, building B2, Zhongan chuanggu Science Park, No. 900, Wangjiang West Road, high tech Zone, Hefei City, Anhui Province Applicant after: Anhui huami Health Technology Co.,Ltd. Address before: Room 189, building H2, phase II, innovation industrial park, 2800 innovation Avenue, high tech Zone, Hefei City, Anhui Province Applicant before: ANHUI HUAMI INTELLIGENT TECHNOLOGY CO.,LTD. |
|
| CB02 | Change of applicant information | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210330 |
|
| WW01 | Invention patent application withdrawn after publication |