CN106580282A - Human body health monitoring device, system and method - Google Patents
Human body health monitoring device, system and method Download PDFInfo
- Publication number
- CN106580282A CN106580282A CN201610934679.0A CN201610934679A CN106580282A CN 106580282 A CN106580282 A CN 106580282A CN 201610934679 A CN201610934679 A CN 201610934679A CN 106580282 A CN106580282 A CN 106580282A
- Authority
- CN
- China
- Prior art keywords
- wearer
- health
- wearable device
- data
- health monitoring
- 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.)
- Pending
Links
Classifications
-
- 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
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- 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
- A61B5/7235—Details of waveform analysis
-
- 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
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Physiology (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Psychiatry (AREA)
- Artificial Intelligence (AREA)
- Cardiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Fuzzy Systems (AREA)
- Evolutionary Computation (AREA)
- Pulmonology (AREA)
- Mathematical Physics (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a human body health monitoring device, system and method. The device comprises a data acquiring unit, which is used for acquiring parameter information involved in wearer health, safety and vital signs and collected by a wearable device through a wireless transmission module; a health analysis unit, which is used for filtering the data source for health analysis according to the information of the wearer of the wearable device, adjusting weight of every parameter according to each parameter information obtained in the data acquiring unit, and selecting an optimal weighted value, and analyzing outlier in the obtained data by using a clustering algorithm in a distributed mining algorithm and using the data as an abnormal value, and pushing the abnormal value index to a corresponding preset mobile phone; an emotion analyzing unit, which uses a searched sample as a training set, generating a classifying model by a machine learning method, screening the optimal classification algorithm, and performing the emotional classification on every wearer individual and identifying the emotional state of every wearer. The invention can improve the accuracy of health evaluation and identify the emotional state of the wearer.
Description
Technical field
The present invention relates to intelligent terminal technical field, more particularly to a kind of body health monitoring devices, system and method.
Background technology
With the fast development of information technology, increasing electronic equipment is applied to the every field of people's life,
Wearable device has also progressively come into the actual life of people from original conceptual design.At present, wearable device has become
The one big focus in market, and species is increasing, occurs in that the such as various Wearable devices of intelligent watch, Intelligent bracelet.
Not only more conventional electronic product is convenient carries with for wearable device, and the practical work(for becoming increasingly abundant also is provided for people
Can, its appearance greatly changes modern lifestyle, motion mode and leisure way.
At present, with the fast development of Internet of Things, MEMS, smart mobile phone and Intelligent worn device, on the market by
Gradually there are many intelligence wearing products to can be used for monitoring human health status, for example, some Intelligent worn devices can be monitored
The indexs such as body temperature, blood pressure, sleep state, the walking step number of people, the Chinese patent Shen of such as Application No. 201410348992.7
A kind of intelligent sign monitoring wrist wearable device and blood pressure measuring method please be provide, it can be with the blood of real-time monitoring wearer
The data such as pressure, body temperature.
However, above-mentioned prior art but has the disadvantage that：The detection method accuracy rate of safety and Health is relatively low, wrong report, leakage
Report rate is higher；The data such as body temperature, blood pressure, walking step number are simply simply obtained, the big data for not carrying out system is analyzed and pre-
Survey, it is impossible to effectively judgement is made to health；Analysis in terms of for human emotion without reference to.
The content of the invention
To overcome the shortcomings of that above-mentioned prior art is present, one of present invention purpose is to provide a kind of measuring of human health dress
Put, system and method, it carries out data by the personal information to wearer and by the health indicator acquired in sensor
Mining algorithm is calculated, and can improve the degree of accuracy of health evaluating and the affective state of identification wearer.
It is that, up to above-mentioned purpose, the present invention proposes a kind of body health monitoring devices, including：
Data capture unit, using what wireless transport module acquisition wearable device was collected wearer's health, peace are related to
Each parameter information of complete and vital signs；
Health analysis unit, according to the information filtering of wearable device wearer the data source for health analysis, pin are gone out
The each parameter information obtained according to the data capture unit to the data source, constantly the weight of adjustment parameters, selects optimum
Weighted value, and using the clustering algorithm in distributed libray algorithm, the outlier in data acquired is analyzed as exceptional value, will
Exceptional value index pushes to corresponding preset mobile phone；
Sentiment analysis unit, by the use of collect come sample as training set, classification mould is produced by the way of machine learning
Type, filters out optimal classification algorithm, and to each wearer's individuality emotional semantic classification is carried out, and identifies the emotion shape of each wearer
State.
Further, the parameter information at least adds including the body temperature of wearer, blood pressure, palmic rate, sound decibel, movement
Speed.
Further, the health analysis unit is using analytic hierarchy process (AHP) or expert graded or Fuzzy Analysis Method or maximum entropy
Technology law or method of characteristic or PCA or Grey Incidence or method of characteristic carry out weight analysis to each parameter, obtain
The corresponding weight of each index.
Further, the health analysis unit is adopted based on the systemic clustering of Euclidean distance, in analyzing data
Outlier is used as exceptional value.
Further, the sentiment analysis unit is classified using Decision tree classification.
Further, the body health monitoring devices also include an alarm unit, by determining for getting from wearable device
Position data are compared with the position of preset mobile phone, when both exceed safe distance, are sent prompting and are reported to the police to preset mobile phone
Information.
To reach above-mentioned purpose, the present invention also provides a kind of human body health monitoring system, including：
Wearable device, for being related to wearer's health, life and security feature by each sensor collection wearer
Parameter, and by wireless transport module timing be sent to body health monitoring devices；
Body health monitoring devices, obtain the wearer that is related to that the wearable device collects and are good for using wireless transport module
The data of health, life and security feature, according to the information filtering of wearable device wearer the data source for health analysis is gone out,
According to the parameter for obtaining, constantly the weight of adjustment parameters, selects optimal weights value, and using in distributed libray algorithm
Clustering algorithm, analyzes the outlier in data acquired as exceptional value, and exceptional value index is pushed to into corresponding preset mobile phone,
And the body health monitoring devices also by the use of the sample collected as training set, produce classification mould by the way of machine learning
Type, filters out optimal classification algorithm, and to each wearer's individuality emotional semantic classification is carried out, and analyzes the emotion shape of each wearer
State.
Further, the wearable device is at least provided with temperature sensor, gravity sensor, humidity sensor, heart rate
Sensor, sound decibel sensor, blood pressure sensor.
Further, the wearable device is additionally provided with GPS location sensor, for obtain wearer positional information simultaneously
It is sent to the body health monitoring devices.
To reach above-mentioned purpose, the present invention also provides a kind of measuring of human health method, comprises the steps：
Step one, using what wireless transport module acquisition wearable device was collected wearer's health, life and peace are related to
The parameter information of full feature；
Step 2, goes out the data source for health analysis, according to acquisition according to the information filtering of wearable device wearer
Be related to wearer's health, the parameter information of life and security feature, constantly the weight of adjustment parameters, selects optimal weights
Value, and using the clustering algorithm in distributed libray algorithm, the outlier in data acquired is analyzed as exceptional value, will be abnormal
Value index pushes to corresponding preset mobile phone；
Step 3, by the use of the sample collected as training set, produces disaggregated model by the way of machine learning, filters out
Optimal classification algorithm, to each wearer's individuality emotional semantic classification is carried out, and identifies the affective state of each wearer.
Compared with prior art, a kind of body health monitoring devices of the invention, system and method are by the individual of wearer
People's information and data mining algorithm calculating is carried out by the health indicator acquired in sensor, the standard of health evaluating can be improved
Exactness and the affective state of identification wearer.
Description of the drawings
Fig. 1 is a kind of configuration diagram of body health monitoring devices of the invention；
Fig. 2 is the program schematic diagram of SSPS (PCA) in present pre-ferred embodiments；
Fig. 3 is the importance obtained using decision Tree algorithms in spass moduler softwares in the specific embodiment of the invention
Predict the outcome schematic diagram；
Fig. 4 is to analogue data in the specific embodiment of the invention using decision Tree algorithms bag in spass moduler softwares
It is analyzed the result figure for obtaining；
Fig. 5 is a kind of system architecture diagram of human body health monitoring system of the invention；
Fig. 6 is the detail structure chart of wearable device in present pre-ferred embodiments；
The step of Fig. 7 is a kind of measuring of human health method of the invention flow chart.
Specific embodiment
Below by way of specific instantiation and embodiments of the present invention are described with reference to the drawings, those skilled in the art can
The further advantage and effect of the present invention are understood easily by content disclosed in the present specification.The present invention also can be different by other
Instantiation implemented or applied, the every details in this specification also can based on different viewpoints with application, without departing substantially from
Various modifications and change are carried out under the spirit of the present invention.
Fig. 1 is a kind of configuration diagram of body health monitoring devices of the invention.As shown in figure 1, a kind of human body of the invention
Health monitoring device, is applied to big data analysis platform, including：Data capture unit 101, health analysis unit 102, emotion point
Analysis unit 103.
Data capture unit 101, using wireless transport module body temperature, the blood of the wearer that wearable device is collected are obtained
The data such as pressure, palmic rate, sound decibel, translational acceleration；Health analysis unit 102, according to wearable device wearer's
Information filtering goes out the data source for health analysis, according to the continuous acquisition of data capture unit 101 including body temperature, blood pressure, the heart
Frequency hopping rate, sound decibel, the isoparametric data message of translational acceleration, constantly the weight of adjustment parameters, selects more afterwards
Optimal weights value, and using the clustering algorithm in distributed libray algorithm, the outlier in data is analyzed as exceptional value, will
Exceptional value index pushes to preset mobile phone, in the present invention, as a example by monitoring the health of child wearer, measuring of human health dress
Put using distributed structure/architecture and stream process technology to receive and process the data from each child's Intelligent worn device transmission, it is first
Guardian is first needed to upload the essential information of child wearer, including sex, date of birth, birthplace etc., for health analysis,
Then filter out first in identical sex, areal life and the child of close date of birth as data source, then in conjunction with every
The health and fitness information of its all kinds of child for constantly gathering includes the ginseng such as body temperature, blood pressure, palmic rate, sound decibel, translational acceleration
Number, constantly the weight of adjustment parameters, selects more afterwards optimal weights value, and is calculated using the cluster in distributed libray algorithm
Method, analyzes the outlier in data as exceptional value, and exceptional value index is pushed to into corresponding guardian's smart mobile phone；Emotion point
Analysis unit 103, by observing judgement and collecting the affective state of child, including happiness, anger, grief and joy, anxiety, fear etc., while obtaining
Other essential informations under the affective state, and by the use of collect come sample as training set, produced by the way of machine learning
Disaggregated model, sorting algorithm model here includes Bayes's classification, binary tree sort, SVMs, KNN algorithms etc., sieve
Optimal classification algorithm is selected, each wearer (such as child) individuality is classified, that is, analyze the emotion of each wearer
State.
In the present invention, the main each parameter by obtaining of health analysis unit 102 carries out being recycled after weight analysis divides
Clustering method in cloth mining algorithm, analyzes the outlier in data as exceptional value.Specifically, weight is a phase
To concept, for a certain index.The weight of a certain index refers to relative importance of the index in the overall evaluation.
In present pre-ferred embodiments, the sensor of wearable device can get child's body temperature, blood pressure, palmic rate, sound decibel,
The parameters such as translational acceleration, parameter amount is big, and weight analysis can help judge influence degree of each index to total system, while
Also have great significance for sample is more accurately clustered.It is true using accurate, rational method based on the importance of weight
Determine the difficult problem that weight is that the present invention is explored.The common method of weight analysis has levels analytic approach, expert graded, fuzzy analysis
Method, maximum-entropy technique method, method of characteristic, PCA, Grey Incidence, method of characteristic etc., be with method of characteristic below
Example, using SPSS (PCA) Eigenvalues analysis are carried out, and obtain the corresponding weight of each index.
(1) assume to sample per hour a secondary data, sampling total time is 5 hours, initial data and the standardized knots of SPSS
Fruit is shown in Table one and table two.
The initial data of table one
The sample standardization result of table two
| Sample | Body temperature | Blood pressure | Heart rate | Gravity |
| 1 | -.53785 | .74514 | -1.13937 | -1.15087 |
| 2 | -.35449 | 1.03173 | -.89168 | -.46035 |
| 3 | -.41561 | .31525 | .09908 | .80561 |
| 4 | 1.78467 | -1.26101 | .84214 | -.46035 |
| 5 | -.47673 | -.83112 | 1.08983 | 1.26596 |
(2) it is using the result of SPSS principal component analysis
The population variance of explanation
Extracting method：Principal component analysis.
The SPSS principal component analysis results of table three
Then by formula：
Can be following table in the hope of the weighted value of each variable：
| Name variable | Weight |
| Body temperature | 0.6515 |
| Blood pressure | 0.3255 |
| Heart rate | 0.02275 |
| Gravity | 0.00025 |
The weight of each variable of table four
The clustering method that health analysis unit 102 is adopted is illustrated in further detail below：
Cluster analysis is on the premise of classification situation is unknown, data structure to be classified.According to many of a collection of sample
Individual observation index, finding out some can measure the statistic of similarity degree between sample or variable, with these statistics as classification
Foundation, find out the larger sample of some similarity degrees (or index) and be polymerized to a class, other similarity degree is larger
Sample (or index) is polymerized to a class, finishes until all samples (or index) are all polymerized, and forming one has little to big point
Class.According to the difference of clustering object, cluster can be divided into Q types and R types, wherein Q types are the cluster between sample, and R types are change
Cluster between amount.Hierarchical Clustering, K mean cluster etc. can be divided into according to the difference of clustering method.
In the present invention, the task of outlier detection is to find the object dramatically different with most of other objects.Based on poly-
The outlier detection method of class has two kinds：One is to abandon the tuftlet away from other clusters, and two is based on the cluster of prototype.
The present invention is using the systemic clustering based on Euclidean distance.The basic think of of hierarchical clustering (namely Hierarchical Clustering)
Think first to constitute a class by itself each sample, then according to some way measures the relatives' degree between all samples, most like
Sample be first polymerized to a group, measure the relatives' degree between remaining sample and group once again, and will current immediate sample in
Group is polymerized to a class, so repeatedly, till all samples are polymerized to a class.
Hypothesis has n sample, and each sample measures P index (variable), then can obtain firsthand information battle array and be：
Wherein xij(i=1,2 ..., n；J=1,2 ..., p) be i-th sample j-th index observation data.
Euclidean distance formula is：
Add weight after Euclidean distance formula be：
According to table two and table four, can obtain Euclidean distance square matrices is：
As can be seen that X from matrix1,X2Between distance it is most short, the two can be combined into a class, be designated as X6If,
Then by X6,X3,X4,X5The Euclidean distance square matrices of composition are：
Find out from matrix, X6,X3Distance is most short, therefore the two can be classified as a class, be designated as X7,
If
Then by X7,X4,X5The Euclidean distance square matrices of composition are：
Then X7,X5A class can be classified as, therefore final classification results are two classes, are respectively { X1,X2,X3,X5, { X4}。
According to the detection method one of discrete point, it is known that X4For discrete point.
In the present invention, sentiment analysis unit 103, by the use of collect come sample as training set, using machine learning
Mode produces sorting algorithm model, filters out optimal classification algorithm, and each child's individuality is classified.
Specifically, classificating requirement must clear and definite each classification in advance information, according to the feature or attribute of text, divide
To in existing classification.Conventional algorithm has a Decision tree classification, simple Bayesian Classification Arithmetic, based on SVMs
(SVM) grader, neural network, fuzzy classifier method etc..
In present pre-ferred embodiments, using Decision tree classification.Decision tree classification is with tree-shaped result presentation class
Result.Each decision point realizes a test function with discrete output, is designated as branch.It is an object of the present invention to realizing
The emotional semantic classification of different childs, it is assumed that known training set is as follows：
The sentiment analysis training set of table five
Then the purpose of present invention research is, it is known that body temperature, blood pressure, during Heart Rate States, how to judge child's emotion now.
Modeling such as Fig. 2 first in SPSS (PCA) model：
Data are randomly divided into into training set and test set in subregion, as shown in Table 6.
The data partition table of table six
| Sample | Emotional category | Body temperature | Blood pressure | Heart rate | Subregion | |
| 1 | 1 | It is glad | 36.50 | 119.00 | 88.00 | 1_ is trained |
| 2 | 2 | It is sad | 36.40 | 110.00 | 73.00 | 1_ is trained |
| 3 | 3 | It is glad | 36.10 | 118.00 | 81.00 | 2_ is tested |
| 4 | 4 | It is sad | 36.30 | 111.00 | 70.00 | 2_ is tested |
| 5 | 5 | It is glad | 36.60 | 117.00 | 83.00 | 1_ is trained |
| 6 | 6 | It is sad | 36.50 | 111.00 | 81.00 | 1_ is trained |
| 7 | 7 | It is glad | 36.50 | 118.00 | 82.00 | 2_ is tested |
| 8 | 8 | It is sad | 36.20 | 110.00 | 77.00 | 1_ is trained |
Fig. 3 be the specific embodiment of the invention in using the decision Tree algorithms in spass moduler softwares obtain it is important
Property predicts the outcome.In general, decision Tree algorithms predictive variable importance primary concern is that independent variable inspection related to dependent variable
The probable value (p) tested, p is less, and importance is bigger.Fig. 3 its actually horizontal histogram, scale represents variable from 0 to 1
The proportion of importance, value is bigger, and variable is more important, and Fig. 3 lower sections can regard a graduated scale as, and the left side is least important to represent straight
Scale 0 in square figure, the most important scale 1 represented in histogram in the right, different variables can be represented in same scale, more
Plus easily find out the size of variable importance, from figure 3, it can be seen that blood pressure variable affects maximum to emotional semantic classification, body temperature,
Heart rate affects less to sentiment analysis.
Fig. 4 is to analogue data in the specific embodiment of the invention using decision Tree algorithms bag in spass moduler softwares
It is analyzed the result figure for obtaining.Generally, the step of decision Tree algorithms are as follows：Read fileinfo, statistics file number；Utilize
Property value and rule set up decision tree；Post pruning, exports decision tree.According to above-mentioned steps in the specific embodiment of the invention
Analogue data is analyzed, and emotional category decision tree as shown in Figure 4 is obtained.
From the analysis result of table seven, it can be seen that be analyzed according to the method, it is 100% to test correct probability.Cause
This can train set analysis with the analysis method of decision tree to the emotion of table five.
The sentiment analysis accuracy of table seven
| ' subregion ' | 1_ is trained | 2_ is tested | ||
| Correctly | 5 | 100% | 3 | 100% |
| Mistake | 0 | 0% | 0 | 0% |
| Amount to | 5 | 3 |
Certainly, for different sorting techniques, the accuracy and reasonability of classification are not quite similar, the present invention not as
Limit.
It is preferred that data capture unit 101 can be with from wearable device acquisition location data, the health of the present invention
Monitoring device also includes an alarm unit, and the location data for getting is compared with the position of preset mobile phone, when both surpass
When crossing safe distance, to preset mobile phone prompting and warning message are sent.
Fig. 5 is a kind of system architecture diagram of human body health monitoring system of the invention.As shown in figure 5, a kind of human body of the invention
Health monitoring systems, including：Wearable device 50 and body health monitoring devices 51.
Wherein, wearable device 50 is used to be related to wearer's health, life and peace by each sensor collection wearer
The parameter of full feature, and body health monitoring devices 51 are sent to by wireless transport module timing.Fig. 6 is preferably real for the present invention
Apply the detail structure chart of wearable device in example.Specifically, wearable device 50 includes CPU, is wirelessly transferred mould
The sensor of block and various parameters for being related to wearer's health, life and security feature for collection, for example, wearable device
50 are provided with temperature sensor, gravity sensor, humidity sensor, heart rate sensor, sound decibel sensor, blood pressure sensing
Device, wherein, temperature sensor is used for recording wearer's body temperature and local environment temperature, when body temperature and local environment temperature anomaly
When, can in time send alarm；Whether the acceleration of gravity that gravity sensor is used for recording suffered by wearer is normal, prevents from wearing
Wearer falls down, falls；Heart rate sensor is used for recording wearer's changes in heart rate situation, as health and sentiment analysis parameter, supplies
Big data analysis platform carries out mining analysis；Sound decibel sensor, for recording wearer's sob decibel information；Humidity sensor
Device, for recording humidity and ambient humidity with wearer；Blood pressure sensor, it is wearable for recording blood pressure data
Equipment 50 these data can uploaded to the people that is applied to big data cloud analysis platform per three minutes using wireless transport module
Body health monitoring device 51.It is preferred that wearable device 50 is additionally provided with GPS location sensor, for obtaining the position of wearer
Confidence ceases and is sent to body health monitoring devices 51, so that body health monitoring devices 51 believe it with the position of preset mobile phone
Breath is compared, and when both distances exceed safe distance, is produced prompting and warning message to preset mobile phone.
Body health monitoring devices 51, using wireless transport module the body of the wearer that wearable device is collected is obtained
Temperature, blood pressure, palmic rate, sound decibel, translational acceleration etc. are related to the data of wearer's health, life and security feature, root
Go out the data source for health analysis according to the information filtering of wearable device wearer, according to continuous acquisition including body temperature, blood
The parameters such as pressure, palmic rate, sound decibel, translational acceleration, constantly the weight of adjustment parameters, selects more afterwards optimum power
Weight values, and using the clustering algorithm in distributed libray algorithm, the outlier in data is analyzed as exceptional value, by exceptional value
Index pushes to preset mobile phone, meanwhile, body health monitoring devices 51 judge and collect the affective state of child, bag by observation
Happiness, anger, grief and joy, anxiety, fear etc. are included, while obtain other essential informations under the affective state, and using collecting the sample work that comes
For training set, disaggregated model is produced by the way of machine learning, sorting algorithm model here includes Bayes's classification, y-bend
Tree classification, SVMs, KNN algorithms etc., filter out optimal classification algorithm, and each child's individuality is classified, that is, analyze
Go out the affective state of each child.
The step of Fig. 7 is a kind of measuring of human health method of the invention flow chart.As shown in fig. 7, a kind of human body of the invention
Health monitor method, comprises the steps：
Step 701, using wireless transport module body temperature, blood pressure, the heartbeat of the wearer that wearable device is collected are obtained
Frequency, sound decibel, translational acceleration etc. are related to the data of wearer's health, life and security feature；
Step 702, goes out the data source for health analysis, according to continuous according to the information filtering of wearable device wearer
Obtain including body temperature, blood pressure, palmic rate, sound decibel, the isoparametric data message of translational acceleration, constantly adjust each
The weight of parameter, selects more afterwards optimal weights value, and using the clustering algorithm in distributed libray algorithm, in analyzing data
Outlier as exceptional value, exceptional value index is pushed to into preset mobile phone, in the present invention, to monitor the strong of child wearer
As a example by health, body health monitoring devices are received using distributed structure/architecture and stream process technology and processed from each child's intelligence
Wearable device transmission data, it is necessary first to guardian upload child wearer essential information, including sex, the date of birth,
Birthplace etc., for health analysis, then filters out first in identical sex, areal life and the child of close date of birth
As data source, then in conjunction with daily constantly the health and fitness information of all kinds of childs of collection includes body temperature, blood pressure, palmic rate, sound
The parameters such as cent shellfish, translational acceleration, the continuous weight of adjustment parameters selects more afterwards optimal weights value, and using point
Clustering algorithm in cloth mining algorithm, analyzes the outlier in data as exceptional value, and exceptional value index is pushed to into phase
Answer guardian's smart mobile phone.
Step 703, by the use of collect come sample as training set, disaggregated model is produced by the way of machine learning, this
In sorting algorithm model include Bayes's classification, binary tree sort, SVMs, KNN algorithms etc., filter out optimal classification
Algorithm, classifies to each wearer's individuality.
It is preferred that the measuring of human health method of the present invention also comprises the steps：
The positional information of wearer is obtained, it is compared with the positional information of preset mobile phone, when both distances exceed
During safe distance, prompting and warning message are produced to preset mobile phone.
In sum, a kind of body health monitoring devices of the invention, system and method are by the personal information to wearer
And data mining algorithm calculating is carried out by the health indicator acquired in sensor, can improve health evaluating the degree of accuracy and
The affective state of identification wearer.
Any those skilled in the art can repair under the spirit and the scope without prejudice to the present invention to above-described embodiment
Decorations and change.Therefore, the scope of the present invention, should be as listed by claims.
Claims (10)
1. a kind of body health monitoring devices, including：
Data capture unit, using wireless transport module obtain wearable device collect be related to wearer health, safety and
Each parameter information of vital signs；
Health analysis unit, goes out the data source for health analysis, for this according to the information filtering of wearable device wearer
Each parameter information that data source is obtained according to the data capture unit, constantly the weight of adjustment parameters, selects optimal weights
Value, and using the clustering algorithm in distributed libray algorithm, the outlier in data acquired is analyzed as exceptional value, will be abnormal
Value index pushes to corresponding preset mobile phone；
Sentiment analysis unit, by the use of the sample collected as training set, produces disaggregated model by the way of machine learning, screens
Go out optimal classification algorithm, emotional semantic classification is carried out to each wearer's individuality, identify the affective state of each wearer.
2. a kind of body health monitoring devices as claimed in claim 1, it is characterised in that：The parameter information at least includes wearing
The body temperature of person, blood pressure, palmic rate, sound decibel, translational acceleration.
3. a kind of body health monitoring devices as claimed in claim 2, it is characterised in that：The health analysis unit adopts level
Analytic approach or expert graded or Fuzzy Analysis Method or maximum-entropy technique method or method of characteristic or PCA or grey pass
Connection method or method of characteristic carry out weight analysis to each parameter, obtain the corresponding weight of each index.
4. a kind of body health monitoring devices as claimed in claim 3, it is characterised in that：The health analysis unit is adopted and is based on
The systemic clustering of Euclidean distance, analyzes the outlier in data acquired.
5. a kind of body health monitoring devices as claimed in claim 4, it is characterised in that：The sentiment analysis unit adopts decision-making
Tree classification method carries out emotional semantic classification.
6. a kind of body health monitoring devices as claimed in claim 2, it is characterised in that：The body health monitoring devices are also wrapped
An alarm unit is included, the location data got from wearable device is compared with the position of preset mobile phone, when both surpass
When crossing safe distance, to preset mobile phone prompting and warning message are sent.
7. a kind of human body health monitoring system, including：
Wearable device, for gathering the ginseng for being related to wearer's health, life and security feature of wearer by each sensor
Number, and body health monitoring devices are sent to by wireless transport module timing；
Body health monitoring devices, using wireless transport module obtain that the wearable device collects be related to wearer's health,
The data of life and security feature, according to the information filtering of wearable device wearer the data source for health analysis, root are gone out
According to the parameter for obtaining, the continuous weight of adjustment parameters selects optimal weights value, and using distributed libray algorithm in it is poly-
Class algorithm, analyzes the outlier in data acquired as exceptional value, and exceptional value index is pushed to into corresponding preset mobile phone, and
And the body health monitoring devices produce classification mould also by the use of the sample collected as training set by the way of machine learning
Type, filters out optimal classification algorithm, and to each wearer's individuality emotional semantic classification is carried out, and analyzes the emotion shape of each wearer
State.
8. a kind of human body health monitoring system as claimed in claim 7, it is characterised in that：The wearable device at least provided with
Temperature sensor, gravity sensor, humidity sensor, heart rate sensor, sound decibel sensor, blood pressure sensor.
9. a kind of human body health monitoring system as claimed in claim 8, it is characterised in that：The wearable device is additionally provided with
GPS location sensor, for obtaining the positional information of wearer and being sent to the body health monitoring devices.
10. a kind of measuring of human health method, comprises the steps：
Step one, using what wireless transport module acquisition wearable device was collected wearer's health, life and Special safety are related to
The parameter information levied；
Step 2, according to the information filtering of wearable device wearer the data source for health analysis is gone out, according to relating to for obtaining
And the parameter information of wearer's health, life and security feature, constantly the weight of adjustment parameters, selects optimal weights value,
And using the clustering algorithm in distributed libray algorithm, the outlier in data acquired is analyzed as exceptional value, by exceptional value
Index pushes to corresponding preset mobile phone；
Step 3, by the use of the sample collected as training set, produces disaggregated model by the way of machine learning, filters out optimum
Sorting algorithm, to each wearer's individuality emotional semantic classification is carried out, and identifies the affective state of each wearer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610934679.0A CN106580282A (en) | 2016-10-25 | 2016-10-25 | Human body health monitoring device, system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610934679.0A CN106580282A (en) | 2016-10-25 | 2016-10-25 | Human body health monitoring device, system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106580282A true CN106580282A (en) | 2017-04-26 |
Family
ID=58556260
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610934679.0A Pending CN106580282A (en) | 2016-10-25 | 2016-10-25 | Human body health monitoring device, system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106580282A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107714056A (en) * | 2017-09-06 | 2018-02-23 | 上海斐讯数据通信技术有限公司 | A kind of wearable device of intellectual analysis mood and the method for intellectual analysis mood |
| CN108492890A (en) * | 2018-05-25 | 2018-09-04 | 广州嘉元华健康电子科技有限公司 | A kind of health condition monitoring system and method |
| CN108711204A (en) * | 2018-05-18 | 2018-10-26 | 长安大学 | A kind of the driving abnormality detection system and method for comprehensive people-Che-road multi-source information |
| WO2018210344A1 (en) * | 2017-05-19 | 2018-11-22 | 匡海云 | System for monitoring health conditions in public place and issuing alert |
| CN109583796A (en) * | 2019-01-08 | 2019-04-05 | 河南省灵山信息科技有限公司 | A kind of data digging system and method for Logistics Park OA operation analysis |
| CN109643579A (en) * | 2017-07-03 | 2019-04-16 | 深圳市汇顶科技股份有限公司 | Method for evaluating quality and device, method for establishing model and module, wearable device |
| CN109636694A (en) * | 2018-12-21 | 2019-04-16 | 广州智而康网络科技有限公司 | A kind of method and system of child's health data management |
| CN110058692A (en) * | 2019-04-22 | 2019-07-26 | 武汉久乐科技有限公司 | A kind of wearing data processing method and device based on machine learning |
| CN110432864A (en) * | 2019-07-19 | 2019-11-12 | 成都中医药大学 | A kind of tcm constitution dynamic change discrimination method and system |
| CN110674107A (en) * | 2019-08-30 | 2020-01-10 | 广西农业职业技术学院 | Data screening processing system |
| CN110857891A (en) * | 2018-08-22 | 2020-03-03 | 上海宜链物联网有限公司 | Intelligent body temperature monitoring device and method |
| CN110881987A (en) * | 2019-08-26 | 2020-03-17 | 首都医科大学 | Old person emotion monitoring system based on wearable equipment |
| CN111671405A (en) * | 2020-05-29 | 2020-09-18 | 昭苏县西域马业有限责任公司 | Saddle with health detection device, health detection system and method |
| CN112399819A (en) * | 2018-07-02 | 2021-02-23 | 3M创新有限公司 | Sensing system and method for monitoring a time-dependent process |
| CN113626607A (en) * | 2021-09-17 | 2021-11-09 | 平安银行股份有限公司 | Abnormal work order identification method and device, electronic equipment and readable storage medium |
| CN113761722A (en) * | 2021-08-18 | 2021-12-07 | 航天科工海鹰集团有限公司 | Spacecraft multi-working-condition service life prediction method based on PCA |
| CN116269242A (en) * | 2023-05-17 | 2023-06-23 | 广州培生智能科技有限公司 | Old person health status real-time monitoring system based on internet |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090001146A (en) * | 2007-06-29 | 2009-01-08 | 장형종 | System and method for detecting abnormality of biosignal using neural network with weighted fuzzy membership funtions |
| CN101853428A (en) * | 2009-04-03 | 2010-10-06 | 范晓 | Remote characteristic query system and method |
| CN102406507A (en) * | 2010-09-20 | 2012-04-11 | 天津职业技术师范大学 | Method for monitoring fatigue degree of driver based on human body physiological signal |
| CN204500673U (en) * | 2014-12-23 | 2015-07-29 | 华南理工大学 | Based on pregnant and lying-in women's health monitoring system of mobile Internet |
| CN204680066U (en) * | 2015-04-23 | 2015-09-30 | 中国人民解放军第一八一医院 | A kind of patient information wrist strap |
| WO2015127361A8 (en) * | 2014-02-23 | 2015-11-12 | Interdigital Patent Holdings, Inc. | Cognitive and affective human machine interface |
| CN105212899A (en) * | 2015-09-21 | 2016-01-06 | 李永川 | Health sleep type remote monitoring service system |
| CN105260745A (en) * | 2015-09-30 | 2016-01-20 | 西安沧海网络科技有限公司 | Information push service system capable of carrying out emotion recognition and prediction based on big data |
| CN105740621A (en) * | 2016-01-29 | 2016-07-06 | 江阴中科今朝科技有限公司 | Moving monitoring and intelligent aged nursing health cloud platform of human body behavior data |
| CN105982678A (en) * | 2015-02-12 | 2016-10-05 | 上海宽带技术及应用工程研究中心 | Method for judging mood according to heart rate and breath |
-
2016
- 2016-10-25 CN CN201610934679.0A patent/CN106580282A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090001146A (en) * | 2007-06-29 | 2009-01-08 | 장형종 | System and method for detecting abnormality of biosignal using neural network with weighted fuzzy membership funtions |
| CN101853428A (en) * | 2009-04-03 | 2010-10-06 | 范晓 | Remote characteristic query system and method |
| CN102406507A (en) * | 2010-09-20 | 2012-04-11 | 天津职业技术师范大学 | Method for monitoring fatigue degree of driver based on human body physiological signal |
| WO2015127361A8 (en) * | 2014-02-23 | 2015-11-12 | Interdigital Patent Holdings, Inc. | Cognitive and affective human machine interface |
| CN204500673U (en) * | 2014-12-23 | 2015-07-29 | 华南理工大学 | Based on pregnant and lying-in women's health monitoring system of mobile Internet |
| CN105982678A (en) * | 2015-02-12 | 2016-10-05 | 上海宽带技术及应用工程研究中心 | Method for judging mood according to heart rate and breath |
| CN204680066U (en) * | 2015-04-23 | 2015-09-30 | 中国人民解放军第一八一医院 | A kind of patient information wrist strap |
| CN105212899A (en) * | 2015-09-21 | 2016-01-06 | 李永川 | Health sleep type remote monitoring service system |
| CN105260745A (en) * | 2015-09-30 | 2016-01-20 | 西安沧海网络科技有限公司 | Information push service system capable of carrying out emotion recognition and prediction based on big data |
| CN105740621A (en) * | 2016-01-29 | 2016-07-06 | 江阴中科今朝科技有限公司 | Moving monitoring and intelligent aged nursing health cloud platform of human body behavior data |
Non-Patent Citations (1)
| Title |
|---|
| 陶建格: "《定量管理模型与方法》", 21 December 2013 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018210344A1 (en) * | 2017-05-19 | 2018-11-22 | 匡海云 | System for monitoring health conditions in public place and issuing alert |
| CN109643579A (en) * | 2017-07-03 | 2019-04-16 | 深圳市汇顶科技股份有限公司 | Method for evaluating quality and device, method for establishing model and module, wearable device |
| CN109643579B (en) * | 2017-07-03 | 2023-08-11 | 深圳市汇顶科技股份有限公司 | Quality evaluation method and device, model building method and module and wearable device |
| CN107714056A (en) * | 2017-09-06 | 2018-02-23 | 上海斐讯数据通信技术有限公司 | A kind of wearable device of intellectual analysis mood and the method for intellectual analysis mood |
| CN108711204A (en) * | 2018-05-18 | 2018-10-26 | 长安大学 | A kind of the driving abnormality detection system and method for comprehensive people-Che-road multi-source information |
| CN108711204B (en) * | 2018-05-18 | 2020-07-31 | 长安大学 | Driving abnormity detection system and method integrating human-vehicle-road multi-source information |
| CN108492890A (en) * | 2018-05-25 | 2018-09-04 | 广州嘉元华健康电子科技有限公司 | A kind of health condition monitoring system and method |
| CN108492890B (en) * | 2018-05-25 | 2024-03-08 | 广州嘉元华健康电子科技有限公司 | Human health state monitoring system and method |
| CN112399819A (en) * | 2018-07-02 | 2021-02-23 | 3M创新有限公司 | Sensing system and method for monitoring a time-dependent process |
| CN110857891A (en) * | 2018-08-22 | 2020-03-03 | 上海宜链物联网有限公司 | Intelligent body temperature monitoring device and method |
| CN109636694A (en) * | 2018-12-21 | 2019-04-16 | 广州智而康网络科技有限公司 | A kind of method and system of child's health data management |
| CN109583796A (en) * | 2019-01-08 | 2019-04-05 | 河南省灵山信息科技有限公司 | A kind of data digging system and method for Logistics Park OA operation analysis |
| CN110058692A (en) * | 2019-04-22 | 2019-07-26 | 武汉久乐科技有限公司 | A kind of wearing data processing method and device based on machine learning |
| CN110058692B (en) * | 2019-04-22 | 2022-02-22 | 武汉久乐科技有限公司 | Wearable data processing method and device based on machine learning |
| CN110432864A (en) * | 2019-07-19 | 2019-11-12 | 成都中医药大学 | A kind of tcm constitution dynamic change discrimination method and system |
| CN110881987A (en) * | 2019-08-26 | 2020-03-17 | 首都医科大学 | Old person emotion monitoring system based on wearable equipment |
| CN110674107A (en) * | 2019-08-30 | 2020-01-10 | 广西农业职业技术学院 | Data screening processing system |
| CN111671405A (en) * | 2020-05-29 | 2020-09-18 | 昭苏县西域马业有限责任公司 | Saddle with health detection device, health detection system and method |
| CN113761722A (en) * | 2021-08-18 | 2021-12-07 | 航天科工海鹰集团有限公司 | Spacecraft multi-working-condition service life prediction method based on PCA |
| CN113626607B (en) * | 2021-09-17 | 2023-08-25 | 平安银行股份有限公司 | Abnormal work order identification method and device, electronic equipment and readable storage medium |
| CN113626607A (en) * | 2021-09-17 | 2021-11-09 | 平安银行股份有限公司 | Abnormal work order identification method and device, electronic equipment and readable storage medium |
| CN116269242B (en) * | 2023-05-17 | 2023-07-18 | 广州培生智能科技有限公司 | Old person health status real-time monitoring system based on internet |
| CN116269242A (en) * | 2023-05-17 | 2023-06-23 | 广州培生智能科技有限公司 | Old person health status real-time monitoring system based on internet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106580282A (en) | Human body health monitoring device, system and method | |
| US20190365332A1 (en) | Determining wellness using activity data | |
| CN107506602A (en) | A kind of big data health forecast system | |
| WO2021068781A1 (en) | Fatigue state identification method, apparatus and device | |
| WO2016075529A1 (en) | System and method for detecting and quantifying deviations from physiological signals normality | |
| Xu et al. | Intelligent emotion detection method based on deep learning in medical and health data | |
| US20160007165A1 (en) | Method and system for identifying exceptions of people behavior | |
| CN107609477A (en) | It is a kind of that detection method is fallen down with what Intelligent bracelet was combined based on deep learning | |
| CN117598700A (en) | Intelligent blood oxygen saturation detection system and method | |
| CN109770913A (en) | A kind of abnormal gait recognition methods based on reverse transmittance nerve network | |
| CN117643461A (en) | Heart rate intelligent monitoring system and method based on artificial intelligence | |
| Beily et al. | A sensor based on recognition activities using smartphone | |
| CN117198468A (en) | Intervention scheme intelligent management system based on behavior recognition and data analysis | |
| Mallol-Ragolta et al. | harAGE: a novel multimodal smartwatch-based dataset for human activity recognition | |
| Khan et al. | Evaluating machine learning techniques on human activity recognition using accelerometer data | |
| Abdullah et al. | Parkinson’s Disease Symptom Detection using Hybrid Feature Extraction and Classification Model | |
| Silva et al. | Fusion of clinical, self-reported, and multisensor data for predicting falls | |
| CN112155520A (en) | Monitoring device and method based on body temperature data, electronic equipment and storage medium | |
| CN116486819A (en) | Pig respiratory disease monitoring method and system and farm inspection robot | |
| KR101483218B1 (en) | Activity diagnosis apparatus | |
| Ji et al. | Human Motion Pattern Recognition Based on Nano-sensor and Deep Learning | |
| Yashaswini et al. | Stress detection using deep learning and IoT | |
| Saranya et al. | Human body temperature and face mask audit system for COVID protocol | |
| CN111209816B (en) | Non-contact fatigue driving detection method based on regular extreme learning machine | |
| Luna-Perejon et al. | Sampling Frequency Evaluation on Recurrent Neural Networks Architectures for IoT Real-time Fall Detection Devices. |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170426 |