CN106691402A - Fatigue level analysis method and device based on pulse characteristics - Google Patents
Fatigue level analysis method and device based on pulse characteristics Download PDFInfo
- Publication number
- CN106691402A CN106691402A CN201611178555.0A CN201611178555A CN106691402A CN 106691402 A CN106691402 A CN 106691402A CN 201611178555 A CN201611178555 A CN 201611178555A CN 106691402 A CN106691402 A CN 106691402A
- Authority
- CN
- China
- Prior art keywords
- fatigue
- degree
- pulse
- pulse data
- label
- 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
-
- 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
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- 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
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
- G06F2218/14—Classification; Matching by matching peak patterns
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Psychiatry (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Evolutionary Computation (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Social Psychology (AREA)
- Psychology (AREA)
- Hospice & Palliative Care (AREA)
- Educational Technology (AREA)
- Developmental Disabilities (AREA)
- Child & Adolescent Psychology (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Cardiology (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
Abstract
An embodiment of the invention discloses a fatigue level analysis method and device based on pulse characteristics, wherein the method comprises: acquiring a pulse data fatigue level classifier model through an initial first pulse data training classifier, wherein first pulse data includes a first fatigue level label; predicting a second fatigue level label corresponding to newly collected second pulse data according to the pulse data fatigue level classifier so as to obtain the second fatigue level label; determining a fatigue level corresponding to the second fatigue level label according to the second fatigue level label. The fatigue level analysis method and device based on pulse characteristics can be carried out by a user to judge and measure the fatigue level quickly and conveniently; therefore, fatigue level measuring accuracy is improved, and a portable fatigue level measurer is brought for users.
Description
Technical field
Embodiment of the present invention is related to field of medical technology, more particularly to a kind of degree of fatigue based on pulse characteristics point
Analysis method and apparatus.
Background technology
Conventionally, as fatigue detecting has, influence factor is more, the features such as define difficulty, this cause it is tired in real time
Detection and judgement become more difficult.
At present, two kinds of subjective assessment and objective evaluation are broadly divided into Fatigue Assessment method.Wherein, subjective assessment method is logical
Roll form development is bothered about, is influenceed larger by subjective factor；Objective evaluation rule is to job performance and life by instrument and equipment
Reason signal is analyzed.
On the one hand, the degree of fatigue testing equipment of specialty, operation are generally required for due to the detection for fatigue and analysis
Complicated, high cost, particularly there are problems that very big in terms of mobility and portability.On the other hand, based on physiological signal
The characteristics of analysis of fatigue has real-time, noninvasive, unperturbed.Therefore, in the prior art, more using physiological signal, for example mainly
There are pulse, electrocardiosignal, and wherein, electrocardiosignal turns into first-selected due to collection relatively convenient.
But, prior art, the scheme that also degree of fatigue is not detected and judged using pulse signal.
In sum, it is in the prior art also without a kind of portable, wearable fatigue detecting equipment and accurately tired
Analysis method.
The content of the invention
Embodiment of the present invention is mainly solving the technical problems that provide a kind of degree of fatigue analysis based on pulse characteristics
Method and apparatus, can bring a kind of good degree of fatigue analysis method based on pulse characteristics of Consumer's Experience and dress to user
Put.
In order to solve the above technical problems, the technical scheme that embodiment of the present invention is used is：There is provided a kind of based on arteries and veins
The degree of fatigue analysis method of feature of fighting, the method includes：
Grader is trained by the first initial pulse data, pulse data degree of fatigue sorter model is obtained, wherein,
First pulse data includes the first degree of fatigue label；
By pulse data degree of fatigue sorter model to corresponding second degree of fatigue of freshly harvested second pulse data
Label is predicted, and obtains the second degree of fatigue label；
Its corresponding degree of fatigue is determined according to the second degree of fatigue label.
Preferably, grader is trained by the first initial pulse data, obtains pulse data degree of fatigue grader mould
Type includes：
The first pulse data is processed by preset algorithm, obtains interim pulse data；
Extract the feature of interim pulse data；
Feature and its first degree of fatigue label are integrated into training set, and grader is trained by training set, divided
Class device model parameter.
Preferably, the first pulse data is processed by preset algorithm, obtaining interim pulse data includes：
By Fourier transformation, the pulse time-domain signal in the first pulse data is converted into pulse frequency-region signal；
CF and its correspondence amplitude are extracted in pulse power spectrum, CF and its correspondence amplitude are classified
Treatment.
Preferably, the feature for extracting interim pulse data includes：
The direct feature and indirect feature of pulse power spectrum are extracted, wherein, direct feature includes main peak amplitude, peak value frequently
Rate, center of gravity amplitude and gravity frequency, indirect feature include general power, low frequency power and the high frequency power ratio of power spectrum.
Preferably, determine that its corresponding degree of fatigue includes according to the second degree of fatigue label：
Degree of fatigue is divided three classes, respectively mental fatigue, spirit are general and energetic；
For three class degree of fatigues determine its correspondence degree of fatigue label respectively.
The invention allows for a kind of degree of fatigue analytical equipment based on pulse characteristics, the device includes：
Tired label initial module, for training grader by the first initial pulse data, obtains pulse data tired
Labor degree sorter model, wherein, the first pulse data includes the first degree of fatigue label；
Tired Tag Estimation module, for by pulse data degree of fatigue sorter model to freshly harvested second Pulse Rate
It is predicted according to corresponding second degree of fatigue label, obtains the second degree of fatigue label；
Degree of fatigue determining module, for determining its corresponding degree of fatigue according to the second degree of fatigue label.
Preferably, tired label initial module include interim pulse data acquiring unit, pulse characteristics extraction unit and
Classifier training unit, wherein,
Interim pulse data acquiring unit is used to process the first pulse data by preset algorithm, obtains interim pulse
Data；
Pulse characteristics extraction unit is used to extract the feature of interim pulse data；
Classifier training unit is used to for feature and its first degree of fatigue label to be integrated into training set, and by training set
Training grader, obtains sorter model parameter.
Preferably, interim pulse data acquiring unit is additionally operable to：
By Fourier transformation, the pulse time-domain signal in the first pulse data is converted into pulse frequency-region signal；
CF and its correspondence amplitude are extracted in pulse power spectrum, CF and its correspondence amplitude are classified
Treatment.
Preferably, pulse characteristics extraction unit is additionally operable to：
The direct feature and indirect feature of pulse power spectrum are extracted, wherein, direct feature includes main peak amplitude, peak value frequently
Rate, center of gravity amplitude and gravity frequency, indirect feature include general power, low frequency power and the high frequency power ratio of power spectrum.
Preferably, degree of fatigue determining module includes degree of fatigue taxon and tired tag determination unit, wherein,
Degree of fatigue taxon is used to be divided three classes degree of fatigue, and respectively mental fatigue, spirit are general and smart
God is full；
Tired tag determination unit is used to determine its correspondence degree of fatigue label respectively for three class degree of fatigues.
Implement the present invention, grader is trained by the first initial pulse data, obtain the classification of pulse data degree of fatigue
Device model, wherein, the first pulse data includes the first degree of fatigue label；By pulse data degree of fatigue sorter model to new
The corresponding second degree of fatigue label of second pulse data of collection is predicted, and obtains the second degree of fatigue label；According to
Two degree of fatigue labels determine its corresponding degree of fatigue.Allow user conveniently and efficiently to degree of fatigue carry out judge and
Measurement, one is, improve measurement degree of fatigue accuracy, two are, be user bring one kind can portable degree of fatigue
Measurement apparatus.
Brief description of the drawings
Fig. 1 is the degree of fatigue analysis method first embodiment flow based on pulse characteristics provided in an embodiment of the present invention
Figure；
Fig. 2 is the degree of fatigue analysis method second embodiment flow based on pulse characteristics provided in an embodiment of the present invention
Figure；
Fig. 3 is the degree of fatigue analysis method 3rd embodiment flow based on pulse characteristics provided in an embodiment of the present invention
Figure；
Fig. 4 is the degree of fatigue analysis method fourth embodiment flow based on pulse characteristics provided in an embodiment of the present invention
Figure；
Fig. 5 is the embodiment flow of degree of fatigue analysis method the 5th based on pulse characteristics provided in an embodiment of the present invention
Figure；
Fig. 6 is the degree of fatigue analytical equipment sixth embodiment structural frames based on pulse characteristics provided in an embodiment of the present invention
Figure.
Specific embodiment
In order that the purpose of the present invention, technical scheme and advantage become more apparent, below in conjunction with drawings and Examples,
The present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only used to explain the present invention, and
It is not used in the restriction present invention.
As long as additionally, technical characteristic involved in invention described below each implementation method is each other not
Constituting conflict can just be combined with each other.
Embodiment 1：
The embodiment of the present invention 1 provides the first preferred embodiment of the degree of fatigue analysis method based on pulse characteristics, such as
Fig. 1 show the degree of fatigue analysis method first embodiment flow chart based on pulse characteristics provided in an embodiment of the present invention.
Refering to Fig. 1, a kind of degree of fatigue analysis method based on pulse characteristics that the present embodiment is provided, this method include with
Lower step：
A kind of degree of fatigue analysis method based on pulse characteristics, the method includes：
S1, grader is trained by initial the first pulse data, obtain pulse data degree of fatigue sorter model, its
In, the first pulse data includes the first degree of fatigue label；
S2, by pulse data degree of fatigue sorter model the second tired journey corresponding to freshly harvested second pulse data
Scale label are predicted, and obtain the second degree of fatigue label；
S3, its corresponding degree of fatigue is determined according to the second degree of fatigue label.
Hereinafter, by from the definition of pulse, the normal range (NR) of pulse, pulse signal the characteristics of and pulse signal feature,
Relevant explanation is done to the pulse that the present embodiment is previously mentioned：
1st, the definition of pulse
Human pulse come from it is aroused in interest, it be human life exist vital sign.
During blood circulation, when ventricular contraction aorta petal is opened, in blood intake sustainer, by heart in itself
Influence and flow through the shadow of various physiologic factors such as vascular resistence, blood viscosity and vessel wall elasticity etc. in arteries at different levels and branch
Ring, part blood can not be immediately entered in vein, the blood injected temporarily is stayed in proximal aorta, cause aortic pressure liter
It is high and expand；The aortic valve closing when ventricular diastole, penetrates blood stopping, and aorta petal is replied and shunk.This one of sustainer
One contracts causes pressure since area is raised, is propagated to distal aorta and its branch in the form of ripple, in artery and superficial artery
Can visually see or finger touches beating.
Pulse waveform includes an ascending branch and a descending branch.The unexpected expansion of artery, one when ascending branch represents ventricular contraction
As rise rapid and smooth, the size of its speed for rising and wave amplitude is received to penetrate the elastic shadow of blood speed, Artery resistance, arterial wall
Ring, the cardiac output at most ascending branch rate of climb is fast, wave amplitude is big.Descending branch represents ventricular diastole.
The characteristics of volume pulsation wave is one of physiology signal, existing general physiological signal, also there is some of its own
Feature.Volume pulsation wave refers to that under heartbeat, blood pressure flows through arteriole, capillary, venule in peripheral vascular etc.
During capilary, the pulsatile change of the volumetric blood of this part blood vessel.Largely, volume pulsation can reflect human vas system
The flow characteristic of many physiological and pathologicals in system.
2nd, the normal range (NR) of pulse
Pulse is arteriopalmus, and pulse frequency is pulse frequency.The pulse of normal person is consistent with heartbeat.Adult normal is 60
To 100 beats/min, often for per minute 70-80 times, averagely about 72 beats/min.The elderly is slower, is 55 to 60 beats/min.Normal human connection
Rate rule, is not in pulse interval time phenomenon different in size.Normal person's pulse is strong and weak impartial, is not in strong and weak alternating
Phenomenon.
The frequency of pulse is influenceed by age and sex, 110-160 times per minute of fetus, baby 120-140 per minute
Secondary, child is per minute 90-100 times, and disease in school age children is per minute 80-90 times.
In addition, moving and can speed pulse when excited；And rest, sleep pulse of can then make slows down.Adult's pulse frequency
It is per minute more than 100 times, referred to as tachycardia；It is per minute to be less than 60 times, referred to as bradycardia.
3rd, the characteristics of pulse signal
Signal is weak, interference is strong.Pulse signal is derived from heart, signal amplitude very little.The usually quantity of microvolt or millivolt
Level.Therefore, the external disturbance such as human body or instrument is just very big relative to pulse signal.Main have an industrial frequency noise, myoelectricity interference,
The noise that finger tip contact noise and human body own activity cause, increased the difficulty for the treatment of and the analysis of pulse wave signal.
Frequency is low.Pulse wave signal is a kind of low frequency signal, and the frequency of the pulse signal of normal person is in 0.01~40Hz models
In enclosing, generally 1Hz, 99% Energy distribution is between 0~10Hz.
Variability.Pulse signal belongs to close to periodic deterministic signal, but is not completely specified, pulse letter
Number different cycles with the Feature change of body, some small changes also occur, especially with the various lifes of human body
The change of reason pathological factor and ambient environmental conditions, its waveform also can correspondingly change.These variability are just further
Increased the complexity of Pulse signal analysis and treatment.
4th, the feature of pulse signal
By the feature of pulse signal, the corresponding relation that can be set up between pulse and fatigue characterizes pulse signal frequency
Characteristic of field has direct feature and indirect feature：Direct feature includes main peak amplitude, crest frequency, center of gravity amplitude, gravity frequency；Between
Connect low frequency of the feature including power spectrum general power TP and power spectrum and compare LF/HF with high frequency power.
In the present embodiment, first, grader is trained by the first initial pulse data, obtains pulse data fatigue journey
Degree sorter model, wherein, the first pulse data includes the first degree of fatigue label.
In the present embodiment, the grader for being used is SVMs (SVM, Support Vector Mach i ne).
Using Training Support Vector Machines grader, comprise the following steps that：
Skin electrical signal data composing training collection of the collection with known degree of fatigue label, sends into SVM, for obtaining
SVM models.Wherein, each group of data of feeding are one by pulse electric signal degree of fatigue label institute group corresponding with the signal
Into two tuples.
Wherein, degree of fatigue label refers to quantify mark to a kind of of degree of fatigue, for example, degree of fatigue is divided into three
Class, respectively mental fatigue, spirit it is general and energetic, respectively, be above-mentioned three classes degree give respectively P1, P2 and
The mark of P3.
Secondly, by pulse data degree of fatigue sorter model the second fatigue corresponding to freshly harvested second pulse data
Degree label is predicted, and obtains the second degree of fatigue label.Comprise the following steps that：
According to the SVM models for obtaining, using SVM models to the new pulse electrical signal data not comprising degree of fatigue label
Carry out Tag Estimation.Wherein, the new pulse electrical signal data not comprising degree of fatigue label refers to, when getting these pulses
During electrical signal data, it is unknown to participate in the degree of fatigue of object.
Finally, its corresponding degree of fatigue is determined according to the second degree of fatigue label.For example, work as being detected by above-mentioned steps
Corresponding when the degree of fatigue label of measurand is P2, its current degree of fatigue is general for spirit.
It is understood that in the present embodiment, the first pulse data includes：The skin telecommunications of known degree of fatigue label
Number (that is, pulse wave primary signal, and the corresponding first degree of fatigue label of the pulse wave primary signal)；Second
Pulse data only includes：The skin electrical signal data pulse wave primary signal of unknown degree of fatigue label is (that is, not comprising the arteries and veins
The corresponding labor degree label of waveform primary signal of fighting).It is to predict using above-mentioned pulse data degree of fatigue sorter model
The corresponding second degree of fatigue label of second pulse data.
The beneficial effect of the present embodiment is to train grader by the first initial pulse data, obtains pulse data
Degree of fatigue sorter model, wherein, the first pulse data includes the first degree of fatigue label；By pulse data degree of fatigue point
Class device model is predicted to the corresponding second degree of fatigue label of freshly harvested second pulse data, obtains the second degree of fatigue
Label；Its corresponding degree of fatigue is determined according to the second degree of fatigue label.Allow user conveniently and efficiently to tired journey
Degree is judged and is measured.
Embodiment 2：
The embodiment of the present invention 2 provides the second preferred embodiment of the degree of fatigue analysis method based on pulse characteristics, such as
Fig. 2 show the degree of fatigue analysis method second embodiment flow chart based on pulse characteristics provided in an embodiment of the present invention.
Refering to a kind of degree of fatigue analysis method based on pulse characteristics that Fig. 2, the present embodiment are provided, in above-described embodiment
On the basis of 1, grader is trained by the first initial pulse data, obtain pulse data degree of fatigue sorter model bag
Include：
S11, the first pulse data is processed by preset algorithm, obtain interim pulse data；
S12, the feature for extracting interim pulse data；
S13, feature and its first degree of fatigue label are integrated into training set, and grader is trained by training set, obtained
To sorter model parameter.
In the present embodiment, by Fourier transformation, first, the time-domain signal conversion in the pulse data that will be collected
Into pulse frequency-region signal；Then, in pulse data extract pulse power spectrum, and in pulse power spectrum extract target frequency and
Its correspondence amplitude, and these frequencies and amplitude are classified；Finally, it is training set by the feature integration of above-mentioned pulse data,
Above-mentioned SVM is trained by training set, so as to obtain SVM models.
The beneficial effect of the present embodiment is that the first pulse data is processed by by preset algorithm, obtains interim
Pulse data, then extracts the feature of interim pulse data, and feature and its first degree of fatigue label finally are integrated into training
Collection, and grader is trained by training set, obtain sorter model parameter.The training to grader is realized, is follow-up use
Grader is analyzed there is provided accurate foundation to new pulse data.
Embodiment 3：
The embodiment of the present invention 3 provides the third preferred embodiment of the degree of fatigue analysis method based on pulse characteristics, such as
Fig. 3 show the degree of fatigue analysis method 3rd embodiment flow chart based on pulse characteristics provided in an embodiment of the present invention.
Refering to a kind of degree of fatigue analysis method based on pulse characteristics that Fig. 3, the present embodiment are provided, in above-described embodiment
On the basis of 2, the first pulse data is processed by preset algorithm, obtaining interim pulse data includes：
S111, by Fourier transformation, the pulse time-domain signal in the first pulse data is converted into pulse frequency-region signal；
S112, extraction CF and its correspondence amplitude in pulse power spectrum, enter to CF and its correspondence amplitude
Row classification is processed.
In the present embodiment, because when the first pulse data is obtained, the signal for being recorded is pulse time-domain signal, in order to
It is easy to subsequent analysis to operate, it is necessary first to which this pulse time-domain signal is converted into pulse frequency-region signal, specific time-domain signal turns
The method of frequency-region signal belongs to prior art, will not be repeated here；Secondly, current embodiment require that extracting pulse in pulse data
Power spectrum, and CF and its correspondence amplitude are extracted in pulse power spectrum, specific extraction scheme will be given in example 4
With explanation.
The beneficial effect of the present embodiment is, by Fourier transformation, by the pulse time-domain signal in the first pulse data
Be converted into pulse frequency-region signal, then, CF and its correspondence amplitude extracted in pulse power spectrum, to CF and its
Correspondence amplitude carries out classification treatment.Realize the extraction to pulse frequency-region signal, and in subsequent analysis its pulse power spectrum
CF and its correspondence amplitude provide data basis.
Embodiment 4：
The embodiment of the present invention 4 provides the 4th preferred embodiment of the degree of fatigue analysis method based on pulse characteristics, such as
Fig. 4 show the degree of fatigue analysis method fourth embodiment flow chart based on pulse characteristics provided in an embodiment of the present invention.
Refering to a kind of degree of fatigue analysis method based on pulse characteristics that Fig. 4, the present embodiment are provided, in above-described embodiment
On the basis of 3, the feature for extracting interim pulse data includes：
S121, the direct feature and indirect feature of extracting pulse power spectrum, wherein, direct feature includes main peak amplitude, peak
Value frequency, center of gravity amplitude and gravity frequency, indirect feature include general power, low frequency power and the high frequency power ratio of power spectrum.
Specifically：
The principal character of pulse power spectrum includes two classes：Direct feature and indirect feature, wherein direct feature includes main peak
Amplitude, crest frequency, center of gravity amplitude, gravity frequency；Indirect feature includes the low frequency and height of power spectrum general power TP and power spectrum
Frequency power ratio LF/HF.The following is indirect frequecy characteristic and the calculation of direct frequecy characteristic.
(1) to the calculating of indirect frequecy characteristic：
Low frequency and calculating of the high frequency power than LF/HF, because the sample frequency of data is 100Hz, 3 minutes altogether, altogether
18000 data are actual only to have used 16384 data.The Fourier transformation that 100Hz carries out 16384 points (should be noted
It is that this 16384 is not data amount check, but the points of Fourier transformation).Each point correspondence 0.0061Hz, then low frequency
Preceding 24 points of 0-0.15Hz correspondences, all sampled points that high frequency the 25th Frequency point of correspondence starts.
1st, the power of preceding 24 data in cumulative calculation power spectrum and, and the 25th Frequency point backward power and.
2nd, by the two power and it is divided by, obtains LF/HF.
3rd, LF/HF and threshold value are compared, obtain the judgement of degree of fatigue.
Preferably, in the present embodiment, the frequency partition to frequency spectrum is：Below 0.04Hz, 0.04Hz-0.15Hz,
More than 0.15Hz, is referred to as very low frequencies (Very Low Frequency power components of HRV
Spectrum, VLF), low frequency (Low Frequency power components of HRV spectrum), high frequency (High
Frequency components of HRV spectrum)
(2) the directly calculating of frequecy characteristic：
1st, during main peak amplitude refers to pulse power spectrum, the maximum amplitude in addition to direct current peak value.
2nd, crest frequency refers to, the corresponding frequency of main peak amplitude.
3rd, during center of gravity amplitude refers to pulse power spectrum, by amplitude be weighted it is average after, the average amplitude being calculated.
4th, gravity frequency refers to, the corresponding frequency of center of gravity amplitude.
Wherein, gravity frequency is the migration situation for evaluating power spectrum curve center of gravity.Wavelength coverage power spectral density plot
Gravity frequency can preferably reflect in frequency spectrum the frequency for accounting for the larger signal component of component, it is also possible to reflect whole pulse work(
The migration situation of rate spectrum.
Aforementioned four feature constitutes the feature description for any one pulse time series, establishes pulse time sequence
Mapping relations between row and feature.
Main peak amplitude is the maximum and its respective frequencies searched in pulse power spectrum, searches for relatively simple, herein no longer
Repeat.
Gravity frequency fgComputing formula it is as follows：
Wherein, p (f) refers to the amplitude of power spectrum, and f refers to the corresponding frequencies of p (f), f1And f2Refer to frequency when calculating gravity frequency
Bound.
Finally, the indirect frequecy characteristic and direct frequecy characteristic that will be obtained as stated above, i.e. main peak amplitude, peak value are frequently
The low frequency and high frequency power of rate, center of gravity amplitude, gravity frequency, power spectrum general power TP and power spectrum enter than LF/HF, feeding SVM
Row training.Wherein, the mode of training will not be repeated here as above described in example.
The beneficial effect of the present embodiment is to extract the direct feature and indirect feature of pulse power spectrum, wherein, it is directly special
Levy including main peak amplitude, crest frequency, center of gravity amplitude and gravity frequency, indirect feature includes general power, the low frequency of power spectrum
Power and high frequency power ratio.The accurate calculating and extraction to the feature in pulse data are realized, for training SVM is provided accurately
Data basis.
Embodiment 5：
The embodiment of the present invention 5 provides the 5th preferred embodiment of the degree of fatigue analysis method based on pulse characteristics, such as
Fig. 5 show the embodiment flow chart of degree of fatigue analysis method the 5th based on pulse characteristics provided in an embodiment of the present invention.
Refering to a kind of degree of fatigue analysis method based on pulse characteristics that Fig. 5, the present embodiment are provided, in above-described embodiment
On the basis of 1, determine that its corresponding degree of fatigue includes according to the second degree of fatigue label：
S31, degree of fatigue is divided three classes, respectively mental fatigue, spirit are general and energetic；
S32, its correspondence degree of fatigue label is determined respectively for three class degree of fatigues.
Preferably, the present embodiment is not limited to be divided three classes degree of fatigue, in order to improve the nicety of grading to degree of fatigue,
Degree of fatigue can be divided into multiclass.
Preferably, the present embodiment can be according to the different type of measurand and testing requirement, can be by degree of fatigue
Classified by other types, for example, on the premise of energetic, being further divided into spirit very full and spiritual general
It is full etc..
The beneficial effect of the present embodiment is to be divided three classes by by degree of fatigue, and respectively mental fatigue, spirit are general
And it is energetic, it is that three class degree of fatigues determine its correspondence degree of fatigue label respectively then.For degree of fatigue detection with
Judge to provide more intuitive result, be easy to recognize and record.
Embodiment 6：
The embodiment of the present invention 6 provides the 6th preferred embodiment of the degree of fatigue analytical equipment based on pulse characteristics, such as
Fig. 6 show the degree of fatigue analytical equipment sixth embodiment structured flowchart based on pulse characteristics provided in an embodiment of the present invention.
Refering to a kind of degree of fatigue analytical equipment based on pulse characteristics that Fig. 6, the present embodiment are provided, the device includes：
Tired label initial module 10, for training grader by the first initial pulse data, obtains pulse data
Degree of fatigue sorter model, wherein, the first pulse data includes the first degree of fatigue label；
Tired Tag Estimation module 20, for by pulse data degree of fatigue sorter model to freshly harvested second pulse
The corresponding second degree of fatigue label of data is predicted, and obtains the second degree of fatigue label；
Degree of fatigue determining module 30, for determining its corresponding degree of fatigue according to the second degree of fatigue label.
Preferably, tired label initial module 10 includes interim pulse data acquiring unit 11, pulse characteristics extraction unit
12 and classifier training unit 13, wherein,
Interim pulse data acquiring unit 11 is used to process the first pulse data by preset algorithm, obtains interim arteries and veins
Fight data；
Pulse characteristics extraction unit 12 is used to extract the feature of interim pulse data；
Classifier training unit 13 is used to for feature and its first degree of fatigue label to be integrated into training set, and by training
Collection training grader, obtains sorter model parameter.
Preferably, interim pulse data acquiring unit 11 is additionally operable to：
By Fourier transformation, the pulse time-domain signal in the first pulse data is converted into pulse frequency-region signal；
CF and its correspondence amplitude are extracted in pulse power spectrum, CF and its correspondence amplitude are classified
Treatment.
Preferably, pulse characteristics extraction unit 12 is additionally operable to：
The direct feature and indirect feature of pulse power spectrum are extracted, wherein, direct feature includes main peak amplitude, peak value frequently
Rate, center of gravity amplitude and gravity frequency, indirect feature include general power, low frequency power and the high frequency power ratio of power spectrum.
Preferably, degree of fatigue determining module 30 includes degree of fatigue taxon 31 and tired tag determination unit 32,
Wherein,
Degree of fatigue taxon 31 is used to be divided three classes degree of fatigue, respectively mental fatigue, spirit it is general and
It is energetic；
Tired tag determination unit 32 is used to determine its correspondence degree of fatigue label respectively for three class degree of fatigues.
The beneficial effect of the present embodiment is to train grader by the first initial pulse data, obtains pulse data
Degree of fatigue sorter model, wherein, the first pulse data includes the first degree of fatigue label；By pulse data degree of fatigue point
Class device model is predicted to the corresponding second degree of fatigue label of freshly harvested second pulse data, obtains the second degree of fatigue
Label；Its corresponding degree of fatigue is determined according to the second degree of fatigue label.Allow user conveniently and efficiently to tired journey
Degree is judged and is measured.Meanwhile, the device also allows for user's carrying, and good experience is brought to user.
Embodiments of the present invention are the foregoing is only, the scope of the claims of the invention is not thereby limited, it is every using this
Equivalent structure or equivalent flow conversion that description of the invention and accompanying drawing content are made, or directly or indirectly it is used in other correlations
Technical field, is included within the scope of the present invention.
Claims (10)
1. a kind of degree of fatigue analysis method based on pulse characteristics, it is characterised in that methods described includes：
Grader is trained by the first initial pulse data, pulse data degree of fatigue sorter model is obtained, wherein, it is described
First pulse data includes the first degree of fatigue label；
By the pulse data degree of fatigue sorter model to corresponding second degree of fatigue of freshly harvested second pulse data
Label is predicted, and obtains the second degree of fatigue label；
Its corresponding degree of fatigue is determined according to the second degree of fatigue label.
2. the degree of fatigue analysis method based on pulse characteristics according to claim 1, it is characterised in that described by first
The the first pulse data training grader for beginning, obtaining pulse data degree of fatigue sorter model includes：
First pulse data is processed by preset algorithm, obtains interim pulse data；Extract the interim Pulse Rate
According to feature；
The feature and its first degree of fatigue label are integrated into training set, and the classification is trained by the training set
Device, obtains sorter model parameter.
3. the degree of fatigue analysis method based on pulse characteristics according to claim 2, it is characterised in that described by default
Algorithm is processed first pulse data, and obtaining interim pulse data includes：
By Fourier transformation, the pulse time-domain signal in first pulse data is converted into pulse frequency-region signal；
CF and its correspondence amplitude are extracted in pulse power spectrum, the CF and its correspondence amplitude are classified
Treatment.
4. the degree of fatigue analysis method based on pulse characteristics according to claim 3, it is characterised in that the extraction institute
The feature for stating interim pulse data includes：
The direct feature and indirect feature of the pulse power spectrum are extracted, wherein, the direct feature includes main peak amplitude, peak value
Frequency, center of gravity amplitude and gravity frequency, the indirect feature include general power, low frequency power and the high frequency power of power spectrum
Than.
5. the degree of fatigue analysis method based on pulse characteristics according to claim 1, it is characterised in that described according to institute
State the second degree of fatigue label and determine that its corresponding degree of fatigue includes：
The degree of fatigue is divided three classes, respectively mental fatigue, spirit are general and energetic；
For the three classes degree of fatigue determines its correspondence degree of fatigue label respectively.
6. a kind of degree of fatigue analytical equipment based on pulse characteristics, it is characterised in that described device includes：
Tired label initial module, for training grader by the first initial pulse data, obtains pulse data fatigue journey
Degree sorter model, wherein, first pulse data includes the first degree of fatigue label；
Tired Tag Estimation module, for by the pulse data degree of fatigue sorter model to freshly harvested second Pulse Rate
It is predicted according to corresponding second degree of fatigue label, obtains the second degree of fatigue label；
Degree of fatigue determining module, for determining its corresponding degree of fatigue according to the second degree of fatigue label.
7. the degree of fatigue analytical equipment based on pulse characteristics according to claim 6, it is characterised in that the fatigue mark
Signing initial module includes interim pulse data acquiring unit, pulse characteristics extraction unit and classifier training unit, wherein,
The interim pulse data acquiring unit is used to process first pulse data by preset algorithm, obtains interim
Pulse data；
The pulse characteristics extraction unit is used to extract the feature of the interim pulse data；
The classifier training unit is used to for the feature and its first degree of fatigue label to be integrated into training set, and by institute
State training set and train the grader, obtain sorter model parameter.
8. the degree of fatigue analytical equipment based on pulse characteristics according to claim 7, it is characterised in that the interim arteries and veins
Data capture unit of fighting is additionally operable to：
By Fourier transformation, the pulse time-domain signal in first pulse data is converted into pulse frequency-region signal；
CF and its correspondence amplitude are extracted in pulse power spectrum, the CF and its correspondence amplitude are classified
Treatment.
9. the degree of fatigue analytical equipment based on pulse characteristics according to claim 8, it is characterised in that the pulse is special
Extraction unit is levied to be additionally operable to：
The direct feature and indirect feature of the pulse power spectrum are extracted, wherein, the direct feature includes main peak amplitude, peak value
Frequency, center of gravity amplitude and gravity frequency, the indirect feature include general power, low frequency power and the high frequency power of power spectrum
Than.
10. the degree of fatigue analytical equipment based on pulse characteristics according to claim 6, it is characterised in that the fatigue
Degree determining module includes degree of fatigue taxon and tired tag determination unit, wherein,
The degree of fatigue taxon is used to be divided three classes the degree of fatigue, respectively mental fatigue, spirit it is general with
And it is energetic；
The tired tag determination unit is used to determine its correspondence degree of fatigue label respectively for the three classes degree of fatigue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611178555.0A CN106691402A (en) | 2016-12-19 | 2016-12-19 | Fatigue level analysis method and device based on pulse characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611178555.0A CN106691402A (en) | 2016-12-19 | 2016-12-19 | Fatigue level analysis method and device based on pulse characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106691402A true CN106691402A (en) | 2017-05-24 |
Family
ID=58938417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611178555.0A Pending CN106691402A (en) | 2016-12-19 | 2016-12-19 | Fatigue level analysis method and device based on pulse characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106691402A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019062595A1 (en) * | 2017-09-30 | 2019-04-04 | 深圳市岩尚科技有限公司 | Method for quality evaluation of photoplethysmogram |
| CN110090029A (en) * | 2018-01-28 | 2019-08-06 | 北京师范大学 | Emotional measurement system, Intelligent bracelet and portable terminal |
| CN110141205A (en) * | 2019-05-27 | 2019-08-20 | 深圳市是源医学科技有限公司 | Resistance to compression data, the test method of fatigue data and device based on HRV technology |
| CN110579622A (en) * | 2019-08-22 | 2019-12-17 | 西安理工大学 | Metal particle flow velocity measuring device and method based on triangular electrode capacitance sensor |
| CN112617772A (en) * | 2021-01-05 | 2021-04-09 | 上海工程技术大学 | Driving fatigue identification method and system based on pulse wave signals |
| CN114246593A (en) * | 2021-12-15 | 2022-03-29 | 山东中科先进技术研究院有限公司 | Electroencephalogram, electrooculogram and heart rate fused fatigue detection method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102271584A (en) * | 2009-10-29 | 2011-12-07 | 松下电器产业株式会社 | Organism fatigue evaluation device and organism fatigue evaluation method |
| WO2015037281A1 (en) * | 2013-09-11 | 2015-03-19 | 株式会社日立システムズ | Screening system for fatigue and stress |
| CN104825174A (en) * | 2015-04-17 | 2015-08-12 | 深圳市元征科技股份有限公司 | Fatigue state detection method and terminal |
| CN105877766A (en) * | 2016-06-21 | 2016-08-24 | 东北大学 | Mental state detection system and method based on multiple physiological signal fusion |
| CN106137181A (en) * | 2015-04-13 | 2016-11-23 | 上海帝仪科技有限公司 | For obtaining the system of the fatigue characteristic of user, method and apparatus |
-
2016
- 2016-12-19 CN CN201611178555.0A patent/CN106691402A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102271584A (en) * | 2009-10-29 | 2011-12-07 | 松下电器产业株式会社 | Organism fatigue evaluation device and organism fatigue evaluation method |
| WO2015037281A1 (en) * | 2013-09-11 | 2015-03-19 | 株式会社日立システムズ | Screening system for fatigue and stress |
| CN106137181A (en) * | 2015-04-13 | 2016-11-23 | 上海帝仪科技有限公司 | For obtaining the system of the fatigue characteristic of user, method and apparatus |
| CN104825174A (en) * | 2015-04-17 | 2015-08-12 | 深圳市元征科技股份有限公司 | Fatigue state detection method and terminal |
| CN105877766A (en) * | 2016-06-21 | 2016-08-24 | 东北大学 | Mental state detection system and method based on multiple physiological signal fusion |
Non-Patent Citations (2)
| Title |
|---|
| 任亚莉: "基于脉搏传感测值和主成分分析对精神疲劳状态的识别", 《中国组织工程研究》 * |
| 蒋建春 等: "基于脉搏IMF时频特征和SVDD的驾驶员疲劳检测", 《计算机科学》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019062595A1 (en) * | 2017-09-30 | 2019-04-04 | 深圳市岩尚科技有限公司 | Method for quality evaluation of photoplethysmogram |
| CN110090029A (en) * | 2018-01-28 | 2019-08-06 | 北京师范大学 | Emotional measurement system, Intelligent bracelet and portable terminal |
| CN110141205A (en) * | 2019-05-27 | 2019-08-20 | 深圳市是源医学科技有限公司 | Resistance to compression data, the test method of fatigue data and device based on HRV technology |
| CN110141205B (en) * | 2019-05-27 | 2022-03-11 | 深圳市是源医学科技有限公司 | HRV technology-based compression data and fatigue data testing method and device |
| CN110579622A (en) * | 2019-08-22 | 2019-12-17 | 西安理工大学 | Metal particle flow velocity measuring device and method based on triangular electrode capacitance sensor |
| CN110579622B (en) * | 2019-08-22 | 2021-08-06 | 西安理工大学 | Metal particle flow velocity measuring device and method based on triangular electrode capacitance sensor |
| CN112617772A (en) * | 2021-01-05 | 2021-04-09 | 上海工程技术大学 | Driving fatigue identification method and system based on pulse wave signals |
| CN112617772B (en) * | 2021-01-05 | 2022-12-27 | 上海工程技术大学 | Driving fatigue identification method and system based on pulse wave signals |
| CN114246593A (en) * | 2021-12-15 | 2022-03-29 | 山东中科先进技术研究院有限公司 | Electroencephalogram, electrooculogram and heart rate fused fatigue detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106691402A (en) | Fatigue level analysis method and device based on pulse characteristics | |
| CN106108889B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| CN102429649B (en) | Continuous blood pressure measuring device | |
| CN102397064B (en) | Continuous blood pressure measuring device | |
| CN102488503B (en) | Continuous blood pressure measurer | |
| CN101732050B (en) | Photoelectric volume wave-based breathing rate monitoring method | |
| CN110251105A (en) | Noninvasive blood pressure measuring method, device, equipment and system | |
| CN105193431A (en) | Device for analyzing mental stress state of human body | |
| CN106691406A (en) | Detection method of vascular elasticity and blood pressure based on single probe photoplethysmography pulse wave | |
| US20180228383A1 (en) | Blood pressure measuring device, blood pressure measurement method and blood pressure measurement program | |
| RU2011122787A (en) | METHODS AND SYSTEMS FOR NON-INVASIVE MEASUREMENT OF GLUCOSE LEVELS | |
| CN110037668A (en) | The system that pulse signal time-space domain binding model judges age, health status and malignant arrhythmia identification | |
| CN113679369B (en) | Evaluation method of heart rate variability, intelligent wearable device and storage medium | |
| CN113520333A (en) | Method, device and equipment for determining core body temperature and readable medium | |
| CN105433934A (en) | Method for personal health examination and evaluation based on vital sign monitor bracelet | |
| CN113907727B (en) | Beat-by-beat blood pressure measurement system and method based on photoplethysmography | |
| Tiwari et al. | Stress and anxiety measurement" in-the-wild" using quality-aware multi-scale hrv features | |
| CN110200642A (en) | A kind of measurement method and terminal of cognitive load and psychological pressure | |
| KR20180043517A (en) | Method and device for the measurement of energy consumption based on vital/motion signals | |
| CN109171674A (en) | Toilet lid and hemodynamic parameter measurement method based on the toilet lid | |
| Dewi et al. | Cardiovascular system modeling using windkessel segmentation model based on photoplethysmography measurements of fingers and toes | |
| Prabhakar et al. | Performance analysis of GMM classifier for classification of normal and abnormal segments in PPG signals | |
| Zhidong | Noninvasive diagnosis of coronary artery disease based on instantaneous frequency of diastolic murmurs and SVM | |
| Shah et al. | Photoplethysmographic based heart rate variability for different physiological conditions | |
| Gharbi et al. | System for body and mind monitoring in coaching process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| 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: 20170524 |