CN109770920A - Intense strain method of discrimination and its system based on wearable ECG signal - Google Patents
Intense strain method of discrimination and its system based on wearable ECG signal Download PDFInfo
- Publication number
- CN109770920A CN109770920A CN201910099443.3A CN201910099443A CN109770920A CN 109770920 A CN109770920 A CN 109770920A CN 201910099443 A CN201910099443 A CN 201910099443A CN 109770920 A CN109770920 A CN 109770920A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- signal
- ecg signal
- wave
- intense strain
- 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
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention patent discloses a kind of intense strain method of discrimination and its system based on wearable ECG signal, it is obtained including electrocardiosignal, Filtering of ECG Signal, electrocardiosignal feature extraction and intense strain differentiate four steps, after to the denoising of collected Filtering of ECG Signal, extract characteristic value, using algorithm of support vector machine, characteristic value collection data are randomly selected as sample group, classifier is trained using sample group, after the completion of training, classified using this classifier to the collected signal of institute, realize the differentiation of intense strain, the judgement of mental status is done in advance, complete effective detection to physiology and spiritual index.
Description
Fields
The present invention relates to ambulatory ecg signal process fields, and in particular to a kind of anxiety based on wearable ECG signal
Mood method of discrimination and its system.
Background technique
Increasingly increase in real time monitoring demand of the modern society people for physical condition, more and more people wish
Can physiology to oneself and mental status have monitoring in real time and understand, to obtain when there is physiology and psychic problems
To giving warning in advance and timely solve.With the continuous progress of science and technology, now for the real-time monitoring of body physiological index
Through very mature, and there are also to be developed and researchs for the Emotion identification of spirit aspect, wherein the study found that electrocardiosignal and anxiety
There is very big connection between mood, the electrocardiosignal under the performance and stress situation of stress reaction can occur accordingly for people
Change.
Anxiety is a kind of normal reaction of the people under stress situation, it is so-called stress (stress) refer to people or organism
Generated a kind of reflection state for adapting to environment under the action of certain environmental stimulus, i.e., in certain social living environment
In, to stimulation and situation that a people can have an impact, after being perceived by it and making subjective assessment, will generate corresponding
Some psychology physiological variations, to make corresponding reaction to stimulation.If it is biggish that this stimulation or situation need people to make
Effort goes to be adapted to, as soon as or even beyond the adaptability that people can bear, at this moment will appear stress.Appropriate stress
The various functions of whole body can be transferred, make the reaction to avert danger, so be on the defensive and adapt to it is compensatory, in some cases
Non-specific adaptation system stress can also be mobilized, enhances the adaptability of body itself, is properly termed as Eustress above;This
It is outer there are also exist to human body the pathologic that threatens stress, there is decompensation, Bu Nengji in the adaptation mechanism of body in such cases
When reply stressor bring stimulation, so as to cause organism metabolism and endocrine disorder.On the whole, under stress situation,
Fierce variation occurs for biochemical system, and adrenaline and each glandular secretion increase, and physical vigor enhancing is in entire body
Mobilize to the fullest extent state, to cope with unexpected mutation, is chronically at stress situation, unfavorable to the health of people, it would be possible to body
Irreversible damage is caused, or even can be dangerous.
In the state of anxiety, due to cardiac sympathetic nerve excitement, while pituitary adrenal hormone secretion increases, in blood
Norepinephrine and epinephrine contents increase, result in increased heart rate, myocardial contractive power increases, periphery drag overall liter
A series of high cardiovascular variations, these variations can be reacted accordingly on electrocardiogram, normal person's usually heart contraction
When, waveform is normal and regular;When nervous, due to palpitating speed, the narrower intervals of wave;Be further aggravated if anxiety, if
T wave in electrocardiogram almost disappears (recovery process of reflection ventricle excitement is abnormal);ST section in electrocardiogram is raised
(usually related with cardiac Troponin level raising);After releasing anxiety, waveform restores normal again.When stress situation continues
It is also possible to induce Stress cardiomyopathy when time is longer or stress reaction is strong, furthermore in the generation of cardiovascular acute events
In, mental emotion stress have been considered as a trigger, become the important original of triggering acute myocardial infarction AMI, sudden cardiac death
Cause, thus if can judge intense strain by obtaining electrocardiosignal, will be to ambulatory ecg signal process field again
One big technological break-through.
Summary of the invention
The present invention is exactly directed to the problems of the prior art, provides a kind of nervous feelings based on wearable ECG signal
Thread method of discrimination and its system, including electrocardiosignal acquisition, Filtering of ECG Signal, electrocardiosignal feature extraction and intense strain
Differentiate four steps, after to the denoising of collected Filtering of ECG Signal, extracts characteristic value, using algorithm of support vector machine,
Characteristic value collection data are randomly selected as sample group, classifier is trained using sample group, after the completion of training, this point
Class device can be used for classifying extracted electrocardiosignal, the differentiation of intense strain be realized, only by for single electrocardiosignal
Acquisition in advance judges mental status to reach with processing, completes effective monitoring to physiology and spiritual index.
To achieve the goals above, the technical solution adopted by the present invention is that: the intense strain based on wearable ECG signal
Method of discrimination, comprising the following steps:
S1, electrocardiosignal obtain: by wearing clothing acquire electrocardiosignal, and convert analog signals into digital signal into
Row is to be processed;
S2, Filtering of ECG Signal: will be by step S1 treated digital signal using the electrocardio letter based on wavelet transformation
Number filtering, the ECG signal denoised, the ECG signal have temporal signatures and time scale feature, at least by
It is divided into PTQRS wave band;
The feature extraction of electrocardiosignal: S3 is alternately sentenced using classics PT algorithm by two groups of noise thresholds and signal threshold value
It is disconnected, corresponding horizontal parameters are adjusted according to judging result, are judged next time when judging next time according to horizontal parameters adjustment
Two groups of threshold values, circulation is until signal is finished by all detections;
S4, intense strain differentiate: it is realized using algorithm of support vector machine, is realized using algorithm of support vector machine, benefit
It uses data in existing database to be trained as sample group using sample group to classifier, after the completion of training, utilizes step
The signal characteristic that S3 is extracted inputs trained training aids as test group, can complete the classification to extracted signal.
As an improvement of the present invention, temporal signatures and time scale feature include at least Q, R, S in the step S2
Peak amplitude, average value, maximum value, minimum value, standard deviation, the interval P-Q, the interval Q-S, the max min at the interval S-T are put down
Mean value and standard deviation.
Improved as another kind of the invention, wavelet transformation is divided into wavelet decomposition in the step S2, the setting of threshold value and
Three steps of signal reconstruction, the size of the threshold value are updated by following formula:
SPKI=0.125PEAKI+0.875SPKI if PEAKI is signal peak
NPKI=0.125PEAKI+0.875NPKI if PEAKI is noise peak
THRESHOLDI1=NPKI+0.25 (SPKI-NPKI)
Wherein, SPKI indicates the peak amplitude of QRS wave;NPKI indicates the peak amplitude of non-QRS wave;THRESHOLD I1 table
Show the threshold value of detected peak value;If detected peak value is greater than THRESHOLD I1, it is considered to be SPKI, otherwise it is assumed that
It is NPKI.
As another improvement of the invention, low-pass filter involved in PT algorithm is integral coefficient in the step S3,
Wherein transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
The high-pass filter being related to subtracts low-pass filter by all-pass filter and constitutes, wherein transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
Y (n)-y (n-1)=- x (n)+x (n-16)-x (n-17)+x (n-32).
As a further improvement of the present invention, PT algorithm further includes derivative filter in the step S3, the derivative
Filter is in order to enhance the slope of P wave, T wave and QRS, transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
It further include square filtering device, the square filtering device further enhances the slope of P wave, T wave and QRS, highlights
Q wave and S wave, the output y (n) of the square filtering device and the relationship of input x (n) are as follows:
Y (n)=x2(n)。
To achieve the goals above, the present invention also the technical solution adopted is that: the nervous feelings based on wearable ECG signal
Thread judgement system, characterized by comprising: ecg signal acquiring module, Filtering of ECG Signal module, electrocardiosignal feature extraction
Module and intense strain discrimination module,
The ecg signal acquiring module is used to acquire electrocardiosignal, and passes after converting analog signals into digital signal
It send to be processed to the progress of Filtering of ECG Signal module;
Denoising of the Filtering of ECG Signal module for electrocardiosignal goes to interfere;
The electrocardiosignal characteristic extracting module is used to extract the characteristic value in electrocardiosignal, then passes through intense strain
Discrimination module, using algorithm of support vector machine, using data in existing database as sample group, using sample group to classification
Device is trained, and after the completion of training, is inputted and is passed through as test group using the signal characteristic that electrocardiosignal characteristic extracting module is extracted
Trained classifier is crossed, the classification results of the electrocardiosignal are obtained.
As an improvement of the present invention, the ecg signal acquiring module include electrocardiosignal front-end acquisition device and
Hardware signal processing device,
The electrocardiosignal front-end acquisition device include at least three stemness electrocardioelectrodes, the stemness electrocardioelectrode by
Conductive fabric is made;
The hardware signal processing device is integrated on sensor node, by the conditioning circuit of electrocardiosignal, microcontroller
Module, power management module and Bluetooth communication modules are constituted;
It will be by the collected electrocardiosignal of electrocardiosignal front-end acquisition device by improving and microcontroller after filter circuit
Device carries out A/D sampling to it, converts analog signals into digital signal, and transmit data to terminal by bluetooth and counted
According to processing.
Compared with prior art, the invention patent the utility model has the advantages that
1, belong to the classification of non-linear type for the differentiation of intense strain, SVM algorithm can be reflected by using non-linear
Penetrating algorithm and converting high-dimensional feature space for the sample of low-dimensional input space linearly inseparable makes its linear separability, so that
High-dimensional feature space carries out linear analysis using nonlinear characteristic of the linear algorithm to sample and is possibly realized;It is based on structure simultaneously
Optimal hyperlane is constructed in feature space on risk minimization theory, is allowed to obtain global optimization, and in entire sample
The expectation in this space meets certain upper bound with some probability.
2, the proposition of this method is perfect carries out emotion based on single physiological signal and sentences method for distinguishing, decreases wearing
The data volume stored needed for formula equipment, and intelligence wearing clothing is personal flexible material, and flexible material allows to wear at any time
Will not be thick and heavy with or inconvenience be brought to wearer, personal material ensure that stemness electrocardioelectrode and skin are kept
Contact is to collect lasting and relatively stable signal.
3, this method and system in addition to be suitable for people's daily life, for the monitoring of old man children, patient monitoring with
Outside, the Working Status Monitoring and military aspect of special work post worker be can be also used for, such as this system monitoring participates in war
The physiology and mental status of food for powder make timely corresponding and processing to send the soldier in battlefield.
4, this method can also be applied in the portability equipment such as wearable watch, bracelet, can greatly enhance people
For own health status sensing capability, and hommization setting facilitate user to be configured according to own situation, effectively
Prompt human body health degree, help user physical condition is assessed.
Detailed description of the invention
Fig. 1 is the working drawing of sensor node in hardware circuit signal processing apparatus of the present invention;
Fig. 2 is the dtr signal figure that the present invention tests;
Fig. 3 is that the signal that the present invention tests preferably is schemed;
Fig. 4 is the signal difference figure that the present invention tests.
Specific embodiment
Below with reference to drawings and examples, the present invention is described in detail.
Embodiment 1
Intense strain method of discrimination based on wearable ECG signal, comprising the following steps:
S1, electrocardiosignal obtain: by wearing clothing acquire electrocardiosignal, and convert analog signals into digital signal into
Row is to be processed.
S2, Filtering of ECG Signal: using the ECG filtering based on wavelet transformation, wavelet transformation is divided into wavelet decomposition, threshold value
Three steps of setting and signal reconstruction, handle, denoised in details of the threshold setting section to part signal
ECG signal, since ECG signal itself can be divided into the wave bands such as PTQRS, the feature of different-waveband is different, is typical
Biomedicine signals with temporal signatures Yu time scale feature are well suited for removing collected signal with wavelet transformation
The problems such as various noise jammings of middle institute's band, Hz noise and baseline drift;It will treated that digital signal is adopted by step S1
With the Filtering of ECG Signal based on wavelet transformation, the ECG signal denoised, the ECG signal has temporal signatures
With time scale feature, it is at least divided into PTQRS wave band.
The feature extraction of electrocardiosignal: S3 is alternately sentenced using classics PT algorithm by two groups of noise thresholds and signal threshold value
It is disconnected, corresponding horizontal parameters are adjusted according to judging result, are judged next time when judging next time according to horizontal parameters adjustment
Two groups of threshold values, circulation is until signal is finished by all detections;
Low-pass filter involved in PT algorithm is integral coefficient, for reducing computation complexity, wherein transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
The cutoff frequency of low-pass filter is 11Hz, has the delay of 5 samples, there is the decaying of 35dB at 60Hz, is designed
The purpose of this low-pass filter is to cross noise filtering.
The high-pass filter that PT algorithm is related to is the drift situation in order to filter off signal at low frequency, high frequency filter by
All-pass filter subtracts low-pass filter composition, transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
Y (n)-y (n-1)=- x (n)+x (n-16)-x (n-17)+x (n-32)
The cutoff frequency of this filter is 5Hz, there is the delay of 16 samples.
The derivative filter of PT algorithm design is used to remove the flip-flop of input, and high frequency obtains linear gain (enhancing
The slope of P wave, T wave and QRS), it can regard a high-pass filter, transmission function as are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
This filter has the delay of 2 samples.
The effect of square filtering device be so that sample value be positive number, further enhance the slope of P wave, T wave and QRS, it is convex
Show Q wave and S wave.
The output y (n) of square filtering device and the relationship of input x (n) are as follows:
Y (n)=x2(n)
In derivative filter in front, QRS complex will generate many waves, and PT algorithm is using integration filter come smooth
Output.The output y (n) of integration filter and the relationship of input x (n) are as follows:
When N value takes excessive, QRS complex and T wave can be flooded.When N takes too small, many waves can be generated in QRS complex.It is right
In the case that sampling frequency is 200Hz, it is 30 suitable that N takes.
The delay of about 21 samples of integration filter.
Adaptive threshold is used to the signal peak and noise peak searched in ECG signal.SPKI represents corresponding QRS
The peak amplitude of wave.NPKI represents the peak amplitude of non-QRS wave.THRESHOLD I1, which is represented, to be used to distinguish detected peak value
Threshold value.
If detected peak value is greater than THRESHOLD I1, it is considered to be SPKI, otherwise it is assumed that being NPKI.
The size of threshold value is updated by following formula:
SPKI=0.125PEAKI+0.875SPKI if PEAKI is signal peak
NPKI=0.125PEAKI+0.875NPKI if PEAKI is noise peak
THRESHOLDI1=NPKI+0.25 (SPKI-NPKI)
The extraction that classical PT algorithm carries out feature is input a signal into after filtering, and PQRST various point locations are being accurately positioned
Afterwards, relevant primitive character value is extracted as parameter using each key point, wherein temporal signatures value includes the peak Q, R, S amplitude, is averaged
The parameters such as value, maximum value, minimum value, standard deviation, the interval P-Q, the interval Q-S, the max min at the interval S-T, average value,
The parameters such as standard deviation.Another important indicator is the characteristic value that HRV (heart rate variability rate) poem rains, and refers to the wink of successive heartbeat
When heart rate or Micro-fluctuations of heartbeat RR interphase by shooting.The measurement of HRV can analyze out how the variation between successive heartbeat is
Change with time change.In addition to temporal signatures value, frequency domain character value is also critically important index.The frequency-domain analysis of HRV
Be from the RR blank signal of electrocardiogram extract frequency domain in terms of parameter, such as peak value frequency, with interior power feature click through
Row statistics, HRV frequency domain character value mainly includes the features such as crest frequency and low frequency (LF), high frequency (HF) and low-and high-frequency ratio.
S4, intense strain differentiate: it is realized using support vector machines (SVM) algorithm, the case where for linearly inseparable,
Converting high-dimensional feature space for the sample of low-dimensional input space linearly inseparable by using non-linear map makes its line
Property can divide so that high-dimensional feature space linear analysis becomes is carried out to the nonlinear characteristic of sample using linear algorithm can
Energy;Second is that it is based on constructing optimal hyperlane in feature space on structural risk minimization theory, so that learner obtains
Global optimization, and certain upper bound is met with some probability in the expectation of entire sample space.It is extracted from existing database
Primitive character value set in randomly select data as sample group (sample group should include the signal and not under intense strain
Two class signal of signal under intense strain), the training of SVM classifier, the feature that will be extracted in previous step are completed using sample group
It is sent into this classifier to classify, obtains the classification results of the electrocardiosignal.Judge this result whether be intense strain and
Alert process is carried out for intense strain.
Embodiment 2
Intense strain judgement system based on wearable ECG signal, comprising: ecg signal acquiring module, electrocardiosignal
Filter module, electrocardiosignal characteristic extracting module and intense strain discrimination module,
The ecg signal acquiring module is used to acquire electrocardiosignal, and passes after converting analog signals into digital signal
It send to be processed to the progress of Filtering of ECG Signal module;The ecg signal acquiring module includes electrocardiosignal front-end acquisition device
With hardware signal processing device,
The electrocardiosignal front-end acquisition device include at least three stemness electrocardioelectrodes, the stemness electrocardioelectrode by
Conductive fabric is made, and such electrode will not generate stimulation to skin, and can wash and not need to replace, and improves practicability
And comfort.Two of them are respectively positioned at below the clavicle of left and right, and third is placed in right abdomen, by measuring body surface potential,
Obtain I lead electrocardiosignal;
The hardware circuit signal processing apparatus is integrated on sensor node, by the conditioning circuit of electrocardiosignal, micro-control
Device module processed, power management module and Bluetooth communication modules are constituted, as shown in Figure 1, collected electrocardiosignal is believed by electrocardio
Number interface is passed to electrocardiosignal conditioning circuit and is filtered and amplifies, and obtains the signal by tentatively filtering, microcontroller is to it
It carries out A/D sampling and converts analog signals into digital signal, transmit data to terminal by Bluetooth communication modules and carry out data
Be further processed;
Power management module is responsible for powering to entire circuit;
Electrocardiosignal interface is for receiving electrocardiosignal;
Electrocardiosignal conditioning circuit builds the method that peripheral circuit is matched with selected chip interior gain using component
Former electrocardiosignal is amplified, uses capacitance resistance appropriate to be configured to high-pass filter to remove unrelated with signal make an uproar
Sound;
Microprocessor module is for converting analog signals into digital signal;
Bluetooth communication modules are transmitted for data;
The function of the filter module of electrocardiosignal and the intense strain discrimination module of electrocardiosignal is realized in terminal part.
Denoising of the Filtering of ECG Signal module for electrocardiosignal goes to interfere, and is filtered using the ECG based on wavelet transformation
Wave, wavelet transformation are divided into wavelet decomposition, and three steps of setting and signal reconstruction of threshold value believe part in threshold setting section
Number details handled, the ECG signal denoised is different since ECG signal itself can be divided into the wave bands such as PTQRS
The feature of wave band is different, is the typical biomedicine signals with temporal signatures and time scale feature, is well suited for using
Wavelet transformation is come the problems such as removing the various noise jammings of institute's band, Hz noise and baseline drift in collected signal;
Intense strain discrimination module based on electrocardiosignal, which is used, carries out algorithm development based on MATLAB platform, in equipment
When collected electrocardiosignal completes transmission, start to carry out the processing to signal, first by collected electrocardiosignal benefit
Be filtered denoising with wavelet transformation, get rid of wearable device collected signal had Hz noise, noise with
And the problem of baseline drift, the feature extraction of electrocardiosignal, the primitive character that will be extracted are carried out followed by classical PT algorithm
It is input to after being saved among the classifier algorithm based on SVM and classifies to the collected signal of institute, as judgement is positive
Normal state then continues time work of circulation, such as determines and currently belongs to intense strain, then alarms to host computer, is come with this
It reminds user to pay close attention to itself spirit and physical health problem in time, is acquired to reach based on wearable device with this
The electrocardiosignal arrived carries out the purpose of intense strain.
The basic principles, main features and advantages of the present invention have been shown and described above.The technology of the industry
Personnel only illustrate the present invention it should be appreciated that the present invention is not limited by examples detailed above described in examples detailed above and specification
Principle, various changes and improvements may be made to the invention without departing from the spirit and scope of the present invention, these variation and
Improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by appended claims and its waits
Jljl is defined.
Claims (7)
1. the intense strain method of discrimination based on wearable ECG signal, it is characterised in that the following steps are included:
S1, electrocardiosignal obtain: acquiring electrocardiosignal by wearing clothing, and convert analog signals into digital signal and carry out wait locate
Reason;
S2, Filtering of ECG Signal: will be by step S1 treated digital signal using the electrocardiosignal filter based on wavelet transformation
Wave, the ECG signal denoised, the ECG signal have temporal signatures and time scale feature, are at least divided into
PTQRS wave band;
The feature extraction of electrocardiosignal: S3 is alternately judged using classics PT algorithm by two groups of noise thresholds and signal threshold value, root
It is judged that result adjusts corresponding horizontal parameters, the two groups of thresholds judged next time are adjusted according to horizontal parameters when judging next time
Value, circulation is until signal is finished by all detections;
S4, intense strain differentiate: it is realized using algorithm of support vector machine, using data in existing database as sample group,
Classifier is trained using sample group, after the completion of training, is inputted using the signal characteristic that step S3 is extracted as test group
Trained training aids can complete the classification to extracted signal.
2. as described in claim 1 based on the intense strain method of discrimination of wearable ECG signal, it is characterised in that the step
In rapid S2 temporal signatures and time scale feature include at least the peak Q, R, S amplitude, average value, maximum value, minimum value, standard deviation,
The interval P-Q, the interval Q-S, the max min at the interval S-T, average value and standard deviation.
3. as described in claim 1 based on the intense strain method of discrimination of wearable ECG signal, it is characterised in that the step
Wavelet transformation is divided into wavelet decomposition in rapid S2, and three steps of setting and signal reconstruction of threshold value, the size of the threshold value is by following formula
It updates:
SPKI=0.125PEAKI+0.875SPKI if PEAKI is signal peak
NPKI=0.125PEAKI+0.875NPKI if PEAKI is noise peak
THRESHOLD I1=NPKI+0.25 (SPKI-NPKI)
Wherein, SPKI indicates the peak amplitude of QRS wave;NPKI indicates the peak amplitude of non-QRS wave;THRESHOLD I1 indicates institute
Detect the threshold value of peak value;If detected peak value is greater than THRESHOLD I1, it is considered to be SPKI, otherwise it is assumed that being
NPKI。
4. as claimed in claim 2 or claim 3 based on the intense strain method of discrimination of wearable ECG signal, it is characterised in that described
Low-pass filter involved in PT algorithm is integral coefficient in step S3, wherein transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
The high-pass filter being related to subtracts low-pass filter by all-pass filter and constitutes, wherein transmission function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
Y (n)-y (n-1)=- x (n)+x (n-16)-x (n-17)+x (n-32).
5. as described in claim 1 based on the intense strain method of discrimination of wearable ECG signal, it is characterised in that the step
PT algorithm further includes derivative filter in rapid S3, and the derivative filter passes to enhance the slope of P wave, T wave and QRS
Delivery function are as follows:
It exports y (n) and inputs the relationship of x (n) are as follows:
Further include square filtering device, the square filtering device further enhances the slope of P wave, T wave and QRS, highlight Q wave and
S wave, the output y (n) of the square filtering device and the relationship of input x (n) are as follows:
Y (n)=x2(n)。
6. the intense strain judgement system based on wearable ECG signal, characterized by comprising: ecg signal acquiring module, the heart
Electric signal filter module, electrocardiosignal characteristic extracting module and intense strain discrimination module,
The ecg signal acquiring module is used to acquire electrocardiosignal, and is sent to the heart after converting analog signals into digital signal
Electric signal filter module carries out to be processed;
Denoising of the Filtering of ECG Signal module for electrocardiosignal goes to interfere;
The electrocardiosignal characteristic extracting module is used to extract the characteristic value in electrocardiosignal, then differentiates mould by intense strain
Block, using data in existing database as sample group, instructs classifier using sample group using algorithm of support vector machine
Practice, after the completion of training, the signal characteristic extracted using electrocardiosignal characteristic extracting module inputs trained as test group
Classifier obtains the classification results of this electrocardiosignal.
7. as claimed in claim 6 based on the intense strain judgement system of wearable ECG signal, it is characterised in that: the heart
Electrical signal collection module includes electrocardiosignal front-end acquisition device and hardware signal processing device,
The electrocardiosignal front-end acquisition device includes at least three stemness electrocardioelectrodes, and the stemness electrocardioelectrode is knitted by conduction
Object is made;
The hardware signal processing device is integrated on sensor node, by the conditioning circuit of electrocardiosignal, micro controller module,
Power management module and Bluetooth communication modules are constituted;
It will be by the collected electrocardiosignal of electrocardiosignal front-end acquisition device by improving and microcontroller pair after filter circuit
It carries out A/D sampling, converts analog signals into digital signal, and transmit data to terminal by bluetooth and carry out at data
Reason.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910099443.3A CN109770920A (en) | 2019-01-31 | 2019-01-31 | Intense strain method of discrimination and its system based on wearable ECG signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910099443.3A CN109770920A (en) | 2019-01-31 | 2019-01-31 | Intense strain method of discrimination and its system based on wearable ECG signal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109770920A true CN109770920A (en) | 2019-05-21 |
Family
ID=66502955
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910099443.3A Pending CN109770920A (en) | 2019-01-31 | 2019-01-31 | Intense strain method of discrimination and its system based on wearable ECG signal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109770920A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110353704A (en) * | 2019-07-12 | 2019-10-22 | 东南大学 | Mood assessments method and apparatus based on wearable ECG monitoring |
CN111134659A (en) * | 2020-01-17 | 2020-05-12 | 乐普(北京)医疗器械股份有限公司 | Method and device for detecting P wave and T wave in electrocardiosignal |
CN114403892A (en) * | 2022-01-24 | 2022-04-29 | 中山大学 | Method, device and equipment for interactively adjusting respiratory feedback based on heart rate variability |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090133064A1 (en) * | 2007-11-19 | 2009-05-21 | Tetsuo Maruyama | Information providing device, information acquisition terminal, broadcast receiving terminal, information providing system, information providing method, and program |
KR20100128023A (en) * | 2009-05-27 | 2010-12-07 | 세종대학교산학협력단 | The emotion recognition system based on biometric signals |
CN104354698A (en) * | 2014-12-12 | 2015-02-18 | 山东交通学院 | System and method for distinguishing stress of driver based on electrocardiosignals and geographic information |
CN104605841A (en) * | 2014-12-09 | 2015-05-13 | 电子科技大学 | Wearable electrocardiosignal monitoring device and method |
CN106963363A (en) * | 2017-03-28 | 2017-07-21 | 西安交通大学 | A kind of mobile phone shell of integrated electrocardio and skin pyroelectric monitor |
-
2019
- 2019-01-31 CN CN201910099443.3A patent/CN109770920A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090133064A1 (en) * | 2007-11-19 | 2009-05-21 | Tetsuo Maruyama | Information providing device, information acquisition terminal, broadcast receiving terminal, information providing system, information providing method, and program |
KR20100128023A (en) * | 2009-05-27 | 2010-12-07 | 세종대학교산학협력단 | The emotion recognition system based on biometric signals |
CN104605841A (en) * | 2014-12-09 | 2015-05-13 | 电子科技大学 | Wearable electrocardiosignal monitoring device and method |
CN104354698A (en) * | 2014-12-12 | 2015-02-18 | 山东交通学院 | System and method for distinguishing stress of driver based on electrocardiosignals and geographic information |
CN106963363A (en) * | 2017-03-28 | 2017-07-21 | 西安交通大学 | A kind of mobile phone shell of integrated electrocardio and skin pyroelectric monitor |
Non-Patent Citations (4)
Title |
---|
S. L. PINGALE: "Using Pan Tompkin’s method, Ecg signal processing and", 《PROCEEDINGS OF IRF INTERNATIONAL CONFERENCE》 * |
WILLIS J. TOMPKINS著,林家瑞,徐邦荃译: "《生物医学数字信号处》", 31 December 2001 * |
华晶: "可穿戴式***心电信号压缩感知与心律失常分类研究", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
荀立晨: "基于心电信号的情感识别研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110353704A (en) * | 2019-07-12 | 2019-10-22 | 东南大学 | Mood assessments method and apparatus based on wearable ECG monitoring |
CN110353704B (en) * | 2019-07-12 | 2022-02-01 | 东南大学 | Emotion evaluation method and device based on wearable electrocardiogram monitoring |
CN111134659A (en) * | 2020-01-17 | 2020-05-12 | 乐普(北京)医疗器械股份有限公司 | Method and device for detecting P wave and T wave in electrocardiosignal |
CN114403892A (en) * | 2022-01-24 | 2022-04-29 | 中山大学 | Method, device and equipment for interactively adjusting respiratory feedback based on heart rate variability |
CN114403892B (en) * | 2022-01-24 | 2024-04-02 | 中山大学 | Method, device and equipment for interactively adjusting respiratory feedback based on heart rate variability |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhai et al. | Automated ECG classification using dual heartbeat coupling based on convolutional neural network | |
Oweis et al. | QRS detection and heart rate variability analysis: A survey | |
Yang et al. | Arrhythmia recognition and classification using combined parametric and visual pattern features of ECG morphology | |
Chua et al. | Application of higher order statistics/spectra in biomedical signals—A review | |
CN105796096B (en) | A kind of heart rate variance analyzing method, system and terminal | |
CN105147248A (en) | Physiological information-based depressive disorder evaluation system and evaluation method thereof | |
Balli et al. | Classification of biological signals using linear and nonlinear features | |
CN204931634U (en) | Based on the depression evaluating system of physiologic information | |
CN105147281A (en) | Portable stimulating, awaking and evaluating system for disturbance of consciousness | |
Kelwade et al. | Radial basis function neural network for prediction of cardiac arrhythmias based on heart rate time series | |
CN110353704B (en) | Emotion evaluation method and device based on wearable electrocardiogram monitoring | |
CN109770920A (en) | Intense strain method of discrimination and its system based on wearable ECG signal | |
Kelwade et al. | Prediction of cardiac arrhythmia using artificial neural network | |
Saminu et al. | Wavelet feature extraction for ECG beat classification | |
Thilagavathy et al. | Real-time ECG signal feature extraction and classification using support vector machine | |
CN113288168A (en) | Wearable fatigue monitoring of intelligence and early warning system | |
Mahesh et al. | ECG arrhythmia classification based on logistic model tree | |
Kamaruddin et al. | Early prediction of Cardiovascular Diseases using ECG signal | |
Dogan et al. | A comprehensive review of computer-based Techniques for R-peaks/QRS complex detection in ECG signal | |
Hugeng et al. | Development of the ‘Healthcor’system as a cardiac disorders symptoms detector using an expert system based on arduino uno | |
Pandit et al. | An efficient abnormal beat detection scheme from ECG signals using neural network and ensemble classifiers | |
Sanamdikar et al. | Extraction of different features of ECG signal for detection of cardiac arrhythmias by using wavelet transformation Db 6 | |
Li et al. | A novel single-lead handheld atrial fibrillation detection system | |
Khandait et al. | Efficient ECG abnormalities recognition using neuro-fuzzy approach | |
Zhi et al. | Wavelet based machine learning techniques for electrocardiogram signal analysis |
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 |
Application publication date: 20190521 |
|
RJ01 | Rejection of invention patent application after publication |