CN105726013A - Electrocardiogram monitoring system with electrocardiosignal quality discrimination function - Google Patents
Electrocardiogram monitoring system with electrocardiosignal quality discrimination function Download PDFInfo
- Publication number
- CN105726013A CN105726013A CN201610056148.6A CN201610056148A CN105726013A CN 105726013 A CN105726013 A CN 105726013A CN 201610056148 A CN201610056148 A CN 201610056148A CN 105726013 A CN105726013 A CN 105726013A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- module
- cardioelectric monitor
- section
- signal
- 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
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
-
- 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
- A61B5/7221—Determining signal validity, reliability or quality
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an electrocardiogram monitoring system with an electrocardiosignal quality discrimination function. The electrocardiogram monitoring system comprises an electrocardiogram monitor and an intelligent terminal. The electrocardiogram monitor is in wireless communication with the intelligent terminal. The intelligent terminal internally comprises a processor and a Bluetooth communication module connected with the processor. The processor comprises a signal collection module, a feature extraction module, a neural network training module and a discrimination module. An independent single-channel electrocardiosignal is converted into three feature values, namely, the QRS energy specific value, the signal kurtosis and the base line energy specific value, before learning through the technological means of the power spectral density quadrature and the kurtosis coefficient, and therefore a discrimination model is accurately established through the artificial neural network learning algorithm in a gradient descent optimizing mode according to the feature values, the discrimination method of the electrocardiosignal quality is achieved by reducing a system model, and therefore whether the electrocardiosignal can be used for diagnosis or not is effectively discriminated.
Description
Technical field
The invention belongs to technical field of medical equipment, be specifically related to a kind of cardioelectric monitor system with electrocardiosignal quality discrimination function.
Background technology
Electrocardiosignal usually can be subject to serious noise and illusion interference, and filtering algorithm many times fine can not must remove these interference, especially because interference signal often has similar frequency content and close form with electrocardiosignal.Therefore interference can reduce electrocardiosignal quality, and affects based on cardiac electrical automatic medical diagnosis on disease thus causing more spurious alarm (false positive) situation.The too poor meeting of such as electrocardiosignal causes substantial amounts of spurious alarm in ICU, and the spurious alarm in ICU is possibly even up to 86%.
Along with gradually stepping up of mankind's life expectancy, modern society's successful aging will become whole world emphasis.World Health Organization (WHO) estimates that more than the 60 years old population in the year two thousand fifty whole world is up to 2,000,000,000, and the old people of 80% will live in low income and middle income country.Cardiovascular disease chronic diseases will become a very big burden and has a strong impact on the quality of life of old people.Portable medical and wearable armarium will be more widely deployed for prevention and the management of chronic disease.Following portable medical will provide for high-quality, at a low price, convenient health management scheme.Such as Holter can be conveniently used when not affecting daily life because of it, and is widely used in the cardiac monitoring of long-range.The dynamic monitoring function of Holter makes it can reach the continuous monitoring even up to a couple of days in 24 hours in Clinical practice.Modern means of communication make electrocardiosignal very convenient must store and are close to real-time must be transferred on the computer of medical personnel or on other mobile terminals.But, the Holter signal under dynamically can be subject to disturbing than the illusion of more serious noise, by the Holter signal of noise pollution, even artificial medical diagnosis on disease automatically is caused very big difficulty.
By in real time electrocardiosignal quality being estimated, user can know that the quality of acquired electrocardiosignal is how immediately.If signal quality is too poor, user will be reminded, and remeasure signal, or check the connection of electrode, or check other noise effects being likely to result in.
The research about electrocardiosignal quality assessment techniques can be found from some documents or patented technology in the recent period, but these researchs are all based on feature point extraction, namely first pass through the characteristic point of a kind of algorithm extraction electrocardiosignal, it is such as in most cases R wave point or QRS complex, then passes through the analysis etc. to the analysis of feature point extraction result, QRS wave shape and obtain the description to signal quality；But one of important factor in order of feature extraction result is signal quality level, feature extraction can be caused error by signal quality problem, by feature extraction result, signal quality level judges also to bring error, these errors all accumulate backward, ultimately cause the inaccuracy analyzing result or even mistake.
It addition, signal quality level is not simply good and bad difference, having obvious fuzzy quality, this meets the mankind cognitive style for signal quality itself.For a specific signal of example, it is likely to, containing the multiple factor affecting quality, how it to be referred in a kind increasingly similar with it, and conventional processing mode is difficult to accomplish this point.
Summary of the invention
Above-mentioned technical problem existing for prior art, the invention provides a kind of cardioelectric monitor system with electrocardiosignal quality discrimination function, can effectively judge whether electrocardiosignal quality meets acceptable requirement, greatly reduce the diagnostic result mistake brought owing to signal quality is relatively low.
A kind of cardioelectric monitor system with electrocardiosignal quality discrimination function, including cardioelectric monitor device and intelligent terminal, described cardioelectric monitor device is connected by wireless telecommunications with intelligent terminal；
Described cardioelectric monitor device includes monitor main body and several electrocardiogram acquisition electrodes, is provided with main control module, Signal-regulated kinase and bluetooth communication module in described monitor main body；Electrocardiogram acquisition electrode is connected with Signal-regulated kinase, and Signal-regulated kinase is connected with main control module, and main control module is connected with bluetooth communication module；Described electrocardiogram acquisition electrode, for picking up the faint electrocardiosignal of people's body surface, is sent into Signal-regulated kinase after amplifying Filtering Processing, main control module is carried out sampling and Digital Signal Processing, then pass through bluetooth communication module and electrocardiosignal is transferred to intelligent terminal；
Described intelligent terminal includes bluetooth communication module and processor, and described processor includes:
Signal acquisition module, for collecting, by the bluetooth communication module in intelligent terminal, the electrocardiogram (ECG) data that cardioelectric monitor device provides；Described electrocardiogram (ECG) data be cardioelectric monitor device when collecting the m group electrocardiosignal section based on independent leads passage or user's routine testing in advance cardioelectric monitor device collect single pass electrocardiosignal section；By being manually marked as, 0 or 1,1 expression is satisfied accepts requirement to the quality of described m group electrocardiosignal section, and 0 foot with thumb down accepts requirement, and m is the natural number more than 1；
Characteristic extracting module, namely by calculating, the QRS energy ratio of this electrocardiosignal section, signal kurtosis and baseline energy ratio are obtained for arbitrary electrocardiosignal section for described electrocardiogram (ECG) data being carried out feature extraction, and these three characteristic index is formed the characteristic sequence of this electrocardiosignal section；
Neural metwork training module, for being trained by artificial neural network learning algorithm according to the m stack features sequence obtained of extracting corresponding to above-mentioned m group electrocardiosignal section, obtains the discrimination model about electrocardiosignal quality；
Discrimination module, characteristic sequence corresponding to the single channel electrocardiosignal section that obtained by user's routine testing substitutes into and obtains the corresponding output result about electrocardiosignal quality in above-mentioned discrimination model, and then judges whether the quality of this single channel electrocardiosignal section meets according to this output result and accept requirement.
Described characteristic extracting module calculates the QRS energy ratio of electrocardiosignal section by following formula:
Wherein: P (f) is the power spectral density function of electrocardiosignal section, S is the QRS energy ratio of electrocardiosignal section, and f is frequency.
Described characteristic extracting module calculates the signal kurtosis of electrocardiosignal section by following formula:
Wherein: X (i) is the ith sample value in electrocardiosignal section, N is the sampled point number in electrocardiosignal section, and μ is the average sample value of electrocardiosignal section, and K is the signal kurtosis of electrocardiosignal section.
Described characteristic extracting module calculates the baseline energy ratio of electrocardiosignal section by following formula:
Wherein: P (f) is the power spectral density function of electrocardiosignal section, B is the baseline energy ratio of electrocardiosignal section, and f is frequency.
The artificial neural network learning algorithm that described neural metwork training module adopts is using gradient descent method as optimizing direction.
The detailed process that described neural metwork training module is trained by artificial neural network learning algorithm is as follows:
(1) it is divided into training set and test set and training set more than test set m stack features sequence；
(2) one neutral net being made up of input layer, hidden layer and output layer of structure is initialized；
(3) appoint from training set and take a characteristic sequence and substitute into above-mentioned neural computing and obtain the corresponding output result about electrocardiosignal quality, calculate the cumulative error between the handmarking's mass corresponding to this output result and this characteristic sequence；
(4) by gradient descent method, weight between input layer and hidden layer and between hidden layer and output layer in neutral net is modified according to this cumulative error, and then appoints from training set and take off a characteristic sequence and substitute into revised neutral net；
(5) travel through all characteristic sequences in training set according to step (3) and (4), take cumulative error minimum time corresponding neutral net be discrimination model.
In the neutral net that described neural metwork training module initialization builds, input layer is made up of 3 neurons, and hidden layer is made up of 4 neurons, and output layer is made up of 1 neuron.
In the neutral net that described neural metwork training module initialization builds, the expression formula of neuron function g (z) is as follows:
Wherein: z is argument of function.
The discrimination model that described neural metwork training module obtains for training, characteristic sequence in test set is substituted into one by one this discrimination model and obtains the corresponding output result about electrocardiosignal quality, the output result corresponding to each characteristic sequence is made to compare with handmarking's mass, if the accuracy of test set is be more than or equal to threshold value, then this discrimination model is finally determined；If the accuracy of test set is less than threshold value, then utilize the more electrocardiosignal section sample of cardioelectric monitor device collection, obtain greater number of characteristic sequence through characteristic extracting module and be trained as the input of neutral net.
Described intelligent terminal can be smart mobile phone, panel computer or PC.
Cardioelectric monitor system of the present invention is quadratured by power spectral density and independent single pass electrocardiosignal is converted into QRS energy ratio, signal kurtosis and three eigenvalues of baseline energy ratio by the technological means of coefficient of kurtosis before study, and then utilize the optimal way that artificial neural network learning algorithm declines with gradient to be set up accurately by discrimination model according to eigenvalue, by the reduction to system model, achieve the method for discrimination of electrocardiosignal quality, and then effective whether electrocardiosignal can be used for carrying out diagnosis be made that examination.
Accompanying drawing explanation
Fig. 1 is the structural representation of cardioelectric monitor system of the present invention.
Fig. 2 is the structural representation of intelligent terminal in cardioelectric monitor system of the present invention.
Fig. 3 is the artificial nerve network model schematic diagram in electrocardiosignal quality discrimination process of the present invention.
Detailed description of the invention
In order to more specifically describe the present invention, below in conjunction with the drawings and the specific embodiments, technical scheme is described in detail.
As it is shown in figure 1, the present invention has the cardioelectric monitor system of electrocardiosignal quality discrimination function, including cardioelectric monitor device and smart mobile phone；Wherein:
Cardioelectric monitor device includes monitor main body and multiple electrocardiogram acquisition electrode, is provided with main control module, voltage detection module, bluetooth communication module, Signal-regulated kinase, automatic shutdown module, power management module and driving module in monitor main body；Monitor body surfaces is provided with shift knob and low pressure display lamp 1；Wherein:
Power management module is for providing running voltage for other functional modules in electrocardiogram acquisition electrode and monitor main body.
Electrocardiogram acquisition electrode is connected with Signal-regulated kinase, and it is for picking up the faint electrocardiosignal of people's body surface.
Signal-regulated kinase is connected with main control module, and it is amplified sending main control module to after filtering etc. processes for the faint electrocardiosignal that electrocardiogram acquisition electrode is picked up；In present embodiment, Signal-regulated kinase is sequentially connected with forms by inputting buffer stage, preposition instrument amplifier stage, high pass filter, interstage amplifier section, low pass filter and power frequency notch filter.
Driving module to be connected with main control module and shift knob, it is for driving power management module to electrocardiogram acquisition electrode discharge by main control module, and user can pass through shift knob and start cardioelectric monitor device.
Voltage detection module is connected with power management module and low pressure display lamp 1, and it is for detecting the information of voltage of power management module；The running voltage provided for cardioelectric monitor device when power management module is less than in a preset value situation, and low pressure display lamp 1 is lighted, to point out user cardioelectric monitor device is charged or changes battery.
Automatic shutdown module is connected with power management module and main control module, and it can make cardioelectric monitor device in long-time idle situation, is cut off by the power supply of cardioelectric monitor device, enters resting state, reduces power consumption；In present embodiment, a timer it is provided with in automatic shutdown module, timer is connected with main control module, timer is set with certain time interval (10s), this interval is exceeded when main control module does not have electrocardiosignal, automatic shutdown module automatically by dump, will enter resting state, reduce power consumption.
Bluetooth communication module is connected with main control module, and electrocardiosignal is radioed to smart mobile phone by bluetooth communication module by main control module.In present embodiment, bluetooth communication module follows bluetooth standard protocol；Module supports the interface such as UART, USB, SPI, PCM, SPDIF, and supports SPP bluetooth serial ports agreement, has that cost is low, volume is little, low in energy consumption, transmitting-receiving susceptiveness advantages of higher, and only need to be equipped with fraction of peripheral cell can be achieved with its power.
As in figure 2 it is shown, include processor and bluetooth communication module in present embodiment in smart mobile phone, bluetooth communication module is connected with processor；Processor includes signal acquisition module, characteristic extracting module, neural metwork training module and discrimination module；Wherein:
Signal acquisition module collects, by bluetooth communication module, the electrocardiogram (ECG) data that cardioelectric monitor device provides；Electrocardiogram (ECG) data be cardioelectric monitor device when collecting the m group electrocardiosignal section based on independent leads passage or user's routine testing in advance cardioelectric monitor device collect single pass electrocardiosignal section；By being manually marked as, 0 or 1,1 expression is satisfied accepts requirement to the quality of m group electrocardiosignal section, and 0 foot with thumb down accepts requirement, and m is the natural number more than 1.Present embodiment is derived from CinCChallenge2011 (hereinafter referred to as CinC) for the data base of Algorithm Analysis.CinC data base comprises the 12 medical records of channel standard of 1000 10 seconds durations.In medical 12 passages of standard, only 8 passages are independent.Therefore, 8 passages is chosen in each record by us, is respectively as follows: passage I, II, V1, V2, V3, V4, V5, V6.The data base thus constituted comprises 8000 single channel electrocardiographic recordings.Each single channel recording is then through by manually passing judgment on, being labeled as acceptable and unacceptable two kinds according to its signal quality.
Characteristic extracting module carries out feature extraction for the electrocardiogram (ECG) data that signal acquisition module collection is obtained and namely obtains the QRS energy ratio of this electrocardiosignal section, signal kurtosis and baseline energy ratio for arbitrary electrocardiosignal section by calculating, and these three characteristic index forms the characteristic sequence of this electrocardiosignal section；Circular is as follows:
(1) QRS energy ratio is calculated；This characterizing definition is the ratio of QRS wave shape energy and the energy of electrocardiosignal.
First electrocardiosignal is done spectrum analysis, then calculate the energy of 5-15Hz frequency range and the relative ratio of 5-40Hz band energy.Wherein 5-15Hz corresponds roughly to the energy of QRS wave shape, and 5-40Hz is about as much as the energy that electrocardiosignal is overall；As shown by the following formula:
The energy of QRS wave is concentrated mainly in the frequency bandwidth of 10Hz and center width is 10Hz, and when occurring that myoelectricity disturbs, the radio-frequency component in signal can increase, then energy ratio will reduce；And when there is the dislocation of electrode of a class QRS wave, then energy ratio can dramatically increase.
(2) kurtosis is calculated；Kurtosis is also called coefficient of kurtosis, also becomes quadravalence standard square in statistics.Kurtosis is used for characterizing distribution curve at meansigma methods place peak value height.Represent one section of electrocardiosignal with X, represent the average of signal with μ, represent the standard variance of signal with σ；Below equation is used for seeking kurtosis:
One clean its kurtosis of intact electrocardiosignal more than 5, and if exist myoelectricity interference or, baseline drift, Hz noise or Gauss distribution random noise, its kurtosis will be less than 5.
(3) baseline energy ratio is calculated；This characterizing definition is the business between the energy of 1-40Hz frequency range and 0-40Hz band energy.
The frequency of baseline drift is about 0.15-0.3Hz, so can effectively be characterized whether there is the baseline interference that impact is bigger by baseline energy ratio.
Neural metwork training module, for being trained by artificial neural network learning algorithm according to the m stack features sequence obtained of extracting corresponding to above-mentioned m group electrocardiosignal section, obtains the discrimination model about electrocardiosignal quality；Implement process as follows:
(1) single pass electrocardiosignal training sample set is divided into training set and test set；
(2) neural network model is set up according to artificial neural network learning algorithm: neural network model has input layer, hidden layer and output layer three layers, the input and output of input layer are three-channel correlation coefficient, it is attached by formula (4) between layers, the neuron activation functions of hidden layer and output layer is formula (5), output layer is made up of 1 neuron, hidden layer is made up of 4 neurons, simultaneously by the weights coefficient initialization of each interlayer；Fig. 3 is the artificial nerve network model set up.
(3) one group of sample in the training set of electrocardio training sample is input to the neutral net under current weight coefficient, calculates the output of each node of input layer, hidden layer and output layer successively.
(4) the cumulative error E between output layer output and the expected result of electrocardio training sample of all electrocardio training samples is calculated according to formula (6)train, according to gradient descent method, revise hidden layer and each internodal weights coefficient of output layer with formula (7), revise input layer and each internodal weights coefficient of hidden layer with formula (8).
Wherein: E is cumulative error,Export through the kth of the output layer of neutral net for single training sample, okFor the kth expected result of single training sample, m is training set total sample number, and p is output layer output sum；
Wherein: whoT () is the weights coefficient that the t time sample is input to during neutral net between hidden layer and output layer,For the output through the output layer of neutral net of the single training sample, o is the expected result of single training sample, xhFor the output of hidden layer, α is learning rate；
Wherein: wihT () is the weights coefficient that the t time sample is input to during neutral net between input layer and hidden layer, xiFor the output of input layer, α is learning rate；
(5) travel through the training set of all electrocardiosignal training samples with step (3) and step (4), then get EtrainWeights coefficient sets time minimum, and test with test set neutral net, if the accuracy of test is higher than threshold value, train；If it is not, increase electrocardiosignal training sample, and repeat step (3)～(5).In present embodiment, learning rate α=0.05.
Discrimination module substitutes into for the characteristic sequence corresponding to single channel electrocardiosignal section obtained by user's routine testing and obtains the corresponding output result about electrocardiosignal quality in above-mentioned discrimination model.The i.e. weights proportion according to each layer of neutral net, the system function of reduction electrocardiosignal quality evaluation discrimination model, by system function to whether electrocardiosignal quality discrimination accepts: first, calculate its QRS energy ratio, kurtosis and baseline energy ratio respectively according to the electrocardiosignal section of a certain autonomous channel；Then, these three eigenvalue substitution discrimination model will obtain the corresponding output result about link position state；Finally, determine the electrocardiosignal quality of this autonomous channel according to this output result whether to can accept.
When differentiating that in the acceptable situation of electrocardiosignal quality, ecg signal data can be further processed by processor, processing mode includes ECG Signal Analysis diagnosis, ecg signal data storage and ecg signal data remote transmission.Wherein, ecg signal data remote transmission is, by wireless network (wifi or 3G/4G), electrocardiogram (ECG) data is uploaded to cloud server, in order to the server dump to individual's excavation of ecg signal data and data.
The above-mentioned description to embodiment is to be understood that for ease of those skilled in the art and apply the present invention.Above-described embodiment obviously easily can be made various amendment by person skilled in the art, and General Principle described herein is applied in other embodiments without through performing creative labour.Therefore, the invention is not restricted to above-described embodiment, those skilled in the art's announcement according to the present invention, the improvement made for the present invention and amendment all should within protection scope of the present invention.
Claims (10)
1. having a cardioelectric monitor system for electrocardiosignal quality discrimination function, including cardioelectric monitor device and intelligent terminal, described cardioelectric monitor device is connected by wireless telecommunications with intelligent terminal；It is characterized in that:
Described cardioelectric monitor device includes monitor main body and several electrocardiogram acquisition electrodes, is provided with main control module, Signal-regulated kinase and bluetooth communication module in described monitor main body；Electrocardiogram acquisition electrode is connected with Signal-regulated kinase, and Signal-regulated kinase is connected with main control module, and main control module is connected with bluetooth communication module；Described electrocardiogram acquisition electrode, for picking up the faint electrocardiosignal of people's body surface, is sent into Signal-regulated kinase after amplifying Filtering Processing, main control module is carried out sampling and Digital Signal Processing, then pass through bluetooth communication module and electrocardiosignal is transferred to intelligent terminal；
Described intelligent terminal includes bluetooth communication module and processor, and described processor includes:
Signal acquisition module, for collecting, by the bluetooth communication module in intelligent terminal, the electrocardiogram (ECG) data that cardioelectric monitor device provides；Described electrocardiogram (ECG) data be cardioelectric monitor device when collecting the m group electrocardiosignal section based on independent leads passage or user's routine testing in advance cardioelectric monitor device collect single pass electrocardiosignal section；By being manually marked as, 0 or 1,1 expression is satisfied accepts requirement to the quality of described m group electrocardiosignal section, and 0 foot with thumb down accepts requirement, and m is the natural number more than 1；
Characteristic extracting module, namely by calculating, the QRS energy ratio of this electrocardiosignal section, signal kurtosis and baseline energy ratio are obtained for arbitrary electrocardiosignal section for described electrocardiogram (ECG) data being carried out feature extraction, and these three characteristic index is formed the characteristic sequence of this electrocardiosignal section；
Neural metwork training module, for being trained by artificial neural network learning algorithm according to the m stack features sequence obtained of extracting corresponding to above-mentioned m group electrocardiosignal section, obtains the discrimination model about electrocardiosignal quality；
Discrimination module, characteristic sequence corresponding to the single channel electrocardiosignal section that obtained by user's routine testing substitutes into and obtains the corresponding output result about electrocardiosignal quality in above-mentioned discrimination model, and then judges whether the quality of this single channel electrocardiosignal section meets according to this output result and accept requirement.
2. cardioelectric monitor system according to claim 1, it is characterised in that: described characteristic extracting module calculates the QRS energy ratio of electrocardiosignal section by following formula:
Wherein: P (f) is the power spectral density function of electrocardiosignal section, S is the QRS energy ratio of electrocardiosignal section, and f is frequency.
3. cardioelectric monitor system according to claim 1, it is characterised in that: described characteristic extracting module calculates the signal kurtosis of electrocardiosignal section by following formula:
Wherein: X (i) is the ith sample value in electrocardiosignal section, N is the sampled point number in electrocardiosignal section, and μ is the average sample value of electrocardiosignal section, and K is the signal kurtosis of electrocardiosignal section.
4. cardioelectric monitor system according to claim 1, it is characterised in that: described characteristic extracting module calculates the baseline energy ratio of electrocardiosignal section by following formula:
Wherein: P (f) is the power spectral density function of electrocardiosignal section, B is the baseline energy ratio of electrocardiosignal section, and f is frequency.
5. cardioelectric monitor system according to claim 1, it is characterised in that: the artificial neural network learning algorithm that described neural metwork training module adopts is using gradient descent method as optimizing direction.
6. cardioelectric monitor system according to claim 1, it is characterised in that: the detailed process that described neural metwork training module is trained by artificial neural network learning algorithm is as follows:
(1) it is divided into training set and test set and training set more than test set m stack features sequence；
(2) one neutral net being made up of input layer, hidden layer and output layer of structure is initialized；
(3) appoint from training set and take a characteristic sequence and substitute into above-mentioned neural computing and obtain the corresponding output result about electrocardiosignal quality, calculate the cumulative error between the handmarking's mass corresponding to this output result and this characteristic sequence；
(4) by gradient descent method, weight between input layer and hidden layer and between hidden layer and output layer in neutral net is modified according to this cumulative error, and then appoints from training set and take off a characteristic sequence and substitute into revised neutral net；
(5) travel through all characteristic sequences in training set according to step (3) and (4), take cumulative error minimum time corresponding neutral net be discrimination model.
7. cardioelectric monitor system according to claim 6, it is characterised in that: in the neutral net that described neural metwork training module initialization builds, input layer is made up of 3 neurons, and hidden layer is made up of 4 neurons, and output layer is made up of 1 neuron.
8. cardioelectric monitor system according to claim 6, it is characterised in that: in the neutral net that described neural metwork training module initialization builds, the expression formula of neuron function g (z) is as follows:
Wherein: z is argument of function.
9. cardioelectric monitor system according to claim 6, it is characterized in that: the discrimination model that described neural metwork training module obtains for training, characteristic sequence in test set is substituted into one by one this discrimination model and obtains the corresponding output result about electrocardiosignal quality, the output result corresponding to each characteristic sequence is made to compare with handmarking's mass, if the accuracy of test set is be more than or equal to threshold value, then this discrimination model is finally determined；If the accuracy of test set is less than threshold value, then utilize the more electrocardiosignal section sample of cardioelectric monitor device collection, obtain greater number of characteristic sequence through characteristic extracting module and be trained as the input of neutral net.
10. cardioelectric monitor system according to claim 1, it is characterised in that: described intelligent terminal is smart mobile phone, panel computer or PC.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610056148.6A CN105726013A (en) | 2016-01-27 | 2016-01-27 | Electrocardiogram monitoring system with electrocardiosignal quality discrimination function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610056148.6A CN105726013A (en) | 2016-01-27 | 2016-01-27 | Electrocardiogram monitoring system with electrocardiosignal quality discrimination function |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105726013A true CN105726013A (en) | 2016-07-06 |
Family
ID=56247768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610056148.6A Pending CN105726013A (en) | 2016-01-27 | 2016-01-27 | Electrocardiogram monitoring system with electrocardiosignal quality discrimination function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105726013A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106691437A (en) * | 2017-01-26 | 2017-05-24 | 浙江铭众科技有限公司 | Fetal heart rate extraction method based on maternal electrocardiosignals |
| CN106889981A (en) * | 2017-01-26 | 2017-06-27 | 浙江铭众科技有限公司 | A kind of intelligent terminal for extracting fetal heart frequency |
| CN107296600A (en) * | 2017-05-17 | 2017-10-27 | 哈尔滨工业大学（威海） | A kind of quick 12 lead electrocardiosignal method for evaluating quality |
| CN107753012A (en) * | 2016-08-19 | 2018-03-06 | 中国科学院上海微***与信息技术研究所 | A kind of mcg-signalses method for evaluating quality, system and server |
| CN108158578A (en) * | 2017-12-14 | 2018-06-15 | 北京顺源开华科技有限公司 | Noise segments recognition methods, ECG signal processing method and processing device |
| CN108399369A (en) * | 2018-02-02 | 2018-08-14 | 东南大学 | Electrocardio beat sorting technique based on Distributed Calculation and deep learning |
| CN108470158A (en) * | 2018-03-08 | 2018-08-31 | 华南理工大学 | A method of it finding error minimal network for dynamic ECG data and calculates structure |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5827196A (en) * | 1997-07-15 | 1998-10-27 | Hewlett-Packard Company | Method and system for providing characterizations of waveform representations of heart function |
| CN101467879A (en) * | 2007-12-26 | 2009-07-01 | 深圳迈瑞生物医疗电子股份有限公司 | Method and device for processing multi-lead synchronized electrocardiosignal |
| CN102512158A (en) * | 2011-12-31 | 2012-06-27 | 济南汇医融工科技有限公司 | Electrocardiosignal quality evaluation method and device based on high-dimensional fuzzy recognition |
-
2016
- 2016-01-27 CN CN201610056148.6A patent/CN105726013A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5827196A (en) * | 1997-07-15 | 1998-10-27 | Hewlett-Packard Company | Method and system for providing characterizations of waveform representations of heart function |
| CN101467879A (en) * | 2007-12-26 | 2009-07-01 | 深圳迈瑞生物医疗电子股份有限公司 | Method and device for processing multi-lead synchronized electrocardiosignal |
| CN102512158A (en) * | 2011-12-31 | 2012-06-27 | 济南汇医融工科技有限公司 | Electrocardiosignal quality evaluation method and device based on high-dimensional fuzzy recognition |
Non-Patent Citations (1)
| Title |
|---|
| 寇明春: "心电信号质量评估方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107753012A (en) * | 2016-08-19 | 2018-03-06 | 中国科学院上海微***与信息技术研究所 | A kind of mcg-signalses method for evaluating quality, system and server |
| CN106691437A (en) * | 2017-01-26 | 2017-05-24 | 浙江铭众科技有限公司 | Fetal heart rate extraction method based on maternal electrocardiosignals |
| CN106889981A (en) * | 2017-01-26 | 2017-06-27 | 浙江铭众科技有限公司 | A kind of intelligent terminal for extracting fetal heart frequency |
| CN107296600A (en) * | 2017-05-17 | 2017-10-27 | 哈尔滨工业大学（威海） | A kind of quick 12 lead electrocardiosignal method for evaluating quality |
| CN108158578A (en) * | 2017-12-14 | 2018-06-15 | 北京顺源开华科技有限公司 | Noise segments recognition methods, ECG signal processing method and processing device |
| CN108399369A (en) * | 2018-02-02 | 2018-08-14 | 东南大学 | Electrocardio beat sorting technique based on Distributed Calculation and deep learning |
| CN108399369B (en) * | 2018-02-02 | 2021-10-19 | 东南大学 | Electrocardio beat classification method based on distributed computation and deep learning |
| CN108470158A (en) * | 2018-03-08 | 2018-08-31 | 华南理工大学 | A method of it finding error minimal network for dynamic ECG data and calculates structure |
| CN108470158B (en) * | 2018-03-08 | 2020-05-12 | 华南理工大学 | Method for searching error minimum network computing structure for dynamic ECG data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105726013A (en) | Electrocardiogram monitoring system with electrocardiosignal quality discrimination function | |
| CN105595990A (en) | Intelligent terminal device for evaluating and distinguishing quality of electrocardiosignal | |
| CN107951485B (en) | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study | |
| US20200367800A1 (en) | Method for identifying driving fatigue based on cnn-lstm deep learning model | |
| US11783924B2 (en) | ECG information processing method and ECG workstation | |
| CN108836323B (en) | Learning state monitoring system based on electroencephalogram analysis and using method thereof | |
| EP3698709A1 (en) | Electrocardiogram information processing method and electrocardiogram workstation system | |
| CN111544015B (en) | Cognitive power-based control work efficiency analysis method, device and system | |
| CN105496392B (en) | A kind of have electrode and connect the three of discrimination function and lead cardioelectric monitor system | |
| CN105725966A (en) | Electrocardiosignal quality discrimination method based on neural network model | |
| CN111466876B (en) | Alzheimer's disease auxiliary diagnosis system based on fNIRS and graph neural network | |
| CN106419864A (en) | Cloud expert system for TCM diagnosis | |
| CN109009102B (en) | Electroencephalogram deep learning-based auxiliary diagnosis method and system | |
| CN111598453B (en) | Control work efficiency analysis method, device and system based on execution force in virtual scene | |
| CN111598451A (en) | Control work efficiency analysis method, device and system based on task execution capacity | |
| CN106236027A (en) | Depressed crowd's decision method that a kind of brain electricity combines with temperature | |
| CN106166065A (en) | A kind of wearable electrocardio health interacting platform based on social networks and its implementation | |
| CN106667436A (en) | Sleep diagnosis method and system | |
| CN114886388A (en) | Evaluation method and device for quality of electroencephalogram signal in anesthesia depth monitoring process | |
| CN105943022B (en) | A kind of cardioelectric monitor system that there are three leads to reconstruct 12 lead function | |
| CN208677390U (en) | A kind of brain wave acquisition instrument | |
| CN113907709A (en) | Portable sleep monitoring system based on ear EEG | |
| CN109431499B (en) | Botanic person home care auxiliary system and auxiliary method | |
| CN113208629A (en) | Alzheimer disease screening method and system based on EEG signal | |
| Chang | Research on prediction algorithm of employees’ psychological stress based on multifeature fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160706 |