CN108030488A - The detecting system of arrhythmia cordis based on convolutional neural networks - Google Patents
The detecting system of arrhythmia cordis based on convolutional neural networks Download PDFInfo
- Publication number
- CN108030488A CN108030488A CN201711238517.4A CN201711238517A CN108030488A CN 108030488 A CN108030488 A CN 108030488A CN 201711238517 A CN201711238517 A CN 201711238517A CN 108030488 A CN108030488 A CN 108030488A
- Authority
- CN
- China
- Prior art keywords
- ecg
- convolutional neural
- electrocardiogram
- data
- neural networks
- 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
- A61B5/7235—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
- A61B2576/00—Medical imaging apparatus involving image processing or analysis
- A61B2576/02—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part
- A61B2576/023—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part for the heart
Abstract
The present invention provides a kind of detecting system of the arrhythmia cordis based on convolutional neural networks, including:Segmentation module, the K leads electrocardiogram (ECG) data for the first patient to acquisition carry out segment processing in chronological order, obtain multiple K leads electrocardiogram (ECG) data sections, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;Detection module, for the multiple K leads electrocardiogram (ECG) data section to be inputted trained convolutional neural networks model respectively successively according to the sequencing of time, obtains the type that first patient suffers from arrhythmia cordis.The detecting system of arrhythmia cordis provided by the invention based on convolutional neural networks, by the way that convolutional neural networks are combined with electrocardiogram (ECG) data, the feature extraction and classifying of arrhythmia cordis is permeated step, fully excavate the multi-lead and timing information of electrocardiogram, highly reliable prediction is made to case, so as to improve the accuracy of arrhythmia detection.
Description
Technical field
The present invention relates to nerual network technique field, more particularly to a kind of inspection of the arrhythmia cordis based on convolutional neural networks
Examining system.
Background technology
With the rapid development and extensive use of computer technology, the health of the mankind has been given play to more in computer-aided diagnosis
Carry out more important effect.
In the prior art, the method by detection of the computer-aided diagnosis system to arrhythmia cordis is as follows:First, according to
Sample of users data, are trained statistical model using support vector machines (Support Vector Machine, SVM) algorithm
Study;Then, electrocardiogram (ECG) data of the patient within a sampling period is gathered, analyzes and extracts the multinomial characteristic in electrocardiogram (ECG) data
According to average value and variance of each characteristic within the sampling period is calculated;By average value, variance and multinomial characteristic
Combination, obtains corresponding first multi-C vector of patient;By the first multi-C vector and predetermined user's types of arrhythmia
Statistical model is matched, and determines the types of arrhythmia of the patient.
Since in the detecting system of the prior art, study is trained to statistical model using algorithm of support vector machine, and
SVM is commonly used in the classification of two class problems, bad for multicategory classification problem effect, and types of arrhythmia up to more than ten is planted, and is led
Cause the analysis result to arrhythmia cordis inaccurate.
The content of the invention
(1) technical problems to be solved
The object of the present invention is to provide a kind of detecting system of the arrhythmia cordis based on convolutional neural networks, solve existing
The technical problem of the testing result inaccuracy of arrhythmia detection system in technology.
(2) technical solution
In order to solve the above-mentioned technical problem, on the one hand, the present invention provides a kind of arrhythmia cordis based on convolutional neural networks
Detecting system, including:
Segmentation module, the K leads electrocardiogram (ECG) data for the first patient to acquisition carry out segment processing, obtain in chronological order
Multiple K leads electrocardiogram (ECG) data sections are taken, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
Detection module, for the multiple K leads electrocardiogram (ECG) data section to be inputted respectively successively according to the sequencing of time
Trained convolutional neural networks model, obtains the type that first patient suffers from arrhythmia cordis.
Further, further include:
Training module, for obtaining training sample set, the training sample set includes multigroup training sample, every group of training sample
This includes multiple K leads electrocardiogram (ECG) data sections of a sample patient;
Based on the training sample set, the convolutional neural networks model is trained.
Further, the convolutional neural networks model includes:Input unit, convolutional neural networks unit, time domain average
Unit, linear classification unit and output unit, the convolutional neural networks unit include N number of subelement, and each subelement includes M
A convolutional layer, M and N are positive integer.
Further, further include:
Spectrum Conversion module, for carrying out Spectrum Conversion to each K leads electrocardiogram (ECG) data section respectively, obtains multiple through overfrequency
The K lead electrocardiogram (ECG) data sections of spectral transformation;
Correspondingly, the detection module, for by the multiple K leads electrocardiogram (ECG) data section according to the time sequencing according to
Secondary to input trained convolutional neural networks model respectively, obtaining the type that first patient suffers from arrhythmia cordis is specially:
Detection module, for will be the multiple suitable according to the priority of time by the K lead electrocardiogram (ECG) data sections of Spectrum Conversion
Sequence inputs trained convolutional neural networks model respectively successively, obtains the type that first patient suffers from arrhythmia cordis.
Further, further include:
Filter module, the K lead electrocardiogram (ECG) datas for the first patient to acquisition are filtered.
Further, further include:
Data augmentation module, for carrying out data augmentation to the training sample set;
Correspondingly, it is described to be based on the training sample set, the convolutional neural networks model is trained specially:
Based on the training sample set after data augmentation, the convolutional neural networks model is trained.
On the other hand, the present invention provides a kind of electronic equipment for arrhythmia detection, including:
Memory and processor, the processor and the memory complete mutual communication by bus;It is described to deposit
Reservoir is stored with the programmed instruction that can be performed by the processor, and it is as follows that the processor calls described program instruction to be able to carry out
Step:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
Another further aspect, the present invention provide a kind of computer program product, and the computer program product includes being stored in non-
Computer program in transitory computer readable storage medium, the computer program include programmed instruction, when described program refers to
When order is computer-executed, the computer is set to perform following steps:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
Another aspect, the present invention provide a kind of computer-readable recording medium, are stored thereon with computer program, the meter
Calculation machine program realizes following steps when being executed by processor:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
(3) beneficial effect
The detecting system of arrhythmia cordis provided by the invention based on convolutional neural networks, by by convolutional neural networks with
Electrocardiogram (ECG) data is combined, and the feature extraction and classifying of arrhythmia cordis is permeated step, electrocardiogram is fully excavated more and leads
Connection and timing information, make case highly reliable prediction, so as to improve the accuracy of arrhythmia detection.
Brief description of the drawings
Fig. 1 is the detecting system schematic diagram of the arrhythmia cordis based on convolutional neural networks according to the embodiment of the present invention;
Fig. 2 is the detecting system signal of the arrhythmia cordis based on convolutional neural networks according to another embodiment of the present invention
Figure;
Fig. 3 is the schematic diagram of the convolutional neural networks model according to the embodiment of the present invention;
Fig. 4 is the schematic diagram of the convolutional neural networks subelement according to the embodiment of the present invention;
Fig. 5 is the structure diagram of the electronic equipment provided in an embodiment of the present invention for arrhythmia detection.
Embodiment
In order to make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, implement below in conjunction with the present invention
Attached drawing in example, is clearly and completely described the technical solution in the embodiment of the present invention, it is clear that described embodiment
It is part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiment of the present invention, those of ordinary skill in the art
All other embodiments obtained without making creative work, belong to the scope of protection of the invention.
Embodiment 1:
Fig. 1 is the detecting system schematic diagram of the arrhythmia cordis based on convolutional neural networks according to the embodiment of the present invention, such as
Shown in Fig. 1, the embodiment of the present invention provides a kind of detecting system of the arrhythmia cordis based on convolutional neural networks, including segmentation module
10 and detection module 20, wherein,
Segmentation module 10 is used to carry out segment processing in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition,
Multiple K leads electrocardiogram (ECG) data sections are obtained, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
Detection module 20 is used to according to the sequencing of time input the multiple K leads electrocardiogram (ECG) data section respectively successively
Trained convolutional neural networks model, obtains the type that first patient suffers from arrhythmia cordis.
Further, further include:
Training module, for obtaining training sample set, the training sample set includes multigroup training sample, every group of training sample
This includes multiple K leads electrocardiogram (ECG) data sections of a sample patient;
Based on the training sample set, the convolutional neural networks model is trained.
Further, the convolutional neural networks model includes:Input unit, convolutional neural networks unit, time domain average
Unit, linear classification unit and output unit, the convolutional neural networks unit include N number of subelement, and each subelement includes M
A convolutional layer, M and N are positive integer.
Further, further include:
Spectrum Conversion module, for carrying out Spectrum Conversion to each K leads electrocardiogram (ECG) data section respectively, obtains multiple through overfrequency
The K lead electrocardiogram (ECG) data sections of spectral transformation;
Correspondingly, the detection module, for by the multiple K leads electrocardiogram (ECG) data section according to the time sequencing according to
Secondary to input trained convolutional neural networks model respectively, obtaining the type that first patient suffers from arrhythmia cordis is specially:
Detection module, for will be the multiple suitable according to the priority of time by the K lead electrocardiogram (ECG) data sections of Spectrum Conversion
Sequence inputs trained convolutional neural networks model respectively successively, obtains the type that first patient suffers from arrhythmia cordis.
Further, further include:
Filter module, the K lead electrocardiogram (ECG) datas for the first patient to acquisition are filtered.
Further, further include:
Data augmentation module, for carrying out data augmentation to the training sample set;
Correspondingly, it is described to be based on the training sample set, the convolutional neural networks model is trained specially:
Based on the training sample set after data augmentation, the convolutional neural networks model is trained.
Specifically, the detecting system of the arrhythmia cordis provided in an embodiment of the present invention based on convolutional neural networks is big including two
Part, Part I:Segmentation module 10;Part II:Detection module 20.
, it is necessary to pass through the training module in system before being detected using convolutional neural networks model to arrhythmia cordis
Convolutional neural networks are trained, it is specific as follows:
First, by training module, the K lead electrocardio numbers of when small (such as 24) were obtained in a period of time of multiple sample patients
According to K is positive integer, can select the quantity of lead in practical application as the case may be.
Due in gatherer process electrocardiogram (ECG) data be easily subject to it is various interference (such as electrostatic interference, Hz noise, high frequency are done
Disturb, myoelectricity interference etc.) influence, so needing that it is carried out to filter out interference processing, arrhythmia cordis is examined with reducing interference to the greatest extent
The influence of survey, by the filter module in system, carries out filtering out interference processing (such as filtering out baseline to do to K leads electrocardiogram (ECG) data
Disturb), obtain filtering out the K lead electrocardiogram (ECG) datas after interference.
Suitable length (such as 10 seconds) is chosen, is in chronological order split into the K lead electrocardiogram (ECG) datas filtered out after disturbing more
A K leads electrocardiogram (ECG) data section.Multiple K leads electrocardiogram (ECG) data sections of each sample patient just constitute one group of training sample, multigroup
Training sample just constitutes training sample set, and every group of training sample is the K lead electrocardiogram (ECG) datas to sample patient by medical practitioners
Mark obtains after being analyzed, the type of the arrhythmia cordis of mark, including nodal tachycardia, sinus bradycardia, Dou Xingxin
Restrain uneven, sinus arrest, sino atrial block, escape beat, escape rhythm, atrial premature beats, ventricular extrasystole, auricular flutter, the heart
Atrial fibrillation moves, ventricular flutter, ventricular fibrillation, sinoatrial block, intra-auricular conductional block, atrioventricular block, intraventricular conduction
Retardance etc..
In addition, in order to improve training quality, it is necessary to carry out data augmentation processing to training sample set, i.e., by system
Multiple data segments that training sample is concentrated are carried out the processing such as time migration, the instruction for concentrating training sample by data augmentation module
Practice sample to increase, based on the training sample set after data augmentation, the convolutional neural networks model is trained, so that
Achieve the purpose that to improve training quality.
In order to further improve training quality, training sample can also be concentrated by the Spectrum Conversion module in system
Each data segment carry out Spectrum Conversion (such as Short Time Fourier Transform), obtain higher-dimension electrocardiogram (ECG) data, thus reach raising training
The purpose of quality.Whether need to carry out Spectrum Conversion in practical application, can be selected according to actual conditions.
Including filter module 30, data augmentation module 40, Spectrum Conversion module 50 and training module 60 based on convolution god
The detecting system of arrhythmia cordis through network is as shown in Figure 2.
Operated more than, complete the collection of training sample set.
Then, based on the training sample set, the convolutional neural networks model is trained.
After being trained completion to convolutional neural networks, trained convolutional neural networks model is obtained, i.e. the rhythm of the heart loses
Normal detection model, by the detection module 20 in system, loses the rhythm of the heart using obtained trained convolutional neural networks model
Often it is detected, it is specific as follows:
First, the K lead electrocardiogram (ECG) datas of when small (such as 24) in a period of time of the first patient are obtained, then by filtering mould
Block carries out K leads electrocardiogram (ECG) data to filter out interference processing (such as filtering out baseline interference), obtains filtering out the K lead electrocardio numbers after interference
According to.By the segmentation module 10 in system, the K leads after disturbing will be filtered out in chronological order by choosing suitable length (such as 10 seconds)
Electrocardiogram (ECG) data splits into multiple K leads electrocardiogram (ECG) data sections.
Then, it is by the detection module 20 in system, the multiple K leads electrocardiogram (ECG) data section is suitable according to the priority of time
Sequence inputs trained convolutional neural networks model respectively successively, obtains the type that first patient suffers from arrhythmia cordis.
Wherein, the convolutional neural networks model includes:Input module, convolutional neural networks module, time domain average module,
Linear classification module and output module, the convolutional neural networks module include N number of submodule, and each submodule includes M volume
Lamination, M and N are positive integer.
Fig. 3 is the schematic diagram of the convolutional neural networks model according to the embodiment of the present invention, as shown in figure 3, convolutional Neural net
Network model includes:Input unit 100, convolutional neural networks unit 200, time domain average unit 300,400 and of linear classification unit
Output unit 500, wherein, the convolutional neural networks unit 200 includes N number of subelement, and each subelement includes M convolution
Layer, M and N are positive integer.The concrete structure of the convolutional neural networks unit 200 can be according to actual need in practical applications
Determine.
Fig. 4 is the schematic diagram of the convolutional neural networks subelement according to the embodiment of the present invention, as shown in figure 4, the convolution
Neutral net subelement includes, convolutional layer A, down-sampling layer A, convolutional layer B, down-sampling layer B, convolutional layer C, down-sampling layer C, finally
Two layers is to carry out classification output using full articulamentum.
8 lead electrocardiogram without Spectrum Conversion are regarded as two dimensional image, input layer dimension is 8*1900;Wherein,
A height of the 8 of the two-dimentional electrocardiogram picture of input, width 1900;Convolutional layer A, shares 8 characteristic faces, and each characteristic face uses a 3*
18 convolution kernels, wherein, a height of the 3 of convolution kernel, width 18, this layer exports the characteristic face of 8 6*1883 sizes;Down-sampling layer A is adopted
Take 1*7 to sample core, produce the characteristic face of 8 6*269;Convolutional layer B uses the core of 3*12, produces the characteristic face of 14 4*258;Under
Sample level B uses the core of 1*6, produces the characteristic face of 14 4*43;Convolutional layer C uses the convolution kernel of 3*8,20 2*36's of generation
Characteristic face;Down-sampling layer C equally uses the core of 1*6, produces the characteristic face of 20 2*6;Last two layers is to use multilayer perceptron
Carry out classification output.
Each input feature vector face (equivalent to 1 characteristic face of input layer) is mapped to multiple defeated by each different convolution kernels
Go out characteristic face, and form each neuron and the same local area phase in each input feature vector face in same output characteristic face
Even, and weights are shared, but the neuron weights in different output characteristic faces are not shared;Then each output characteristic face is grasped by scaling
Make (interval is maximized or is spaced and is averaged) down-sampling and reduce size.
The detecting system of arrhythmia cordis provided by the invention based on convolutional neural networks, by by convolutional neural networks with
Electrocardiogram (ECG) data is combined, and the feature extraction and classifying of arrhythmia cordis is permeated step, electrocardiogram is fully excavated more and leads
Connection and timing information, make case highly reliable prediction, so as to improve the accuracy of arrhythmia detection.
Embodiment 2:
Fig. 5 is the structure diagram of the electronic equipment provided in an embodiment of the present invention for arrhythmia detection, such as Fig. 5 institutes
Show, the equipment includes:Processor (processor) 801, memory (memory) 802 and bus 803;
Wherein, processor 801 and memory 802 complete mutual communication by the bus 803;
Processor 801 is used to call the programmed instruction in memory 802, to perform following steps:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
Embodiment 3:
The embodiment of the present invention discloses a kind of computer program product, and the computer program product includes being stored in non-transient
Computer program on computer-readable recording medium, the computer program include programmed instruction, when described program instructs quilt
When computer performs, computer is able to carry out following steps:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
Embodiment 4:
The embodiment of the present invention provides a kind of non-transient computer readable storage medium storing program for executing, the non-transient computer readable storage
Medium storing computer instructs, and the computer instruction makes the computer perform following steps:
Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads hearts
Electric data segment, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;
The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolution god respectively successively according to the sequencing of time
Through network model, the type that first patient suffers from arrhythmia cordis is obtained.
One of ordinary skill in the art will appreciate that:The embodiment such as device described above and equipment is only schematic
, wherein the unit illustrated as separating component may or may not be physically separate, shown as unit
The component shown may or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
In network unit.Some or all of module therein can be selected to realize the mesh of this embodiment scheme according to the actual needs
's.Those of ordinary skill in the art are not in the case where paying performing creative labour, you can to understand and implement.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that:It still may be used
To modify to the technical solution described in foregoing embodiments, or equivalent substitution is carried out to which part technical characteristic;
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical solution spirit and
Scope.
Claims (9)
- A kind of 1. detecting system of the arrhythmia cordis based on convolutional neural networks, it is characterised in that including:Segmentation module, the K leads electrocardiogram (ECG) data for the first patient to acquisition carry out segment processing in chronological order, obtain more A K leads electrocardiogram (ECG) data section, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;Detection module, for the sequencing input training respectively successively by the multiple K leads electrocardiogram (ECG) data section according to the time Good convolutional neural networks model, obtains the type that first patient suffers from arrhythmia cordis.
- 2. system according to claim 1, it is characterised in that further include:Training module, for obtaining training sample set, the training sample set includes multigroup training sample, every group of training sample bag Multiple K leads electrocardiogram (ECG) data sections containing a sample patient;Based on the training sample set, the convolutional neural networks model is trained.
- 3. system according to claim 1, it is characterised in that the convolutional neural networks model includes:Input unit, volume Product neutral net unit, time domain average unit, linear classification unit and output unit, the convolutional neural networks unit include N A subelement, each subelement include M convolutional layer, and M and N are positive integer.
- 4. system according to claim 1, it is characterised in that further include:Spectrum Conversion module, for carrying out Spectrum Conversion to each K leads electrocardiogram (ECG) data section respectively, obtains multiple by frequency spectrum change The K lead electrocardiogram (ECG) data sections changed;Correspondingly, the detection module, for the multiple K leads electrocardiogram (ECG) data section to be divided successively according to the sequencing of time Trained convolutional neural networks model is not inputted, and obtaining the type that first patient suffers from arrhythmia cordis is specially:Detection module, for by the multiple K lead electrocardiogram (ECG) data sections by Spectrum Conversion according to the time sequencing according to It is secondary to input trained convolutional neural networks model respectively, obtain the type that first patient suffers from arrhythmia cordis.
- 5. system according to claim 1, it is characterised in that further include:Filter module, the K lead electrocardiogram (ECG) datas for the first patient to acquisition are filtered.
- 6. system according to claim 2, it is characterised in that further include:Data augmentation module, for carrying out data augmentation to the training sample set;Correspondingly, it is described to be based on the training sample set, the convolutional neural networks model is trained specially:Based on the training sample set after data augmentation, the convolutional neural networks model is trained.
- A kind of 7. electronic equipment for arrhythmia detection, it is characterised in that including:Memory and processor, the processor and the memory complete mutual communication by bus;The memory The programmed instruction that can be performed by the processor is stored with, the processor calls described program instruction to be able to carry out walking as follows Suddenly:Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads electrocardio numbers According to section, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolutional Neural net respectively successively according to the sequencing of time Network model, obtains the type that first patient suffers from arrhythmia cordis.
- 8. a kind of computer program product, it is characterised in that the computer program product includes being stored in non-transient computer Computer program on readable storage medium storing program for executing, the computer program include programmed instruction, when described program is instructed by computer During execution, the computer is set to perform following steps:Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads electrocardio numbers According to section, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolutional Neural net respectively successively according to the sequencing of time Network model, obtains the type that first patient suffers from arrhythmia cordis.
- 9. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the computer program quilt Processor realizes following steps when performing:Segment processing is carried out in chronological order to the K leads electrocardiogram (ECG) data of the first patient of acquisition, obtains multiple K leads electrocardio numbers According to section, the equal length of each K leads electrocardiogram (ECG) data section, K is positive integer;The multiple K leads electrocardiogram (ECG) data section is inputted into trained convolutional Neural net respectively successively according to the sequencing of time Network model, obtains the type that first patient suffers from arrhythmia cordis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711238517.4A CN108030488A (en) | 2017-11-30 | 2017-11-30 | The detecting system of arrhythmia cordis based on convolutional neural networks |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711238517.4A CN108030488A (en) | 2017-11-30 | 2017-11-30 | The detecting system of arrhythmia cordis based on convolutional neural networks |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108030488A true CN108030488A (en) | 2018-05-15 |
Family
ID=62094298
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711238517.4A Pending CN108030488A (en) | 2017-11-30 | 2017-11-30 | The detecting system of arrhythmia cordis based on convolutional neural networks |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108030488A (en) |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108039203A (en) * | 2017-12-04 | 2018-05-15 | 北京医拍智能科技有限公司 | The detecting system of arrhythmia cordis based on deep neural network |
CN108647785A (en) * | 2018-05-17 | 2018-10-12 | 普强信息技术(北京)有限公司 | A kind of neural network method for automatic modeling, device and storage medium |
CN109243615A (en) * | 2018-10-12 | 2019-01-18 | 东软集团股份有限公司 | Determine the method, apparatus and storage medium of percussion information |
CN109480816A (en) * | 2018-12-18 | 2019-03-19 | 安徽华米信息科技有限公司 | Rhythm of the heart monitoring method, device, electronic equipment and computer readable storage medium |
CN109711544A (en) * | 2018-12-04 | 2019-05-03 | 北京市商汤科技开发有限公司 | Method, apparatus, electronic equipment and the computer storage medium of model compression |
CN110236520A (en) * | 2019-05-20 | 2019-09-17 | 上海数创医疗科技有限公司 | ECG type recognition methods and device based on double convolutional neural networks |
CN110464334A (en) * | 2019-07-23 | 2019-11-19 | 苏州国科视清医疗科技有限公司 | A kind of electrocardiographic abnormality detection method based on composite depth learning network |
WO2020047750A1 (en) * | 2018-09-04 | 2020-03-12 | 深圳先进技术研究院 | Arrhythmia detection method and apparatus, electronic device, and computer storage medium |
CN111067505A (en) * | 2019-12-25 | 2020-04-28 | 北京多唯阳光科技有限责任公司 | Arrhythmia detection method, device and storage medium |
CN111297327A (en) * | 2020-02-20 | 2020-06-19 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
CN111407261A (en) * | 2020-03-31 | 2020-07-14 | 京东方科技集团股份有限公司 | Method and device for measuring periodic information of biological signal and electronic equipment |
CN111588349A (en) * | 2020-05-28 | 2020-08-28 | 京东方科技集团股份有限公司 | Health analysis device and electronic equipment |
CN111723622A (en) * | 2019-03-22 | 2020-09-29 | 安徽华米信息科技有限公司 | Heart beat classification method and device, wearable device and storage medium |
WO2020226783A1 (en) * | 2019-05-06 | 2020-11-12 | Medtronic, Inc. | Category-based review and reporting of episode data |
CN112043260A (en) * | 2020-09-16 | 2020-12-08 | 杭州师范大学 | Electrocardiogram classification method based on local mode transformation |
CN112131907A (en) * | 2019-06-24 | 2020-12-25 | 华为技术有限公司 | Method and device for training classification model |
CN113040781A (en) * | 2021-03-11 | 2021-06-29 | 广州市康源图像智能研究院 | I-lead electrocardiogram data identification method and system |
US11355244B2 (en) | 2019-05-06 | 2022-06-07 | Medtronic, Inc | Machine learning based depolarization identification and arrhythmia localization visualization |
US11443852B2 (en) | 2019-05-06 | 2022-09-13 | Medtronic, Inc. | Reduced power machine learning system for arrhythmia detection |
US11475998B2 (en) | 2019-05-06 | 2022-10-18 | Medtronic, Inc. | Data preparation for artificial intelligence-based cardiac arrhythmia detection |
US11583687B2 (en) | 2019-05-06 | 2023-02-21 | Medtronic, Inc. | Selection of probability thresholds for generating cardiac arrhythmia notifications |
US11617533B2 (en) | 2019-05-06 | 2023-04-04 | Medtronic, Inc. | Visualization of arrhythmia detection by machine learning |
US11679268B2 (en) | 2018-10-05 | 2023-06-20 | Medtronic, Inc. | Multi-tier prediction of cardiac tachyarrythmia |
US11696718B2 (en) | 2019-05-06 | 2023-07-11 | Medtronic, Inc. | Arrhythmia detection with feature delineation and machine learning |
CN111723622B (en) * | 2019-03-22 | 2024-04-26 | 安徽华米信息科技有限公司 | Heart beat classification method, heart beat classification device, wearable equipment and storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102779234A (en) * | 2012-06-28 | 2012-11-14 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram classified processing method and device |
CN105748063A (en) * | 2016-04-25 | 2016-07-13 | 山东大学齐鲁医院 | Intelligent arrhythmia diagnosis method based on multiple-lead and convolutional neural network |
CN106202968A (en) * | 2016-07-28 | 2016-12-07 | 北京博源兴康科技有限公司 | The data analysing method of cancer and device |
CN106344005A (en) * | 2016-10-28 | 2017-01-25 | 张珈绮 | Mobile ECG (electrocardiogram) monitoring system and monitoring method |
CN106725426A (en) * | 2016-12-14 | 2017-05-31 | 深圳先进技术研究院 | A kind of method and system of electrocardiosignal classification |
CN106901723A (en) * | 2017-04-20 | 2017-06-30 | 济南浪潮高新科技投资发展有限公司 | A kind of electrocardiographic abnormality automatic diagnosis method |
CN107137072A (en) * | 2017-04-28 | 2017-09-08 | 北京科技大学 | A kind of ventricular ectopic beating detection method based on 1D convolutional neural networks |
CN107184198A (en) * | 2017-06-01 | 2017-09-22 | 广州城市职业学院 | A kind of electrocardiosignal classifying identification method |
CN107203692A (en) * | 2017-05-09 | 2017-09-26 | 哈尔滨工业大学(威海) | The implementation method of atrial fibrillation detection based on depth convolutional neural networks |
CN107516075A (en) * | 2017-08-03 | 2017-12-26 | 安徽华米信息科技有限公司 | Detection method, device and the electronic equipment of electrocardiosignal |
-
2017
- 2017-11-30 CN CN201711238517.4A patent/CN108030488A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102779234A (en) * | 2012-06-28 | 2012-11-14 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram classified processing method and device |
CN105748063A (en) * | 2016-04-25 | 2016-07-13 | 山东大学齐鲁医院 | Intelligent arrhythmia diagnosis method based on multiple-lead and convolutional neural network |
CN106202968A (en) * | 2016-07-28 | 2016-12-07 | 北京博源兴康科技有限公司 | The data analysing method of cancer and device |
CN106344005A (en) * | 2016-10-28 | 2017-01-25 | 张珈绮 | Mobile ECG (electrocardiogram) monitoring system and monitoring method |
CN106725426A (en) * | 2016-12-14 | 2017-05-31 | 深圳先进技术研究院 | A kind of method and system of electrocardiosignal classification |
CN106901723A (en) * | 2017-04-20 | 2017-06-30 | 济南浪潮高新科技投资发展有限公司 | A kind of electrocardiographic abnormality automatic diagnosis method |
CN107137072A (en) * | 2017-04-28 | 2017-09-08 | 北京科技大学 | A kind of ventricular ectopic beating detection method based on 1D convolutional neural networks |
CN107203692A (en) * | 2017-05-09 | 2017-09-26 | 哈尔滨工业大学(威海) | The implementation method of atrial fibrillation detection based on depth convolutional neural networks |
CN107184198A (en) * | 2017-06-01 | 2017-09-22 | 广州城市职业学院 | A kind of electrocardiosignal classifying identification method |
CN107516075A (en) * | 2017-08-03 | 2017-12-26 | 安徽华米信息科技有限公司 | Detection method, device and the electronic equipment of electrocardiosignal |
Non-Patent Citations (3)
Title |
---|
MAN_WORLD: "《参见https://blog.csdn.net/mzpmzk/article/details/78636184》", 26 November 2017 * |
徐其华: "《参见https://www.cnblogs.com/denny402/p/6933172.html》", 2 June 2017 * |
郑泽宇: "《参见https://zhuanlan.zhihu.com/p/30645524》", 1 November 2017 * |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108039203A (en) * | 2017-12-04 | 2018-05-15 | 北京医拍智能科技有限公司 | The detecting system of arrhythmia cordis based on deep neural network |
CN108647785A (en) * | 2018-05-17 | 2018-10-12 | 普强信息技术(北京)有限公司 | A kind of neural network method for automatic modeling, device and storage medium |
WO2020047750A1 (en) * | 2018-09-04 | 2020-03-12 | 深圳先进技术研究院 | Arrhythmia detection method and apparatus, electronic device, and computer storage medium |
US11679268B2 (en) | 2018-10-05 | 2023-06-20 | Medtronic, Inc. | Multi-tier prediction of cardiac tachyarrythmia |
CN109243615A (en) * | 2018-10-12 | 2019-01-18 | 东软集团股份有限公司 | Determine the method, apparatus and storage medium of percussion information |
CN109711544A (en) * | 2018-12-04 | 2019-05-03 | 北京市商汤科技开发有限公司 | Method, apparatus, electronic equipment and the computer storage medium of model compression |
CN109480816A (en) * | 2018-12-18 | 2019-03-19 | 安徽华米信息科技有限公司 | Rhythm of the heart monitoring method, device, electronic equipment and computer readable storage medium |
CN111723622A (en) * | 2019-03-22 | 2020-09-29 | 安徽华米信息科技有限公司 | Heart beat classification method and device, wearable device and storage medium |
CN111723622B (en) * | 2019-03-22 | 2024-04-26 | 安徽华米信息科技有限公司 | Heart beat classification method, heart beat classification device, wearable equipment and storage medium |
US11696718B2 (en) | 2019-05-06 | 2023-07-11 | Medtronic, Inc. | Arrhythmia detection with feature delineation and machine learning |
US11475998B2 (en) | 2019-05-06 | 2022-10-18 | Medtronic, Inc. | Data preparation for artificial intelligence-based cardiac arrhythmia detection |
US11694804B2 (en) | 2019-05-06 | 2023-07-04 | Medtronic, Inc. | Reduced power machine learning system for arrhythmia detection |
US11723577B2 (en) | 2019-05-06 | 2023-08-15 | Medtronic, Inc. | Visualization of arrhythmia detection by machine learning |
WO2020226783A1 (en) * | 2019-05-06 | 2020-11-12 | Medtronic, Inc. | Category-based review and reporting of episode data |
US11776691B2 (en) | 2019-05-06 | 2023-10-03 | Medtronic, Inc. | Machine learning based depolarization identification and arrhythmia localization visualization |
US11617533B2 (en) | 2019-05-06 | 2023-04-04 | Medtronic, Inc. | Visualization of arrhythmia detection by machine learning |
US11583687B2 (en) | 2019-05-06 | 2023-02-21 | Medtronic, Inc. | Selection of probability thresholds for generating cardiac arrhythmia notifications |
US11355244B2 (en) | 2019-05-06 | 2022-06-07 | Medtronic, Inc | Machine learning based depolarization identification and arrhythmia localization visualization |
US11443852B2 (en) | 2019-05-06 | 2022-09-13 | Medtronic, Inc. | Reduced power machine learning system for arrhythmia detection |
CN110236520B (en) * | 2019-05-20 | 2020-09-29 | 上海数创医疗科技有限公司 | Electrocardiogram type recognition device based on double convolution neural network |
CN110236520A (en) * | 2019-05-20 | 2019-09-17 | 上海数创医疗科技有限公司 | ECG type recognition methods and device based on double convolutional neural networks |
CN112131907A (en) * | 2019-06-24 | 2020-12-25 | 华为技术有限公司 | Method and device for training classification model |
WO2020258913A1 (en) * | 2019-06-24 | 2020-12-30 | 华为技术有限公司 | Method and device for training classification model |
CN110464334A (en) * | 2019-07-23 | 2019-11-19 | 苏州国科视清医疗科技有限公司 | A kind of electrocardiographic abnormality detection method based on composite depth learning network |
CN111067505A (en) * | 2019-12-25 | 2020-04-28 | 北京多唯阳光科技有限责任公司 | Arrhythmia detection method, device and storage medium |
CN111297327A (en) * | 2020-02-20 | 2020-06-19 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
CN111297327B (en) * | 2020-02-20 | 2023-12-01 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
CN111407261A (en) * | 2020-03-31 | 2020-07-14 | 京东方科技集团股份有限公司 | Method and device for measuring periodic information of biological signal and electronic equipment |
CN111588349A (en) * | 2020-05-28 | 2020-08-28 | 京东方科技集团股份有限公司 | Health analysis device and electronic equipment |
CN111588349B (en) * | 2020-05-28 | 2023-12-01 | 京东方科技集团股份有限公司 | Health analysis device and electronic equipment |
CN112043260A (en) * | 2020-09-16 | 2020-12-08 | 杭州师范大学 | Electrocardiogram classification method based on local mode transformation |
CN113040781A (en) * | 2021-03-11 | 2021-06-29 | 广州市康源图像智能研究院 | I-lead electrocardiogram data identification method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108030488A (en) | The detecting system of arrhythmia cordis based on convolutional neural networks | |
US11344243B2 (en) | Artificial intelligence self-learning-based static electrocardiography analysis method and apparatus | |
Kumari et al. | An automated detection of heart arrhythmias using machine learning technique: SVM | |
US11529103B2 (en) | Artificial intelligence self-learning-based automatic electrocardiography analysis method and apparatus | |
CN108039203A (en) | The detecting system of arrhythmia cordis based on deep neural network | |
EP3692904A1 (en) | Method and device for self-learning dynamic electrocardiography analysis employing artificial intelligence | |
US9131864B2 (en) | System and method for evaluating an electrophysiological signal | |
US11617528B2 (en) | Systems and methods for reduced lead electrocardiogram diagnosis using deep neural networks and rule-based systems | |
CN109411041A (en) | Ecg information processing method and electro cardio signal workstation system | |
CN106725426A (en) | A kind of method and system of electrocardiosignal classification | |
WO2019161611A1 (en) | Ecg information processing method and ecg workstation | |
CN111000551A (en) | Heart disease risk diagnosis method based on deep convolutional neural network model | |
Farag | A self-contained STFT CNN for ECG classification and arrhythmia detection at the edge | |
CN109948396A (en) | A kind of beat classification method, beat classification device and electronic equipment | |
WO2021071646A1 (en) | Systems and methods for electrocardiogram diagnosis using deep neural networks and rule-based systems | |
CN107480637A (en) | Heart failure based on heart sound feature method by stages | |
Zeng et al. | Deep arrhythmia classification based on SENet and lightweight context transform | |
Yuan et al. | Diagnosing cardiac abnormalities from 12-lead electrocardiograms using enhanced deep convolutional neural networks | |
CN109044338A (en) | atrial fibrillation detection device and storage medium | |
CN111612042B (en) | Detection device for left front branch conduction block, storage medium and terminal | |
Ni | Enhanced IoT-Based Electrocardiogram Monitoring System with Deep Learning | |
Srinivasulu et al. | Novel method to find the parameter for noise removal from multi-channel ecg waveforms | |
Mukherjee et al. | A noise independent approach to QRS-complex detection using CWT | |
Mousa | Optimizing electrocardiogram analysis for efficient heart condition diagnosis | |
Begum et al. | Automatic Diagnostic System for Long-Term ECG Data from Holter Monitor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180515 |