CN106778685A - Electrocardiogram image-recognizing method, device and service terminal - Google Patents
Electrocardiogram image-recognizing method, device and service terminal Download PDFInfo
- Publication number
- CN106778685A CN106778685A CN201710023150.8A CN201710023150A CN106778685A CN 106778685 A CN106778685 A CN 106778685A CN 201710023150 A CN201710023150 A CN 201710023150A CN 106778685 A CN106778685 A CN 106778685A
- Authority
- CN
- China
- Prior art keywords
- electrocardiogram
- image
- deep learning
- learning model
- module
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000013136 deep learning model Methods 0.000 claims abstract description 108
- 238000000605 extraction Methods 0.000 claims abstract description 26
- 239000012634 fragment Substances 0.000 claims abstract description 15
- 238000003909 pattern recognition Methods 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 28
- 230000015654 memory Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 14
- 238000013135 deep learning Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 11
- 201000010099 disease Diseases 0.000 description 6
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 6
- 241000894007 species Species 0.000 description 6
- 238000003860 storage Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 210000004556 brain Anatomy 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 206010003671 Atrioventricular Block Diseases 0.000 description 1
- 208000001871 Tachycardia Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 206010003119 arrhythmia Diseases 0.000 description 1
- 230000006793 arrhythmia Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005194 fractionation Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006794 tachycardia Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Physics & Mathematics (AREA)
- Pathology (AREA)
- Signal Processing (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to ecg analysis technical field, specifically related to a kind of electrocardiogram image-recognizing method, device and service terminal, the electrocardiogram image-recognizing method is applied to service terminal, and the service terminal includes electrocardiogram deep learning model, while the service terminal is used for the input of electrocardiogram image.The electrocardiogram image-recognizing method includes for the electrocardiogram image of input being divided into multiple images fragment, feature extraction is carried out to the multiple image segments using the electrocardiogram deep learning model, according to each image segments in the characteristic use electrocardiogram the multiple image segments of deep learning Model Identification extracted obtaining the electrocardiogram image information of input, and then realize the automatic identification of the electrocardiogram image of input, data volume is have accumulated, while reducing the workload of staff.
Description
Technical field
The present invention relates to ecg analysis technical field, in particular to a kind of electrocardiogram image-recognizing method, device
And service terminal.
Background technology
Deep learning is a new field in machine learning research, and its motivation is to set up, simulate human brain and be analyzed
Study.Deep learning is a kind of data-driven model, can simulate human brain vision mechanism automatically learn it is at all levels to data
Abstract characteristics.
With the fast development of Medical Imaging Technology, medical image analysis step into the big data epoch, and the medical image is main
Refer to electrocardiogram, in the prior art, mainly by being manually analyzed to different types of electrocardiogram and then the species that diagnoses the illness, work
Measure very greatly, while being not easy to accumulate data.Therefore, substantial amounts of electrocardiogram image is divided using the method for deep learning
Analysis, can not only accumulate mass data, the workload of staff be alleviated, while having complied with the development in big data epoch.
The content of the invention
It is an object of the invention to provide a kind of electrocardiogram image-recognizing method, to realize accumulating data, independently to electrocardio
Figure image is analyzed, and mitigates the function of worker workload.
It is autonomous right to realize accumulating data another object of the present invention is to provide a kind of electrocardiogram pattern recognition device
Electrocardiogram image is analyzed, and mitigates the function of worker workload.
Another object of the present invention is to provide a kind of service terminal, to realize accumulating data, independently to electrocardiogram image
It is analyzed, mitigates the function of worker workload.
To achieve these goals, the technical scheme that the embodiment of the present invention is used is as follows：
In a first aspect, the embodiment of the invention provides a kind of electrocardiogram image-recognizing method, service terminal is applied to, it is described
Service terminal includes electrocardiogram deep learning model, and the service terminal is used for the input of electrocardiogram image, and methods described includes：
The electrocardiogram image of input is divided into multiple images fragment；
Feature extraction is carried out to the multiple image segments using the electrocardiogram deep learning model；
According to extract the feature, using in electrocardiogram the multiple image segments of deep learning Model Identification each
Image segments are obtaining the information of the electrocardiogram image of the input.
Further, it is described the electrocardiogram of input is divided into multiple images fragment before also include step：
Predetermined deep learning model；
According to the existing electrocardiogram image of multiple species, using deep learning method, the deep learning model is entered
Row training, obtains electrocardiogram deep learning model.
Further, it is described that the deep learning model is trained, obtain electrocardiogram deep learning model the step of
Including：
The existing electrocardiogram image is divided into multiple images data；
Using deep learning model, the feature of the multiple view data is extracted；
According to the feature of the multiple view data extracted, using the deep learning model, to the multiple image
Data are classified to obtain electrocardiogram deep learning model.
Further, it is described according to the feature extracted, using electrocardiogram the multiple figure of deep learning Model Identification
In photo section each image segments with the information of the electrocardiogram image for obtaining the input the step of include：
According to the feature extracted, the classification belonging to the feature is gone out using electrocardiogram deep learning Model Matching；
Classification according to belonging to the feature obtains the information of electrocardiogram image.
Second aspect, the embodiment of the present invention additionally provides a kind of electrocardiogram pattern recognition device, is applied to service terminal, institute
Stating service terminal includes electrocardiogram deep learning model, and the service terminal is used for the input of electrocardiogram image, described device bag
Include：
Module is split, for the electrocardiogram image of input to be divided into multiple images fragment；
First extraction module, for carrying out feature to the multiple image segments using the electrocardiogram deep learning model
Extract；
Identification module, for according to the feature extracted, using electrocardiogram the multiple figure of deep learning Model Identification
Each image segments is obtaining the information of the electrocardiogram image of the input in photo section.
Further, the electrocardiogram pattern recognition device also includes：
Presetting module, for predetermined deep learning model；
Training module, for the existing electrocardiogram image according to multiple species, using deep learning method, to the depth
Degree learning model is trained, and obtains electrocardiogram deep learning model.
Further, the training module includes：
Segmentation module, for the existing electrocardiogram image to be divided into multiple images data；
Second extraction module, for utilizing deep learning model, extracts the feature of the multiple view data；
Sort module is right using the deep learning model for the feature according to the multiple view data extracted
The multiple view data is classified to obtain electrocardiogram deep learning model.
Further, the identification module also includes：
Matching module, for according to the feature extracted, the feature being gone out using electrocardiogram deep learning Model Matching
Affiliated classification；
Acquisition module, the information of electrocardiogram image is obtained for the classification according to belonging to the feature.
The third aspect, the embodiment of the present invention additionally provides a kind of service terminal, and the service terminal includes electrocardiogram depth
Learning model, the service terminal is used for the input of electrocardiogram image, and the service terminal includes：
Memory；
Processor；And
Electrocardiogram pattern recognition device, the electrocardiogram pattern recognition device be installed on the memory and including one or
Multiple software function modules by the computing device, the electrocardiogram pattern recognition device includes：
Module is split, for the electrocardiogram image of input to be divided into multiple images fragment；
First extraction module, for carrying out feature to the multiple image segments using the electrocardiogram deep learning model
Extract；
Identification module, for according to the feature extracted, using electrocardiogram the multiple figure of deep learning Model Identification
Each image segments is obtaining the information of the electrocardiogram image of the input in photo section.
Further, the electrocardiogram pattern recognition device also includes presetting module, training module, and the presetting module is used
In predetermined deep learning model, the training module is used for the existing electrocardiogram image according to multiple species, using depth
Learning method, is trained to the deep learning model, obtains electrocardiogram deep learning model；The training module also includes dividing
Module, the second extraction module and sort module are cut, the segmentation module is used to for the existing electrocardiogram image to be divided into multiple
View data, second extraction module is used to utilize deep learning model, extracts the feature of the multiple view data, described
Sort module is used for the feature according to the multiple view data extracted, using the deep learning model, to the multiple
View data is classified to obtain electrocardiogram deep learning model, and the identification module also includes matching module and obtains not
Overstate, the matching module is used to, according to the feature extracted, the feature institute is gone out using electrocardiogram deep learning Model Matching
The classification of category, the classification that the acquisition module is used for according to belonging to the feature obtains the information of electrocardiogram image..
A kind of electrocardiogram image-recognizing method provided in an embodiment of the present invention, device and service terminal, the electrocardiogram image
Recognition methods is applied to service terminal, and the service terminal includes electrocardiogram deep learning model, while the service terminal is used for the heart
The input of electrograph image.The electrocardiogram image-recognizing method includes for the electrocardiogram image of input being divided into multiple images fragment, profit
Feature extraction is carried out to the multiple image segments with the electrocardiogram deep learning model, according to the characteristic use electrocardio for extracting
Each image segments to be to obtain the electrocardiogram image information of input in the multiple image segments of figure deep learning Model Identification,
And then the automatic identification of the electrocardiogram image of input is realized, data volume is have accumulated, while reducing the workload of staff.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
Technical scheme in order to illustrate more clearly the embodiments of the present invention, below will be attached to what is used needed for embodiment
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, thus be not construed as it is right
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 shows the structural representation of service terminal provided in an embodiment of the present invention.
Fig. 2 shows the flow chart of electrocardiogram image-recognizing method provided in an embodiment of the present invention.
Fig. 3 shows the flow chart of the sub-step of electrocardiogram image-recognizing method provided in an embodiment of the present invention.
Fig. 4 shows the flow chart of another sub-step of electrocardiogram image-recognizing method provided in an embodiment of the present invention.
Fig. 5 shows the block diagram of electrocardiogram pattern recognition device provided in an embodiment of the present invention.
Fig. 6 shows the square frame of the submodule of the training module of electrocardiogram pattern recognition device provided in an embodiment of the present invention
Schematic diagram.
Fig. 7 shows the square frame of the submodule of the identification module of electrocardiogram pattern recognition device provided in an embodiment of the present invention
Schematic diagram.
Diagram：100- service terminals；110- electrocardiogram pattern recognition devices；120- memories；130- processors；111- is pre-
If module；112- training modules；113- splits module；The extraction modules of 114- first；115- identification modules；1121- splits module；
The extraction modules of 1122- second；1123- sort modules；1151- matching modules；1152- acquisition modules.
Specific embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground description, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be arranged and designed with a variety of configurations herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of selected embodiment of the invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined in individual accompanying drawing, then it need not be further defined and explained in subsequent accompanying drawing.Meanwhile, of the invention
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that indicating or implying relative importance.
Fig. 1 is refer to, is the structural representation of service terminal provided in an embodiment of the present invention 100.The service terminal 100 can
To be, but desktop computer or panel computer are not limited to, electrocardiogram deep learning model is included in the service terminal 100, it is main to use
In the input for realizing electrocardiogram image, identification is analyzed to the electrocardiogram image being input into by electrocardiogram deep learning model,
And then the information of the electrocardiogram image being input into.The service terminal 100 includes electrocardiogram pattern recognition device 110, memory
120 and processor 130.
Memory 120, each element of processor 130 are directly or indirectly electrically connected with each other, to realize the biography of data
Defeated or interaction.For example, these elements can be realized being electrically connected with by one or more communication bus or holding wire each other.The heart
Electrograph pattern recognition device 110 can be stored in the memory including at least one in the form of software or firmware (firmware)
Software function module in 120 or in the operating system (operating system, OS) that is solidificated in service terminal 100.Place
Reason device 130 is used to perform the executable module stored in the memory 120, and such as electrocardiogram pattern recognition device 110 is wrapped
Software function module and computer program for including etc..
Wherein, memory 120 may be, but not limited to, random access memory (Random Access Memory,
RAM), read-only storage (Read Only Memory, ROM), programmable read only memory (Programmable Read-Only
Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only Memory, EPROM),
Electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only Memory, EEPROM) etc..
Wherein, memory 120 is used for storage program, and processor 130 performs the program after execute instruction is received.
Fig. 2 is refer to, is the flow chart of electrocardiogram image-recognizing method provided in an embodiment of the present invention, the electrocardiogram image
Recognition methods is applied to service terminal 100.The method includes：
Step S130, multiple images fragment is divided into by the electrocardiogram image of input.
Electrocardiogram image is the foundation of diagnosis patient disease, and several spies are divided into by by the electrocardiogram image of input
Fixed, with unique properties region, each region correspondence one image segments of correspondence.The image segments are one complete
Waveform, such as P ripples, P-R section, P-Q sections, be divided into multiple images fragment by by a complete electrocardiogram image, in order to more
The feature of image segments is obtained well, consequently facilitating preferably analyzing.
The multiple image segments are carried out feature extraction by step S140 using the electrocardiogram deep learning model.
Feature extraction is carried out to the multiple image segments using the electrocardiogram deep learning model, the image segments
More abstract high-level feature is characterized as, the high-level character representation is the feature of the corresponding waveform of the image segments, the ripple
Shape characteristic present kinds of Diseases, such as P-R sections of the duration can diagnose VPB, arrhythmia cordis, tachycardia.Specifically
For multiple hidden layers of the convolutional neural networks in electrocardiogram deep learning model enter the low-level image feature of the image segments of acquisition
Row multi-level combination, computing, to obtain abstract high-level feature.
Step S150, according to the feature extracted, using electrocardiogram the multiple image sheet of deep learning Model Identification
Each image segments is obtaining the information of the electrocardiogram image of the input in section.
Electrocardiogram deep learning model is identified analysis to each image segments, to obtain the letter of image segments sign
Breath.Specifically, the information characterized by the waveform in electrocardiogram deep learning model analysis each image segments, such as image
Waveform in fragment is P-R sections, and the P-R sections of cycle is more long, then characterizing may have the illnesss such as atrioventricular block.By to every
Waveform in one image segments is analyzed, determine each waveform may corresponding disease, and then by all image sheet segment tables
The information combination levied is got up the information of the electrocardiogram image being input into.By being input into electrocardiogram image to be detected in real time, lead to
Crossing electrocardiogram deep learning model carries out intellectual analysis, so that quick diagnosis go out the disease kind of the electrocardiogram characterization image of the input
Class, dramatically reduces the workload of staff, while by constantly accumulating data so that defeated to analyzing other next time
The electrocardiogram graphical analysis for entering is more convenient.
Before being analyzed to electrocardiogram image using electrocardiogram deep learning model, should also be to existing depth
Practise model to be trained, and then obtain the electrocardiogram deep learning model for meeting the ecg analysis, therefore,
Also included step S110 and S120 before step S130.
Step S110, predetermined deep learning model.
Deep learning model includes convolutional neural networks layer.Convolutional neural networks layer includes multiple hidden layers.The present embodiment
In, by pre-seting the parameter of each hidden layer, obtain the framework of deep learning model.And then, complete predetermined depth study mould
Type.Wherein, parameter can be, but not limited to be include the weighted value of each hidden layer, bias, structure, output layer node number,
The size of convolution kernel and the quantity of convolution kernel etc..
In other embodiments, ripe deep learning mould can also be obtained by initializing ripe deep learning model
The framework of type, and using the framework of ripe deep learning model as default deep learning model framework.Specifically, it is ripe
Deep learning model can be, but not limited to be, GoogleNet or VGG.
Step S120, according to the existing electrocardiogram image of multiple species, using deep learning method, to the depth
Practise model to be trained, obtain electrocardiogram deep learning model.
In the present embodiment, the existing electrocardiogram image according to substantial amounts of multiple types sets up training set.Using depth
Practise model each sample (i.e. existing electrocardiogram image) in training set is split and classified.Specifically, training
Cheng Zhong, deep learning model is also using the hyper parameter of hyper parameter automatic optimization method adjusting training process.It should be noted that super
The configuration of parameter directly affects the training effect of depth model, it is therefore desirable to find appropriate hyper parameter configuration to be trained
Best result can be obtained.In the present embodiment, hyper parameter includes hidden layer node number, activation primitive selection, unsupervised or have
Monitor procedure, the training factor, frequency of training and sparse coefficient etc..Specifically, hyper parameter automatic optimization method is according to current training
Result and training history search optimal hyper parameter configuration in hyper parameter space automatically.
Fig. 3 is refer to, is the flow chart of the sub-step of electrocardiogram image-recognizing method provided in an embodiment of the present invention.The step
Rapid S120 includes step S121, S122 and S123.
Step S121, multiple images data are divided into by the existing electrocardiogram image.
The existing electrocardiogram image is different types of existing electrocardiogram image, by multiple existing electrocardiogram images
It is divided into several regions specific, with unique properties, the view data after one segmentation of each region correspondence, the image
One complete waveform of data characterization, different waveforms characterize different diseases.Torn open by by existing electrocardiogram image
Point, in order to obtain more low-level image features, the low-level image feature may be, but not limited to, the ripple of the waveform in the view data
Length, wave amplitude etc., consequently facilitating the analysis of electrocardiogram image.
Step S122, using deep learning model, extracts the feature of the multiple view data.
The low-level image feature of multiple images data is extracted, and is processed using deep learning model and is generated sample correspondence
View data feature.Specific deep learning model can form more abstract high-level by combining low-level image feature
Feature, the high-level feature is the feature of the view data, such as the feature of the waveform in the view data.The present embodiment
In, the low-level image feature of the view data of acquisition is carried out multilayer by multiple hidden layers of the convolutional neural networks in deep learning model
Secondary combination, computing, to obtain abstract high-level feature.
Step S123, according to the feature of the multiple view data extracted, using deep learning model, to the multiple
View data is classified to obtain electrocardiogram deep learning model.
According to the feature of the multiple view data, view data is classified using deep learning model, i.e., according to
Different view data is divided into different classifications by different features so that the management of view data more specification.Specifically, it is deep
Degree learning model also includes full articulamentum, and full articulamentum is entered according to the feature of view data using classifier algorithm to view data
Row classification, and then obtain electrocardiogram deep learning model.
Fig. 4 is refer to, is the flow chart of another sub-step of electrocardiogram image-recognizing method provided in an embodiment of the present invention.
Specifically, step S150 also includes step S151 and step S152.
Step S151, according to the feature extracted, is gone out described in the feature using electrocardiogram deep learning Model Matching
Classification.
Specifically, according to classification results, using the corresponding classification of characteristic matching this feature extracted.
Step S152, the classification according to the feature obtains the information of electrocardiogram image.
Specifically, according to the corresponding classification of the feature, this feature is carried out with the feature under the classification under the classification
Compare, look for the corresponding information of the corresponding electrocardiogram image segments of the feature, such as corresponding disease type of this feature, and then obtain
The information of whole electrocardiogram image.
Fig. 5 is refer to, is the block diagram of electrocardiogram pattern recognition device 110 provided in an embodiment of the present invention, the heart
Electrograph pattern recognition device 110 is applied to service terminal 100.
The electrocardiogram pattern recognition device 110 includes splitting module 113, the first extraction module 114 and identification module 115.
Module 113 is split, for the electrocardiogram image of input to be divided into multiple images fragment.
In embodiments of the present invention, step S130 can be performed by fractionation module 113.
First extraction module 114, for being carried out to the multiple image segments using the electrocardiogram deep learning model
Feature extraction.
In embodiments of the present invention, step S140 can be performed by the first extraction module 114.
Identification module 115, it is the multiple using electrocardiogram deep learning Model Identification for according to the feature extracted
Each image segments is obtaining the information of the electrocardiogram image of the input in image segments.
In embodiments of the present invention, step S150 can be performed by identification module 115.
The electrocardiogram pattern recognition device 110 also includes presetting module 111 and training module 112.
The presetting module 111, for predetermined deep learning model.
In embodiments of the present invention, step S110 can be performed by presetting module 111.
Training module 112, for the existing electrocardiogram image according to multiple species, using deep learning method, to institute
State deep learning model to be trained, obtain electrocardiogram deep learning model.
In embodiments of the present invention, step S120 can be performed by training module 112.
Fig. 6 is refer to, is the training module 112 in electrocardiogram pattern recognition device 110 provided in an embodiment of the present invention
Block diagram.
The training module 112 includes segmentation module 1121, the second extraction module 1122 and sort module 1123.
Segmentation module 1121, for the existing electrocardiogram image to be divided into multiple images data.
In embodiments of the present invention, step S121 can be performed by segmentation module 1121.
Second extraction module 1122, for utilizing deep learning model, extracts the feature of the multiple view data.
In embodiments of the present invention, step S122 can be performed by the second extraction module 1122.
Sort module 1123 is right using deep learning model for the feature according to the multiple view data extracted
The multiple view data is classified to obtain electrocardiogram deep learning model.
In embodiments of the present invention, step S123 can be performed by sort module 1123.
Fig. 7 is refer to, is the son of the identification module 115 of electrocardiogram pattern recognition device 110 provided in an embodiment of the present invention
The block diagram of module.
The identification module 115 specifically includes matching module 1151 and acquisition module 1152.
Matching module 1151, for according to the feature extracted, being gone out using electrocardiogram deep learning Model Matching described
Classification belonging to feature.
In embodiments of the present invention, step S151 can be performed by matching module 1151.
Acquisition module 1152, the information of electrocardiogram image is obtained for the classification according to belonging to the feature.
In embodiments of the present invention, step S152 can be performed by acquisition module 1152.
In sum, a kind of electrocardiogram image-recognizing method provided in an embodiment of the present invention, device and service terminal, the heart
Electrograph image-recognizing method is applied to service terminal, and the service terminal includes electrocardiogram deep learning model, while the service end
Hold the input for electrocardiogram image.The electrocardiogram image-recognizing method includes for the electrocardiogram image of input being divided into multiple images
The multiple image segments are carried out feature extraction by fragment using the electrocardiogram deep learning model, according to the feature extracted
Using each image segments in electrocardiogram the multiple image segments of deep learning Model Identification obtaining the electrocardiogram of input
Image information, and then the automatic identification of the electrocardiogram image of input is realized, data volume is have accumulated, while reducing staff
Workload.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, it is also possible to pass through
Other modes are realized.Device embodiment described above is only schematical, for example, flow chart and block diagram in accompanying drawing
Show the device of multiple embodiments of the invention, the architectural framework in the cards of method and computer program product,
Function and operation.At this point, each square frame in flow chart or block diagram can represent one the one of module, program segment or code
Part a, part for the module, program segment or code is used to realize holding for the logic function for specifying comprising one or more
Row instruction.It should also be noted that at some as in the implementation replaced, the function of being marked in square frame can also be being different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially be performed substantially in parallel, they are sometimes
Can perform in the opposite order, this is depending on involved function.It is also noted that every in block diagram and/or flow chart
The combination of the square frame in individual square frame and block diagram and/or flow chart, can use the function or the special base of action for performing regulation
Realized in the system of hardware, or can be realized with the combination of computer instruction with specialized hardware.
In addition, each functional module in each embodiment of the invention can integrate to form an independent portion
Divide, or modules individualism, it is also possible to which two or more modules are integrated to form an independent part.
If the function is to realize in the form of software function module and as independent production marketing or when using, can be with
Storage is in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used to so that a computer equipment (can be individual
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the invention.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.Need
Illustrate, herein, such as first and second or the like relational terms be used merely to by an entity or operation with
Another entity or operation make a distinction, and not necessarily require or imply these entities or there is any this reality between operating
The relation or order on border.And, term " including ", "comprising" or its any other variant be intended to the bag of nonexcludability
Contain, so that process, method, article or equipment including a series of key elements are not only including those key elements, but also including
Other key elements being not expressly set out, or it is this process, method, article or the intrinsic key element of equipment also to include.
In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including the key element
Process, method, article or equipment in also there is other identical element.
The preferred embodiments of the present invention are the foregoing is only, is not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.It is all within the spirit and principles in the present invention, made any repair
Change, equivalent, improvement etc., should be included within the scope of the present invention.It should be noted that：Similar label and letter exists
Similar terms is represented in following accompanying drawing, therefore, once being defined in a certain Xiang Yi accompanying drawing, then it is not required in subsequent accompanying drawing
It is further defined and is explained.
The above, specific embodiment only of the invention, but protection scope of the present invention is not limited thereto, and it is any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (10)
1. a kind of electrocardiogram image-recognizing method, is applied to service terminal, it is characterised in that the service terminal includes electrocardiogram
Deep learning model, the service terminal is used for the input of electrocardiogram image, and methods described includes：
The electrocardiogram image of input is divided into multiple images fragment；
Feature extraction is carried out to the multiple image segments using the electrocardiogram deep learning model；
According to the feature extracted, using each image in electrocardiogram the multiple image segments of deep learning Model Identification
Fragment is obtaining the information of the electrocardiogram image of the input.
2. electrocardiogram image-recognizing method as claimed in claim 1, it is characterised in that described that the electrocardiogram of input is divided into many
Also include step before individual image segments：
Predetermined deep learning model；
According to the existing electrocardiogram image of multiple species, using deep learning method, the deep learning model is instructed
Practice, obtain electrocardiogram deep learning model.
3. electrocardiogram image-recognizing method as claimed in claim 2, it is characterised in that described to enter to the deep learning model
The step of row training, acquisition electrocardiogram deep learning model, includes：
The existing electrocardiogram image is divided into multiple images data；
Using deep learning model, the feature of the multiple view data is extracted；
According to the feature of the multiple view data extracted, using the deep learning model, to the multiple view data
Classified to obtain electrocardiogram deep learning model.
4. electrocardiogram image-recognizing method as claimed in claim 1, it is characterised in that described according to the feature extracted,
Using each image segments in electrocardiogram the multiple image segments of deep learning Model Identification obtaining the heart of the input
The step of information of electrograph image, includes：
According to the feature extracted, the classification belonging to the feature is gone out using electrocardiogram deep learning Model Matching；
Classification according to belonging to the feature obtains the information of electrocardiogram image.
5. a kind of electrocardiogram pattern recognition device, is applied to service terminal, it is characterised in that the service terminal includes electrocardiogram
Deep learning model, the service terminal is used for the input of electrocardiogram image, and described device includes：
Module is split, for the electrocardiogram image of input to be divided into multiple images fragment；
First extraction module, carries for carrying out feature to the multiple image segments using the electrocardiogram deep learning model
Take；
Identification module, for according to the feature extracted, using electrocardiogram the multiple image sheet of deep learning Model Identification
Each image segments is obtaining the information of the electrocardiogram image of the input in section.
6. electrocardiogram pattern recognition device as claimed in claim 5, it is characterised in that the electrocardiogram pattern recognition device is also
Including：
Presetting module, for predetermined deep learning model；
Training module, for the existing electrocardiogram image according to multiple species, using deep learning method, to the depth
Practise model to be trained, obtain electrocardiogram deep learning model.
7. electrocardiogram pattern recognition device as claimed in claim 6, it is characterised in that the training module includes：
Segmentation module, for the existing electrocardiogram image to be divided into multiple images data；
Second extraction module, for utilizing deep learning model, extracts the feature of the multiple view data；
Sort module, for the feature according to the multiple view data extracted, using the deep learning model, to described
Multiple images data are classified to obtain electrocardiogram deep learning model.
8. electrocardiogram pattern recognition device as claimed in claim 5, it is characterised in that the identification module also includes：
Matching module, for according to the feature extracted, being gone out belonging to the feature using electrocardiogram deep learning Model Matching
Classification；
Acquisition module, the information of electrocardiogram image is obtained for the classification according to belonging to the feature.
9. a kind of service terminal, it is characterised in that the service terminal includes electrocardiogram deep learning model, the service terminal
For the input of electrocardiogram image, the service terminal includes：
Memory；
Processor；And
Electrocardiogram pattern recognition device, the electrocardiogram pattern recognition device is installed on the memory and including one or more
By the software function module of the computing device, the electrocardiogram pattern recognition device includes：
Module is split, for the electrocardiogram image of input to be divided into multiple images fragment；
First extraction module, carries for carrying out feature to the multiple image segments using the electrocardiogram deep learning model
Take；
Identification module, for according to the feature extracted, using electrocardiogram the multiple image sheet of deep learning Model Identification
Each image segments is obtaining the information of the electrocardiogram image of the input in section.
10. service terminal as claimed in claim 9, it is characterised in that the electrocardiogram pattern recognition device also includes default
Module, training module, the presetting module are used for predetermined deep learning model, and the training module is used for according to multiple species
Existing electrocardiogram image, using deep learning method, is trained to the deep learning model, obtains electrocardiogram depth
Practise model；The training module also includes segmentation module, the second extraction module and sort module, and the segmentation module is used for institute
State existing electrocardiogram image and be divided into multiple images data, second extraction module is used to utilize deep learning model, extracts
The feature of the multiple view data, the sort module is used for the feature according to the multiple view data extracted, and utilizes
The deep learning model, is classified to obtain electrocardiogram deep learning model, the identification to the multiple view data
Module also includes matching module and obtains not overstate that the matching module is used for according to the feature extracted, using electrocardiogram depth
Degree learning model matches the classification belonging to the feature, and the classification that the acquisition module is used for according to belonging to the feature is obtained
The information of electrocardiogram image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710023150.8A CN106778685A (en) | 2017-01-12 | 2017-01-12 | Electrocardiogram image-recognizing method, device and service terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710023150.8A CN106778685A (en) | 2017-01-12 | 2017-01-12 | Electrocardiogram image-recognizing method, device and service terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106778685A true CN106778685A (en) | 2017-05-31 |
Family
ID=58947662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710023150.8A Pending CN106778685A (en) | 2017-01-12 | 2017-01-12 | Electrocardiogram image-recognizing method, device and service terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106778685A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107256393A (en) * | 2017-06-05 | 2017-10-17 | 四川大学 | The feature extraction and state recognition of one-dimensional physiological signal based on deep learning |
| CN107358196A (en) * | 2017-07-12 | 2017-11-17 | 北京卫嘉高科信息技术有限公司 | A kind of sorting technique of heart beat type, device and electrocardiogram equipment |
| CN107374614A (en) * | 2017-07-25 | 2017-11-24 | 海南聚能科技创新研究院有限公司 | The automatic classification storage control method of electrocardiogram and system |
| CN108464827A (en) * | 2018-03-08 | 2018-08-31 | 四川大学 | It is a kind of it is Weakly supervised under electrocardio image-recognizing method |
| CN109009047A (en) * | 2018-05-24 | 2018-12-18 | 东软集团股份有限公司 | A kind of arrhythmia detection method, device and equipment |
| CN109567789A (en) * | 2018-12-03 | 2019-04-05 | 东软集团股份有限公司 | Division processing method, device and the readable storage medium storing program for executing of ECG data |
| WO2019096178A1 (en) * | 2017-11-14 | 2019-05-23 | 深圳码隆科技有限公司 | Fiber detection method and apparatus, and electronic device |
| CN110141214A (en) * | 2019-04-23 | 2019-08-20 | 首都师范大学 | A labeling method for electrocardiogram recognition and its application |
| CN110236522A (en) * | 2019-05-28 | 2019-09-17 | 北京大学深圳研究生院 | Health screening method, system and medical equipment based on single-lead electrocardiogram |
| US10426364B2 (en) | 2015-10-27 | 2019-10-01 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| WO2019186439A1 (en) * | 2018-03-30 | 2019-10-03 | Cardiologs Technologies Sas | Systems and methods for digitizing electrocardiograms |
| CN110575161A (en) * | 2019-08-27 | 2019-12-17 | 复旦大学 | Atrial Fibrillation Analysis and Prediction Method Based on Cardiac Mapping Activation Sequence |
| US10779744B2 (en) | 2015-10-27 | 2020-09-22 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| CN111931717A (en) * | 2020-09-22 | 2020-11-13 | 平安科技（深圳）有限公司 | Semantic and image recognition-based electrocardiogram information extraction method and device |
| WO2021037101A1 (en) * | 2019-08-27 | 2021-03-04 | 中科麦迪人工智能研究院(苏州)有限公司 | Method, device, apparatus, and medium for image information-based ecg analysis |
| CN113344040A (en) * | 2021-05-20 | 2021-09-03 | 深圳索信达数据技术有限公司 | Image classification method and device, computer equipment and storage medium |
| US11331034B2 (en) | 2015-10-27 | 2022-05-17 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| US11672464B2 (en) | 2015-10-27 | 2023-06-13 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
| US11678831B2 (en) | 2020-08-10 | 2023-06-20 | Cardiologs Technologies Sas | Electrocardiogram processing system for detecting and/or predicting cardiac events |
| US11826150B2 (en) | 2017-08-25 | 2023-11-28 | Koninklijke Philips N.V. | User interface for analysis of electrocardiograms |
| US12016694B2 (en) | 2019-02-04 | 2024-06-25 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101766484A (en) * | 2010-01-18 | 2010-07-07 | 董军 | Method and equipment for identification and classification of electrocardiogram |
| CN103815897A (en) * | 2014-02-28 | 2014-05-28 | 吉林大学 | Electrocardiogram characteristic extraction method |
| CN105125206A (en) * | 2015-09-15 | 2015-12-09 | 中山大学 | Intelligent electrocardio monitoring method and device |
| CN105769177A (en) * | 2016-04-01 | 2016-07-20 | 佛山市质量计量监督检测中心 | Electrocardiograph signal examination record analyzing device |
| US20160264969A1 (en) * | 2009-05-01 | 2016-09-15 | Ophthotech Corporation | Methods for treating or preventing ophthalmological diseases |
| CN106108889A (en) * | 2016-07-20 | 2016-11-16 | 杨平 | Electrocardiogram classification method based on degree of depth learning algorithm |
| CN106214145A (en) * | 2016-07-20 | 2016-12-14 | 杨平 | A kind of electrocardiogram classification method based on degree of depth learning algorithm |
| CN106214123A (en) * | 2016-07-20 | 2016-12-14 | 杨平 | A kind of electrocardiogram compressive classification method based on degree of depth learning algorithm |
-
2017
- 2017-01-12 CN CN201710023150.8A patent/CN106778685A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160264969A1 (en) * | 2009-05-01 | 2016-09-15 | Ophthotech Corporation | Methods for treating or preventing ophthalmological diseases |
| CN101766484A (en) * | 2010-01-18 | 2010-07-07 | 董军 | Method and equipment for identification and classification of electrocardiogram |
| CN103815897A (en) * | 2014-02-28 | 2014-05-28 | 吉林大学 | Electrocardiogram characteristic extraction method |
| CN105125206A (en) * | 2015-09-15 | 2015-12-09 | 中山大学 | Intelligent electrocardio monitoring method and device |
| CN105769177A (en) * | 2016-04-01 | 2016-07-20 | 佛山市质量计量监督检测中心 | Electrocardiograph signal examination record analyzing device |
| CN106108889A (en) * | 2016-07-20 | 2016-11-16 | 杨平 | Electrocardiogram classification method based on degree of depth learning algorithm |
| CN106214145A (en) * | 2016-07-20 | 2016-12-14 | 杨平 | A kind of electrocardiogram classification method based on degree of depth learning algorithm |
| CN106214123A (en) * | 2016-07-20 | 2016-12-14 | 杨平 | A kind of electrocardiogram compressive classification method based on degree of depth learning algorithm |
Non-Patent Citations (3)
| Title |
|---|
| MARSCHALL S.RUNGE等: "《奈特心脏病学》", 30 June 2015, 人民军医出版社 * |
| 叶本兰等: "《全国普通高等医学院校五年制临床医学专业"十三五"规划教材 生理学》", 31 August 2016, 中国医药科技出版社 * |
| 颐恒等: "《中西医结合实践技能内部教程》", 31 December 2014, 中国广播影视出版社 * |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11147500B2 (en) | 2015-10-27 | 2021-10-19 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
| US10758139B2 (en) | 2015-10-27 | 2020-09-01 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| US10959660B2 (en) | 2015-10-27 | 2021-03-30 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
| US10779744B2 (en) | 2015-10-27 | 2020-09-22 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| US11134880B2 (en) | 2015-10-27 | 2021-10-05 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| US11672464B2 (en) | 2015-10-27 | 2023-06-13 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
| US10426364B2 (en) | 2015-10-27 | 2019-10-01 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| US11331034B2 (en) | 2015-10-27 | 2022-05-17 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
| CN107256393A (en) * | 2017-06-05 | 2017-10-17 | 四川大学 | The feature extraction and state recognition of one-dimensional physiological signal based on deep learning |
| CN107358196B (en) * | 2017-07-12 | 2020-11-10 | 北京卫嘉高科信息技术有限公司 | Heart beat type classification method and device and electrocardiograph |
| CN107358196A (en) * | 2017-07-12 | 2017-11-17 | 北京卫嘉高科信息技术有限公司 | A kind of sorting technique of heart beat type, device and electrocardiogram equipment |
| CN107374614A (en) * | 2017-07-25 | 2017-11-24 | 海南聚能科技创新研究院有限公司 | The automatic classification storage control method of electrocardiogram and system |
| US11826150B2 (en) | 2017-08-25 | 2023-11-28 | Koninklijke Philips N.V. | User interface for analysis of electrocardiograms |
| WO2019096178A1 (en) * | 2017-11-14 | 2019-05-23 | 深圳码隆科技有限公司 | Fiber detection method and apparatus, and electronic device |
| CN108464827A (en) * | 2018-03-08 | 2018-08-31 | 四川大学 | It is a kind of it is Weakly supervised under electrocardio image-recognizing method |
| US10827938B2 (en) | 2018-03-30 | 2020-11-10 | Cardiologs Technologies Sas | Systems and methods for digitizing electrocardiograms |
| WO2019186439A1 (en) * | 2018-03-30 | 2019-10-03 | Cardiologs Technologies Sas | Systems and methods for digitizing electrocardiograms |
| CN109009047A (en) * | 2018-05-24 | 2018-12-18 | 东软集团股份有限公司 | A kind of arrhythmia detection method, device and equipment |
| CN109009047B (en) * | 2018-05-24 | 2021-10-26 | 东软集团股份有限公司 | Arrhythmia detection method, device and equipment |
| CN109567789A (en) * | 2018-12-03 | 2019-04-05 | 东软集团股份有限公司 | Division processing method, device and the readable storage medium storing program for executing of ECG data |
| US12016694B2 (en) | 2019-02-04 | 2024-06-25 | Cardiologs Technologies Sas | Electrocardiogram processing system for delineation and classification |
| CN110141214A (en) * | 2019-04-23 | 2019-08-20 | 首都师范大学 | A labeling method for electrocardiogram recognition and its application |
| CN110236522A (en) * | 2019-05-28 | 2019-09-17 | 北京大学深圳研究生院 | Health screening method, system and medical equipment based on single-lead electrocardiogram |
| CN110575161A (en) * | 2019-08-27 | 2019-12-17 | 复旦大学 | Atrial Fibrillation Analysis and Prediction Method Based on Cardiac Mapping Activation Sequence |
| CN110575161B (en) * | 2019-08-27 | 2021-12-28 | 复旦大学 | Atrial fibrillation analysis prediction method based on cardiac mapping activation sequence diagram |
| WO2021037101A1 (en) * | 2019-08-27 | 2021-03-04 | 中科麦迪人工智能研究院(苏州)有限公司 | Method, device, apparatus, and medium for image information-based ecg analysis |
| US11678831B2 (en) | 2020-08-10 | 2023-06-20 | Cardiologs Technologies Sas | Electrocardiogram processing system for detecting and/or predicting cardiac events |
| CN111931717A (en) * | 2020-09-22 | 2020-11-13 | 平安科技（深圳）有限公司 | Semantic and image recognition-based electrocardiogram information extraction method and device |
| CN113344040A (en) * | 2021-05-20 | 2021-09-03 | 深圳索信达数据技术有限公司 | Image classification method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106778685A (en) | Electrocardiogram image-recognizing method, device and service terminal | |
| Chambon et al. | DOSED: A deep learning approach to detect multiple sleep micro-events in EEG signal | |
| CN109620152B (en) | MutifacolLoss-densenert-based electrocardiosignal classification method | |
| CN107133651B (en) | The functional magnetic resonance imaging data classification method of subgraph is differentiated based on super-network | |
| CN108256452A (en) | A kind of method of the ECG signal classification of feature based fusion | |
| CN111956212B (en) | Inter-group atrial fibrillation recognition method based on frequency domain filtering-multi-mode deep neural network | |
| CN107194158A (en) | A kind of disease aided diagnosis method based on image recognition | |
| CN111000553A (en) | Intelligent classification method for electrocardiogram data based on voting ensemble learning | |
| CN111785366B (en) | Patient treatment scheme determination method and device and computer equipment | |
| US20240260885A1 (en) | Street greening quality detection method based on physiological activation recognition | |
| CN110141220A (en) | Automatic detection method of myocardial infarction based on multimodal fusion neural network | |
| CN110141219A (en) | Automatic detection method of myocardial infarction based on lead fusion deep neural network | |
| CN112674782B (en) | Device and method for detecting epileptic-like electrical activity of epileptic during inter-seizure period | |
| Kusrini et al. | The effect of Gaussian filter and data preprocessing on the classification of Punakawan puppet images with the convolutional neural network algorithm | |
| Khalil et al. | Diagnosis of blood cells using deep learning | |
| CN106803074A (en) | Chinese herbal medicine recognition methods and device | |
| CN111134664A (en) | Epileptic discharge identification method and system based on capsule network and storage medium | |
| Perumal et al. | Guava leaf disease classification using support vector machine | |
| CN110718301B (en) | Alzheimer's disease auxiliary diagnosis device and method based on dynamic brain function network | |
| CN108836313A (en) | A kind of intelligence chooses the ambulatory ECG analysis method of lead | |
| Al Qatrawi et al. | Classification of chicken diseases using deep learning | |
| CN113288157A (en) | Arrhythmia classification method based on depth separable convolution and improved loss function | |
| Muhali et al. | Classification of Lanzones Tree Leaf Diseases Using Image Processing Technology and a Convolutional Neural Network (CNN) | |
| Kamaruzaman et al. | Systematic literature review: application of deep learning processing technique for fig fruit detection and counting | |
| Firdaus et al. | Detection and classification of citrus diseases based on a combination of features using the densenet-169 model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
| RJ01 | Rejection of invention patent application after publication |