Unlock instant, AI-driven research and patent intelligence for your innovation.

ST segment classification convolutional neural network based on feature selection and its application method

A convolutional neural network and neural network technology, applied in the field of ST segment classification convolutional neural network based on feature selection, can solve the problems of low size, low maturity of ST segment waveform automatic identification, and the occurrence mechanism is not completely clear. Effects of reduced impact, reduced fitting process, good robustness

Active Publication Date: 2020-12-04
SHANGHAI SID MEDICAL CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] (4) U wave: The U wave is located behind the T wave, relatively low and small, and its mechanism of occurrence is not completely clear
At present, there are many classification and recognition algorithms for ECG signals, but the maturity of automatic recognition of ST-segment waveforms is relatively low.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • ST segment classification convolutional neural network based on feature selection and its application method
  • ST segment classification convolutional neural network based on feature selection and its application method
  • ST segment classification convolutional neural network based on feature selection and its application method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] The present embodiment provides a kind of ST segment classification network neural training method based on feature selection, comprising the following steps:

[0066] Step S1: Data collection and preprocessing: collect enough clinical resting 12-lead electrocardiograms of known types to form an ECG signal training set. The types of clinical resting 12-lead electrocardiograms include normal electrocardiograms, ST segment There are four types of high, ST-segment depression, and ST arch elevation, and the number of different types of ECG signals is even; select the leads with ST-segment elevation, ST-segment depression, and ST arch elevation among the 12 leads. The joint signal, and any lead ECG signal of the normal ECG is randomly selected to form the ECG signal training set. The label vectors corresponding to the normal ECG, ST-segment level elevation, ST-segment level depression, and ST arch-back elevation are ( a, b, c, d), only one of a, b, c, d is 1, and the rest ar...

Embodiment 2

[0082] The present embodiment includes a kind of ST segment classification network nerve based on feature selection, including:

[0083] The first convolutional neural network, the second convolutional neural network and the third convolutional neural network, and two independent fully connected layers;

[0084] Layer1-layer7 of the first convolutional neural network consists of a convolutional layer and a pooling layer; the convolutional layer in layer1 contains 5 cores, the size of the convolutional kernel is 29, and the step size in the pooling layer in layer1 and the kernel size are both 2; the layer2 convolutional layer contains 5 kernels, and the convolution kernel size is 15, and the step size and kernel size in the pooling layer in layer2 are both 2; the layer3 convolutional layer contains 5 kernels, and the volume The kernel size is 13, the step size and kernel size of the pooling layer in layer3 are both 2; the layer4 convolutional layer contains 10 kernels, the conv...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

This application relates to a ST segment classification convolutional neural network based on feature selection, by using multiple convolutional neural networks with different structures and filtering out the output of the last pooling layer of multiple convolutional neural networks with high correlation Part of the output of is used as the input of the independent fully connected layer, and finally the ST segment classification type can be determined according to the output of the independent fully connected layer. The ST segment classification network neural network based on feature selection in this application has good robustness, reduces the fitting process, and reduces the impact on the results due to too few training sets.

Description

technical field [0001] The present application belongs to the technical field of electrocardiogram processing, and in particular relates to a ST segment classification convolutional neural network based on feature selection and its application method. Background technique [0002] The ECG is made up of a series of wave groups, each wave group representing each cardiac cycle. A wave group includes P wave, QRS complex, T wave and U wave, such as figure 1 shown. See what each wave represents: [0003] (1) P wave: The excitation of the heart originates from the sinoatrial node, and then conducts to the atrium. The P wave is produced by atrial depolarization and is the first wave in each wave group, which reflects the depolarization process of the left and right atria. The first half represents the right atrium and the second half represents the left atrium. [0004] (2) QRS complex: A typical QRS complex consists of three closely connected waves. The first downward wave is ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08G06N3/04A61B5/0452A61B5/00
CPCG06N3/08A61B5/7235A61B5/7267A61B5/316A61B5/349G06N3/045G06F2218/12G06F18/214
Inventor 朱俊江王雨轩黄浩林彩梅
Owner SHANGHAI SID MEDICAL CO LTD
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com