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

Electrocardiogram waveform segmentation method based on time-frequency analysis and recurrent neural network

A technology of cyclic neural network and electrocardiogram waveform, which is applied in medical science, sensors, diagnostic recording/measurement, etc., can solve the problems of time-consuming, tedious, labeling errors, etc., and achieve the goal of reducing workload, avoiding boring and time-consuming, and improving classification The effect of efficiency and accuracy

Inactive Publication Date: 2021-08-03
GUANGYUAN CENT HOSPITAL
View PDF11 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for an ECG signal with a sampling time of about ten minutes, there are thousands of heartbeat cycles. It is a very time-consuming and tedious task for doctors to manually annotate each area of ​​the ECG signal, and there may be errors in the annotation

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
  • Electrocardiogram waveform segmentation method based on time-frequency analysis and recurrent neural network
  • Electrocardiogram waveform segmentation method based on time-frequency analysis and recurrent neural network
  • Electrocardiogram waveform segmentation method based on time-frequency analysis and recurrent neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] The present embodiment verifies the electrocardiogram waveform segmentation method based on time-frequency analysis and recurrent neural network proposed by the present invention, and the specific contents are as follows:

[0045] The original ECG data set comes from the QT database, which consists of about 15 minutes of ECG recording segments from 105 patients, with a sampling rate of 250 Hz and dual-channel signals. The first few rows of the label table of the ECG data set are shown in Table 1. Each row of the table corresponds to a patient, and each column corresponds to a channel. The label values ​​near the 150th sample of the first channel of the first patient are shown in Table 2. This area marks the end of the QRS complex and the transition to N / A.

[0046] Table 1

[0047] WaveformLabels_Chan1 WaveformLabels_Chan2 Member 1 225000×2table 225000×2table Member 2 225000×2table 225000×2table Member 3 225000×2table 225000×2ta...

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

The invention discloses an electrocardiogram waveform segmentation method based on time-frequency analysis and a recurrent neural network. The method comprises the steps: carrying out the preprocessing of an obtained labeled original electrocardiogram (ECG) data set through a time-frequency analysis method, and dividing the data set into a training set and a test set according to a preset proportion; constructing a recurrent neural network, and sending the recurrent neural network into the training set for training to obtain a trained recurrent neural network; and substituting the test set into the trained recurrent neural network, and selecting an optimal ECG waveform segmentation model in combination with a test result and an evaluation index; and inputting the ECG data sequence to be labeled into the ECG waveform segmentation model to obtain an ECG segmentation result. According to the ECG waveform segmentation method based on the signal processing and deep learning method provided by the invention, each region of the ECG signal can be automatically marked, and the efficiency and accuracy of a doctor or an ECG automatic classification device for diagnosing the heart health state of a patient are improved.

Description

technical field [0001] The invention relates to the related field of electrocardiogram waveform segmentation, in particular to an electrocardiogram waveform segmentation method based on time-frequency analysis and cyclic neural network. Background technique [0002] Electrocardiogram (ECG) is the response of potential signals on both sides of the human myocardial cell membrane, which can reflect the health status of the human heart and is one of the important basis for doctors to diagnose and treat heart diseases. For a single normal heartbeat cycle, the ECG signal can be divided into the following wave forms: P waves (small excursions in front of the QRS complex representing atrial depolarization), QRS complexes (the largest amplitude part of the heartbeat), T waves wave (a small excursion after the QRS complex representing ventricular repolarization). After getting the patient's ECG, the doctor first marks the trend of each waveform (that is, waveform segmentation), so as...

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
IPC IPC(8): A61B5/349A61B5/00
CPCA61B5/7225A61B5/7264A61B5/7203A61B5/7267
Inventor 宋鑫星付贤飞李云鹰马再华张雪芹
Owner GUANGYUAN CENT HOSPITAL