Current Result Document : 4 / 184
| 한글제목(Korean Title) |
1D 컨볼루션 신경망 기반 ECG 부정맥 분류 |
| 영문제목(English Title) |
ECG Arrhythmia Classification based on 1D Convolutional Neural Network |
| 저자(Author) |
주우
강경태
Yu Zhou
Kyungtae Kang
|
| 원문수록처(Citation) |
VOL 48 NO. 02 PP. 0596 ~ 0598 (2021. 12) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
ECG analysis and classification are the basic tools to monitor cardiac activity and patient health today. The main purpose is to detect the appearance of arrhythmias. Many studies have been proposed to diagnose arrhythmia using classification techniques based on various heartbeat characteristics through a diagnostic electrocardiogram (ECG). This study proposes a model of prediction and classification of heartbeat types based on a one-dimensional convolutional neuronal network, which judges 11 types of heartbeat. The original data present an unbalanced classification problem, and the data augmentation technology is used at the pre-processing stage of the data. The ECG recordings from the MIT-BIH arrhythmia database were applied to train the proposed CNN model. The results show that the proposed model achieves 97% accuracy, 99% mean macro recall, 97% mean macro F1-score, and the AUC score for each category was higher than 97%. Our experimental results show that our proposed method is useful for arrhythmia detection.
|
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
