Recurrent neural network-based model for electrocardiogram classification

Name: Panlop Pantuprecharat

Address: Manhanakorn University of Technology.Faculty of Engineering and Technolgy.Department of Electronic Engineering

Name: Prajuab Pawarangkoon

Address: Manhanakorn University of Technology.Faculty of Engineering and Technolgy.Department of Electronic Engineering

Name: Suriya Adirek

Address: Nakhon Sawan Rajabhat University.Faculty of Agriculture and Industrial

keyword: Deep learning

; Electrocardiogram (ECG)

; Recurrent neural network (RNN)

; Gated recurrent unit (GRU)

; Long short-term memory (LSTM)

Abstract: This paper presents the study of deep learning models for electrocardiogram (ECG) classification. Abnormalities of heart diseases can detect and diagnose by ECG signal. Deep learning models have been interested for arrhythmia detection and classification from ECG signal. Recurrent neural networks (RNNs) represent a different category of artificial neural networks that incorporate feedback mechanisms. It makes suitable for capturing temporal dependencies in the time series data from ECG electrical signal. Long short-term memory (LSTM) network and gated recurrent unit (GRU) are types of RNN designs. The objective of this paper is to compare the performance of conventional RNN with LSTM and GRU models for capturing the long-term relationships in ECG data. Simulation results show that GRU model can classify the heart's diseases more accuracy than simple RNN and LSTM model.

Mahanakorn University of Technology. Library

Address: BANGKOK

Email: mutelibrary@gmail.com

Created: 2025

Modified: 2025-11-20

Issued: 2025-11-20

บทความ/Article

application/pdf

BibliograpyCitation : Engineering Transactions Vol. 28, no.1 (58) (Jan.-Dec. 2025), p. 19-25

eng

©copyrights Mahanakorn University of Technology

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	v.28 n.1 p. 19-25.pdf	1.35 MB

ใช้เวลา

0.033049 วินาที