PC-based ECG signal analysis using artificial neural network

Author

Chang, Ching-su ; Chen, Hsing-ton ; Tan, Tan-Hsu ; Chen, Yung-fu

Author_Institution

Dept. of Electr. Eng., Nat. Taipei Univ. of Technol., Taipei

Volume

6

fYear

2008

fDate

12-15 July 2008

Firstpage

3334

Lastpage

3340

Abstract

This study develops a real-time personal computer-based (PC-based) ECG signal measurement and monitoring system to allow PC users to monitor their health status at any time. The system includes a signal processing unit and a health evaluation unit. The signal processing unit consists of a front-end ECG signal amplification circuit, electrocardiogram (ECG) signal acquisition circuit, USB power supply circuit and a control circuit. The second unit is a software tool developed on a PC for evaluating userpsilas health based on heart rate variability (HRV) data. A back-propagation neural network is trained by using the features of HRV message to classify ECG signals for further diagnosis via telemedicine. Experimental results indicate that the proposed system has practical potential in health monitoring and evaluation for PC users.

Keywords

backpropagation; electrocardiography; medical signal processing; neural nets; telemedicine; ECG signal monitoring system; PC-based ECG signal analysis; artificial neural network; back-propagation neural network; electrocardiogram; health evaluation unit; heart rate variability data; telemedicine; Artificial neural networks; Circuits; Computerized monitoring; Electrocardiography; Heart rate variability; Real time systems; Signal analysis; Signal processing; Time measurement; Universal Serial Bus; Personal computer (PC); electrocardiogram (ECG) signal measurement and monitoring; health evaluation; heart rate variability (HRV); neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2008 International Conference on

Conference_Location

Kunming

Print_ISBN

978-1-4244-2095-7

Electronic_ISBN

978-1-4244-2096-4

Type

conf

DOI

10.1109/ICMLC.2008.4620981

Filename

4620981