The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of CIRCUITS, SYSTEMS and SIGNAL PROCESSING
Transactions ID Number: 19-920
Full Name: Alireza Sahab
Position: Assistant Professor
Age: ON
Sex: Male
Address: Motahhari Ave. - Arjang Ave. - Saiiah All. - No.60 - 2nd Floor - Bandar Anzali - Guilan - Iran
Country: IRAN
Tel: 9111832163
Tel prefix: 0098
Fax: 00981412222150
E-mail address: sahab@iau-lahijan.ac.ir
Other E-mails: ali.reza.sahab@gmail.com
Title of the Paper: an automatic diagnostic machine for ecg arrhythmias classification based on wavelet transformation and neural networks
Authors as they appear in the Paper: Ali reza Sahab, Yadollah Mehrzad Gilmalek
Email addresses of all the authors: sahab@iau-lahijan.ac.ir,jaber.mehrzad@gmail.com
Number of paper pages: 8
Abstract: The objective of this paper is to design a heart arrhythmias diagnosis instrument that has very low complicated computations. Therefore, a ECG classifier system based on discreet wavelet (DW) transformation and multi layer Perceptron neural network is presented. There is a new Idea in this paper in which signal is pre-processed in order to omit its noises firstly, then, using DW, 6db of signal is divided into eight levels and the minimum, maximum, variance and standard deviation of the signal are obtained. In addition, time features of the signal are obtained. Then combining time features with discrete wavelet output features an array of them are made to be used as final features in order to teach and test a 3-layer MLP neural network. Finally, using 255 heart signal samples existed in MIT-BIH data base, designed Classifier is taught and tested and in its best performance accuracy of 98% percentage have been obtained for three different heart arrhythmias of ECG sig!
nals include; RBBB,LBBB and normal heart rhythm.
Keywords: Diagnostic Machine, ECG, Classifier, Wavelet Transformation, Heart Arrhythmia, Neural Networks, Features
EXTENSION of the file: .pdf
Special (Invited) Session: ECG arrhythmias classification using wavelet transform and neural networks
Organizer of the Session: 202-292
How Did you learn about congress:
IP ADDRESS: 109.109.42.4
