Shirin Badizadegan - rControl and Intelligent Processing Center of Excellence, Electrical and Computer Engineering Department, University of Tehran, Tehran
Hamid Soltanian-Zadeh - rControl and Intelligent Processing Center of Excellence, Electrical and Computer Engineering Department, University of Tehran, Tehranو tlmage Analysis Laboratory, Radiology Department, Henry Ford Heatth System, Detroit, MI ۴g۲۰۲, USA

lavelet-based methods hove led to the most accuratefeature extraction results, especially in the non-stationary signls, among the recent proposed methods. The methods proposed so far have used standard and unmatched wavelet bases such as Daubechies and quadratic spline. In this work, two new matched wavelet bases with the maximum gain criterion are designed and used for analyzing and R peak detection in the ECG signols. A modified wavelet-based R peak detection algorithm is ,also presented and evaluated using MIT-BIH Arrhythmia database. Experimental results show both'excellent performance and high speed of the proposed methods.

Wavelets transform, matched wavelets. coding gain, R peak detection, optimization