Ceylan, RahimeOzbay, Yuksel2020-03-262020-03-262011978-988-19251-4-52078-0958https://hdl.handle.net/20.500.12395/27047World Congress on Engineering (WCE 2011) -- JUL 06-08, 2011 -- Imperial Coll, London, UNITED KINGDOMBundle branch blocks are very important for the heart treatment immediately. Left and right bundle branch blocks represent an independent predictor in which underlying cardiac disease that needs to be treated. In this study, we presented a model of wavelet neural network for classification of bundle branch blocks. The proposed wavelet neural network was implemented using Morlet and Mexican hat wavelet functions as activation function in hidden layer. ECG data in this study were formed by taken from MIT-BIH ECG Arrhythmia Database. Training and test data consist of three different beat types, which are belong to ECG signal classes of normal, right bundle branch block and left bundle branch block. The performed experimental studies were demonstrated that wavelet neural network designed by Mexican hat wavelet was successful than other network which designed by Morlet wavelet.eninfo:eu-repo/semantics/closedAccessWavelet neural networkECGclassificationQRS detectionConference Object10031007N/AWOS:000393012800016N/A