Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Systems and Biomedical Engineering

We presented several approaches to enhance the performance of arrhythmia CAD system by introducing modifications on feature extraction and classification methods. In the first approach, spectral correlation coefficients were estimated for arrhythmias of interest. In the other three approaches, novel Bayesian and KNN ensembles employing the structures of one-versus-all and one-versus-one used for multi-class SVM classifier were introduced. The proposed methods outperformed all studies in comparison. In addition, the testing time follows AAMI standards

Issued also as CD