DefinePK

Abstract

In this work selection and combination technique is used to enhance the performance of classifiers. The main objective of the work is to select efficient classifiers, from a group of seven classifiers, based on the error rate and combining them, to determine multiple heart diseases simultaneously. The classifiers are combined efficiently based on the decision of each classifier thus improving the classification accuracy of the signal. Heart ailments are the major health problem in the current scenario. These ailments in heart may even lead to death when not found at correct time. In recent years rate of patients with heart disease is increasing due to stress and unhygienic activities etc. This work is mainly done to develop an efficient detection system of multiple heart ailments. Three types of ECG signal, Normal, Arrhythmia and Sudden cardiac arrest signal is taken from physionet database is used as input signal.