Double constrained genetic algorithm for ECG signal classification
Elangovan Ramanujam, S. Venkata Padmavathi · 2016
Analyzing cardiovascular activity under abnormal heart beat is an intricate and vital job to the medical experts and complicated to novice persons. Electrocardiogram is a way to measure or diagnose abnormal heart rhythms to spot heart disease in human beings. These streaming medical signals can be well analyzed or diagnosed only with the prior knowledge. This paper deals with ECG signal analysis based on Time Series Motif as feature using Genetic Algorithm with double constraints. Genetic Algorithm has been employed to extract the features of various lengths from ECG signals. The proposed technique classifies the ECG signals into two classes (normal and abnormal) the classifier performance is measured in terms of Sensitivity, Specificity and Accuracy. Experimental results on standard MIT-BIH arrhythmia database shows that the proposed approach achieves 97.78% of accuracy.