Efficient Combination Scheme for Arrhythmia Detection Using ECG Signals

Arrhythmia is a health problem and it is a drastic cause for many kinds of heart diseases. Some arrhythmia may even cause to death. The main objective of this work is to combine the decision of each classifier in an efficient way to improve the classification accuracy of the classifiers to predict the disease. As different classifiers provide different opinion to the target system, the combined decision will provide more robust and accurate result. The HRV parameters obtained from the ECG of normal and the arrhythmia affected patients is used for classification. Combination of classifiers like k-Nearest Neighbor (k-NN), Principal Component Analysis (PCA), Probabilistic neural network, Decision tree and Adaptive boost classifier is used in this work. The choice or judgment of the efficient classifiers is selected and combined using majority voting system for detection of final classification.