A Comparative Analysis of Biomedical Data Mining Models for Cardiac Signal Classification

摘要

Application of machine learning in healthcare sector is increasing day-by-day. It can be very useful for automated and early diagnosis of different diseases. In the proposed work, authors have compared the classification performance of three different classifiers for cardiac signal classification. The ECG data is collected form Physionet database. Relevant features are extracted from the original signal by applying dual tree complex wavelet transform (DTCWT). Multi-layer perceptron (MLP), radial basis function (RBFN), and support vector machine (SVM) classifiers are considered for classifying the cardiac signal. From the result, it can be observed that, SVM is performing better as compare to other two types of classifiers.