Diagnosis of coronary artery diseases using classification algorithms based on wavelet transforms

Abstract

One of the primary drivers of the death in the world is cardiovascular diseases (CAD) which is a major threat in developing and developed countries. The fundamental drivers in CAD results in clogging of the coronary lumen consequently occlusion, and after that prompts to myocardial dead tissue (MI) or sudden heart attach which causes death. It is difficult to ascertain that a certain person has been affected by CAD, since there are bunch of parameters has been involved to ascertain the conclusion. Classification has been done using wavelet transform to classify the certain parameters. We analyzed following methods such as NB, Logistic, SMO, RBF Network, K-star, Multiclass Classifier, Conjunctive rule, Decision table, DTNB, LAD Tree, LMT, NB Tree, Random forest and Random Tree calculations has been associated with extensive fragment of the surveys. This database has been generated from UCI machine learning database. In this paper, we used 10-fold cross validation with 14 attributes and calculations of TP rate, FP rate, Precision, Recall, F-measure, ROC and Accuracy are analyzed practically. As a result, the Logistic, SMO and LMT algorithms has yield to improve the high accuracy rate of 77.0%.