Cardiac Arrhythmia Classifier

The irregularities in the heartbeat are called arrhythmias and can be an essential subject for hearthealth. The artifacts in the mechanical structure or the electrical signaling system of the heart result in theseirregularities. Arrhythmias depending on the different artifacts, cause different patterns onelectroencephalogram recordings of patients. Early diagnosis of cardiac arrhythmia is quite crucial to savingpatient lives. The main goal of this study is to detect the presence of cardiac arrhythmia and classify it intosixteen different groups from the electrocardiogram (ECG) recordings. First, the classificationperformances of four well-known classification algorithms, namely, multilayer perceptron (MLP), supportvector machines (SVM), random forest (RF), and K-means clustering, have been obtained using animbalanced electrocardiogram (ECG) dataset. The classifier performance criteria such as accuracy andprecision are used to determine the best classifier structure for each arrhythmia class. Then, to obtain abetter classifier performance, an ensemble network which is a cascaded form of two RF classifiers, isconstructed. The designed ensemble network consisting of two RF networks results in an accuracy of73.63%.