Intelligent Heart Rate Classification with Adaptive Neuro-Fuzzy Inference System Approach

The human heart is considered among the most crucial internal organs found in the body. An ECG signal is defined as the widespread name given to any electrical signal that is generated by the circulatory system*. By using the ECG signal, one can estimate anomalies present in a heart. This study is aimed at grouping the electrocardiogram (ECG) signals, and thereby using an adaptive neuro fuzzy inference system (ANFIS) algorithm. In this study, the ANFIS which is associated with the back propagation approach is used in order to perform the categorising task. ANFIS can be developed from the integration of fuzzy logic which is a qualitative technique with the neural network aspect in allowing adaptability. Attribute selection is done before the classification exercise is conducted. Four different types of ECG beats can be retrieved from the datasets available at PhysioBank. Classification of the cardiac signals is done employing four ANFIS classifiers. The fifth ANFIS decoder is used to get a higher discrimination ratio in order to classify the electrocardiograms accurately.