Combining Nonlinear Fractal Transformation and Neural Network Based Classifier for Cardiac Arrhythmias Recognition

−This paper proposes a method for cardiac arrhythmias recognition using fractal transformation (FT) and neural network based classifier. Iterated function system (IFS) uses the non-linear interpolation in the map and FT with fractal dimension (FD) is used to construct various fractal patterns, including supra-ventricular ectopic beat, bundle branch ectopic beat, and ventricular ectopic beat. Probabilistic neural network (PNN) is used to recognize normal heartbeat and multiple cardiac arrhythmias. The proposed classifier is tested using the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) arrhythmia database. Compared with other method, the results will show the efficiency of the proposed method, and also show high accuracy for recognizing electrocardiogram (ECG) signals. Keywords⎯Fractal Transformation (FT), Iterated Function System (IFS), Non-linear Interpolation, Fractal Dimension (FD), Probabilistic Neural Network (PNN), Electrocardiogram (ECG).