Computational Intelligence Techniques on Cardiac Arrhythmia

The book contributes to the study and examination of the efficiency of Computational Intelligence techniques for feature extraction in ECG signal. The 1st chapter discusses ¿Electrocardiogram (ECG), Cardiac Arrhythmias, Wavelet transformations, and some Computational Intelligence techniques¿. The 2nd chapter introduces various techniques for feature extraction in ECG signal from rich literature. The 3rd chapter introduces ¿QRS Detection of ECG ¿ A Statistical Analysis¿ using Discrete and Multi Wavelet Transforms with double differentiation. The 4th chapter deals with the ¿Hybrid Methodologies on Performance identification of various Cardiac Arrhythmias based on Neural Network Classification¿. The 5th chapter deals with Cardiac Arrhythmia Classification Using Probabilistic Neural Networks. It includes denoising, wavelet decomposition, autoregression coefficients. Effective training and classification can be done using Probabilistic Neural Networks for Cardiac Arrhythmia. The model is tested on MIT-BIH database and Error% is taken as a performance criterion. The Probabilistic Neural Networks Approach resulted in accuracy 99.04%.The 6th chapter of the book discusses the results.