ECG Signal Classification using Wavelet Transform and Back Propagation Neural Network

This paper addressed the use of Back Propagation Neural Network for Classification of ECG waveforms using discrete wavelet transform. We have been selected of MIT-BIH arrhythmia database and picked up 45 files out of 48 files of one minute recording where 25 files are considered as normal class and 20 files of abnormal based on Maximum number of beats present in each record. Proposed method used to classify ECG signal data for abnormal class using BPNN. The features are break up in to two classes that is DWT based features and morphological feature of ECG signal which is an input to the classifier. Back Propagation Neural Network (BPNN) was used to classify the ECG data and the system performance is measured on the basis of percentage accuracy. For the Abnormal sample 100% of accuracy is reached whereas 96% of accuracy was achieved for normal ECG sample. The overall system accuracy 97.8% was obtained with the use of BPNN classifier.