Abstract:
Electrocardiogram (ECG)is one of the most important techniques that used for diagnosing cardiac arrhythmias. Automatic detection and classification of ECG signals is paramount since scrutinizing each and every beat is a tedious job specially when we need to record the heart’s electrical activity versus time, this is tediousness leads to increase the human error factor in the cardiologist decisions. In this research, have been used an accurate method of classification and differentiation of Normal and abnormal heartbeats.
1500ECG signals were collected from MIT-BIH Arrhythmia database in Physionet bank in Physionet website. 1000 of these signals are said to be abnormal and the rest 500 signals are normal sinus rhythm ECG. These signals were processed to remove baseline wonder and high frequency noise using band path filter.
The Matlab program was also developed to extract the main feature of the ECG signals during a set period of time. 10 features were extracted and divided into 2 groups : ECG features and statistical features, All the features were presented in an excel file and used in the development of the ANN. the signal has been segmented into smaller samples then detecting the R peak knowing that it is the easiest feature to detect because it’s the highest peak. The rest of the features were collected easily once R peak is known. Artificial Neural Network was used in the classification step. The normal signal was given the number 1 while the abnormal sample was 0 in the construction of ANN. All these data were used in the training of ANN and the results was collected and discussed. The system successfully classified and differentiate between the normal and abnormal with accuracy 99.7%, and between atrial premature beat(APB) and paced beat(PB) with accuracy 95.1%.
