Detection of the QRS complex, P wave and T wave in electrocardiogram

Electrocardiogram (ECG) provides useful information of the condition of heart. Most automatic ECG diagnosis techniques require an accurate detection of the QRS complexes. In this project, two QRS detection methods are studied. The first method, the `So and Chan' method, is based on the maximum slope detection with the QRS onset selected when two successive values of the slope exceed the threshold. The second method is developed by Pan and Tompkins. American Heart Association (AHA) ECG data files are used to test these two QRS detection methods. Our results show that the `So and Chan' method performs better than the `Pan and Tompkins' method. Further development continues on the `So and Chan' method. Based on the information of the identified QRS complexes, the P waves and the T waves are also be detected. ECG classification is then carried out using the autocorrelation. The classification algorithm is trained to recognize four types of beat.