A new method for electrocardiogram study

A new method for the identification of electrocardiogram (ECG) waves and the study of atrial activity is discussed. It was designed to improve the performance of an automatic ECG interpretation program. The P wave detector uses an 80 ms bidirectional energy collector; it is computed for TQ intervals only. The QRS complex detector is based on the Function of Spatial Velocity; it was modified to improve the QRS frontiers identification. The T-wave offset was defined as the minimum difference between the ECG signal and the isoelectric line, this process was made after T-wave peak was reached. Five-hundred ECGs were recorded with two digital electrocardiographs; patients were chosen randomly at a Cardiology Service. Two cardiologists studied each ECG using a software with graphic tools, they identified and measured each wave or signal interval; their criteria were used as a golden rule for event identification. The sensitivity values were 97.12 % for P wave detection; 99.38 % for T-wave offset identification and 100 % for QRS complex detection. The algorithms developed are simple and satisfactory to be implemented in low speed processor platforms because the use only integer arithmetic with low execution time requirements.