ECG image representation of normal sinus rythm

This paper deals with a new image representation for ECG signal. The method consists in decomposing the ECG signal in beats. The ECG segmentation is based on the time location of the peak R of the QRS complex. A matrix parameterized by QRS rank comprises the signal tranches around this time reference. The matrix forms an image that illustrates the ECG signal characteristics to replace the long ECG record signal. This representation is very useful for diagnosis in both normal and heart diseases cases. The different ECG records are given from the Arrhythmias database. In this work, we focus on the normal sinus rhythm to show the performances of this representation.