کنفرانس مهندسی برق ایران

صفحه اصلی / سی و دومین کنفرانس بین المللی مهندسی برق

Accurate Methods for Automatic Detection of Characteristic Points in Electrocardiograms

نویسندگان :

Seyedeh Mersedeh Bagheri¹ Mohammad Pooyan²

1- دانشگاه شاهد تهران 2- دانشگاه شاهد تهران

کلمات کلیدی :

ECG،characteristic points detection،R peak،T wave،S peak،J point

چکیده :

The electrocardiogram (ECG) is a measurement and recording of the electrical signals associated with the heart. Automatic identification of ECG characteristic points (fiducial points) is essential for cardiac disease diagnosis. The characteristic points which are studied in this paper, include R peak, iso-electric level, S peak, J point, P peak and onset, offset and peak of the T wave. We attempted to employ an approach with acceptable accuracy to detect each of the ECG characteristic points. The proposed method to detect R peak consists of a preprocessing stage based on a Shannon energy envelope technique and peak-finding logic stage using Hilbert transform. S peak and J point detection are performed during the process of calculation the angles of the points after R peak. This paper represents a sliding area which is covered by the T wave curve to determine offset of the T wave, and then peak and onset of the T wave could be identified by offset of the T wave. P peak detection is almost similar to T peak detection. The characteristic points detection is validated using European ST-T database, Long Term ST database, and MIT-BIH Normal Sinus Rhythm database, and total accuracy of 98.26 % is achieved.