QRS Detection and PVC Beat Recognition Using a Generalised Teager Energy Operator

-This paper presents a novel algorithm for automatic QRS complex detection and premature ventricular contraction (PVC) beat recognition based on a generalised Teager energy operator (GTEO). The algorithm is divided into two stages: QRS detection and PVC beat recognition. An optimal GTEO order is determined for each stage. A second order GTEO is used for QRS detection, and a seventh order GTEO is used for PVC beat recognition. The proposed algorithm was tested using ECG signals from two recognised arrhythmia databases, the MIT-BIH and the AHA database. The signals chosen contained PVC beats as well as normal beats. Sensitivity and specificity parameters were used to measure the accuracy of the proposed algorithm. The main advantages of using a GTEO are simplicity, robustness and speed. The sensitivities achieved using the proposed algorithm were 99.5% for QRS detection and 97.4% for PVC recognition. The specificities achieved were 99.8% for QRS detection and 99.1% for PVC beat recognition. Key-Words: Teager Energy, ECG, QRS Detection, PVC Recognition, SA Node, Purkinje Fibres.