Multi-lead T wave end detection based on statistical hypothesis testing
Abstract
Automatic detection of electrocardiograms (ECG) waves provides important information for cardiac disease diagnosis. A new T wave end location algorithm based on multi-lead ECG processing is proposed in this paper. A statistical hypothesis testing algorithm is applied to two auxiliary signals computed by filtering and differentiating ECG signals. The performance of the algorithm has been evaluated using the PhysioNet QT database. The standard deviation of the errors between automatic annotations and manual ones are within tolerance accepted by cardiologist.