An automatic threshold-based ECG wave boundaries detector achieved correct T wave morphology identification and boundary location variance within manual referees' variance in 71% of records.
Does a single-lead threshold-based ECG wave boundaries detector accurately identify T wave morphology and boundary location compared to manual referees in the QT database?
A single-lead threshold-based ECG wave boundaries detector correctly identified T wave morphology and boundary location in 71% of records from the QT database.
The authors evaluate a single-lead threshold based ECG wave boundaries detector with a QT database developed for validation purposes. They also identify its different sources of error distinguishing those that come from precision errors in boundary location from those that come from morphology misclassification. They obtain 71 % of records with correct morphology identification of T wave and variance in boundary location within manual referees variance. The remaining records analyzed correspond to signals with poor SNR at the T wave, or morphology discrepancies between algorithm and experts.
Jané et al. (Fri,) conducted a other in ECG waveform limits detection. Single-lead threshold based ECG wave boundaries detector vs. Manual referees was evaluated on Correct morphology identification of T wave and variance in boundary location within manual referees variance. An automatic threshold-based ECG wave boundaries detector achieved correct T wave morphology identification and boundary location variance within manual referees' variance in 71% of records.