The latest edition of Ellegaard Newsletter contains an article on the Challenges to detecting the end of cardiac repolarisation in the Göttingen minipig ECG.
This publication describes how our ECG analyzer ECG30s detects and marks PQRST points on ECG with changing T wave morphology. Sensitivity and predictivity results are expressed for the automated positioning of these fiducial points.
Authors: Florence Koeppel1, Michael Markert2, David Labarre2 and Philippe Zitoun2
1: NOTOCORD Systems, France
2: Boehringer Ingelheim Pharma GmbH & Co KG, Germany
Minipigs exhibit characteristic and unique ECG features. In particular, the morphology of the T wave, which corresponds to cardiac repolarization, is often observed to change on a beat-to-beat basis.
This variability represents a challenge for automated ECG analysis, especially for defining the timing of the T wave start and end.
A new software application named ECG30s was developed specifically for automated analysis of ECGs with changing T wave morphology.
This study was designed to determine the accuracy of P, Q, R, S and T points positioning with this new software. To that end, a signal database was built with 17 ECG recordings, mainly from freely moving Göttingen minipigs. Positioning of fiducial points with the automated software was compared to positioning done manually.
Results show a sensitivity of 80.3%, 85.5%, 97.3%, 89.1% and 80.7%, and a predictivity of 78.7%, 83.8%, 95.3%, 84.0% and 81.4% for P, Q, R, S and T points placement, respectively.
In conclusion, ECG30s ensured consistent detection of T end even with changing T wave morphology.
► Read the whole article