Automatic Classification of 12-, 6-, 4-, 3-, and 2-Lead Electrocardiograms Using Morphological Feature Extraction

Alexander Hammer1, Matthieu Scherpf2, Hannes Ernst1, Jonas Weiß1, Daniel Schwensow1, Martin Schmidt1
1TU Dresden, 2TU Dresden, Institute of Biomedical Engineering, Dresden, Germany


Cardiovascular diseases are the global leading cause of death. Automated electrocardiogram (ECG) analysis can support clinicians to identify abnormal excitation of the heart and prevent premature cardiovascular death. An explainable classification is particularly important for support systems. Our contribution to the PhysioNet/CinC Challenge 2021 (team name: ibmtPeakyFinders) therefore pursues an approach that is based on interpretable features to be as explainable as possible.

To meet the challenge goal of developing an algorithm that works for both 12-lead and reduced lead ECGs, we processed each lead separately. We focused on signal processing techniques based on template delineation that yield the template’s fiducial points to take the ECG waveform morphology into account. In addition to beat intervals and amplitudes obtained from the template, various heart rate variability and QT interval variability features were extracted and supplemented by signal quality indices. Our classification approach utilized a decision tree ensemble in a one-vs-rest approach. The model parameters were determined using an extensive grid search.

Our approach achieved challenge scores of 0.47, 0.47, 0.40, 0.43, and 0.45 on 12-, 6-, 3-, 4-, and 2-lead validation sets, respectively.