We propose a new method for the classification task of distinguishing atrial Fibrillation (AFib) from regular atrial tachycardias including atrial Flutter (AFlu) on the basis of a surface electrocardiogram (ECG). Although recently many approaches for an automatic classification of cardiac arrhythmia were proposed, to our knowledge none of them can distinguish between these two. We discuss reasons why deep learning might not yield satisfactory results for this task. We generate new and clinically interpretable features using mathematical optimization for subsequent use within a machine learning (ML) model. These features are generated from the same input data and by solving an additional regression problem with complicated combinatorial substructures. The resulting model can thus also be seen as a completely novel ML model that incorporates expert knowledge on the pathophysiology of AFlu. Our approach achieved an unprecedented accuracy of 82.84% and an excellent area under the ROC curve of 0.9. One additional advantage of our approach is the inherent interpretability of the classification results. Our features give insight into a possibly occurring multi-level atrioventricular blocking mechanism, which might improve treatment decisions beyond the classification itself. Our research ideally complements existing cardiac arrhythmia classification methods from the literature, which can provide a preclassication, but so far left the important case AFib vs AFlu open.
Citation
submitted to Artificial Intelligence in Medicine
Article
View Expert-Enhanced Machine Learning for Cardiac Arrhythmia Classification