DICTIONARY LEARNING FOR THE SPARSE MODELLING OF ATRIAL FIBRILLATION IN ECG SIGNALS

Boris Mailhé (1) , Rémi Gribonval (1) , Frédéric Bimbot (1) , Mathieu Lemay (2) , Pierre Vandergheynst (3) , Jean-Marc Vésin (2)

1 METISS - Speech and sound data modeling and processing
2 SP Lab - Laboratoire de Traitement du signal [EPFL] / Signal Processing Laboratories
3 LTS2 - EPFL

We propose a new method for ventricular cancellation and atrial modelling in the ECG of patients suffering from atrial fibrillation. Our method is based on dictionary learning. It extends both the average beat subtraction and the sparse source separation approaches. Experiments on synthetic data show that this method can almost completely suppress the ventricular activity, but it generates some artifacts. Contrary to other ventricular cancellations methods, our approach also learns a model for the atrial activity.

Mots clés: ECG; atrial fibrillation; monochannel source separation; dictionary learning; sparse approximation; K-SVD

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Boris Mailhé, Rémi Gribonval, Frédéric Bimbot, Mathieu Lemay, Pierre Vandergheynst, et al.. DICTIONARY LEARNING FOR THE SPARSE MODELLING OF ATRIAL FIBRILLATION IN ECG SIGNALS. ICASSP 2009, Apr 2009, Taipei, Taiwan. pp.465 - 468, ⟨10.1109/ICASSP.2009.4959621⟩. ⟨hal-00466973⟩