Nonnegative Matrix Factorization for Noninvasive Prediction of Catheter Ablation Outcome in Persistent Atrial Fibrillation

Abstract : Despite the increasing popularity of catheter ablation (CA) for treat- ing atrial fibrillation (AF), the identification of patients who would actually benefit from the therapy remains a challenging open issue. This study aims at noninvasively predicting CA outcome by quanti- fying the spatio-temporal variability of the atrial activity signal mea- sured on the standard 12-lead electrocardiogram. The normalized mean square error (NMSE) between consecutive atrial segments and their principal component approximations is computed for each lead, as a recent noninvasive index of AF organization. In the present work, the multilead NMSE array is decomposed by means of a non- negative matrix factorization (NNMF) with two different initializa- tions. The reconstruction error between the original NMSE matrix and its low-rank NNMF approximation is taken as a classification feature. A dataset of persistent AF patients undergoing CA reveals that the proposed feature is able to predict the therapy's outcome with a notably higher level of statistical significance than recent single-lead indices.
Type de document :
Communication dans un congrès
Annual International Conference of the IEEE Acoustics, Speech, and Signal Processing, ICASSP, 2012, Japan. 1, pp.601-604, 2012
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00655798
Contributeur : Olivier Meste <>
Soumis le : lundi 2 janvier 2012 - 14:54:52
Dernière modification le : vendredi 6 juillet 2012 - 17:57:12

Identifiants

  • HAL Id : hal-00655798, version 1

Collections

Citation

Marianna Meo, Vicente Zarzoso, Olivier Meste, Decebal Gabriel Latcu, Nadir Saoudi. Nonnegative Matrix Factorization for Noninvasive Prediction of Catheter Ablation Outcome in Persistent Atrial Fibrillation. Annual International Conference of the IEEE Acoustics, Speech, and Signal Processing, ICASSP, 2012, Japan. 1, pp.601-604, 2012. <hal-00655798>

Partager

Métriques

Consultations de la notice

142