Contributions for the optimal lead placement for the study of atrial fibrillation applying Independent Component Analysis to 64 body surface potential mapping recordings

The interest in the study and analysis of atrial fibrillation has increased significantly in the last decades. A correct estimation of the atrial activity is a crucial previous step for AF analysis. Different methods based on independent component analysis of 12-lead ECG have been proposed. However, they don´t work for a reduced number of leads. The purpose of this study was (1) search of new spatial information of atrial fibrillation waves by means of multi-lead recordings and (2) to quantify the contribution of the estimated atrial activity (AA) on each lead in order to identify the optimal lead placement for successful extraction. Principal component analysis and second order blind identification methods were applied for AA estimation and identification. Contribution of each lead was assessed. A reduced set of leads which allows AA extraction was identified, located (1) to right of VI and (2) below the left precordial zone.