We propose a new Cardiac Electrical Sparse Imaging (CESI) technique to image cardiac activation throughout the three-dimensional (3D) ventricular myocardium from body surface electrocardiogram (ECG) with the aid of heart-torso geometry. Sparse property of cardiac electrical activity along time is utilized in the temporal sparse inverse problem. Computer simulations were carried out to evaluate the performance of this imaging method under various circumstances. A total of 12 single site pacing simulations with various noise/disturbance were employed to evaluate the accuracy, stability and robustness of the technique. The imaged activation sequence is compared with the simulated ones and those from the conventional weighted minimum norm method. The results show that CESI outperformed the conventional method. In summary, we have proposed a novel method for cardiac activation imaging, and our results suggest that CESI has enhanced performance and has the potential to image ventricular activation and assist in clinical management of ventricular arrhythmias.
|Original language||English (US)|
|Number of pages||4|
|Journal||Computing in Cardiology|
|State||Published - Jan 1 2014|
|Event||41st Computing in Cardiology Conference, CinC 2014 - Cambridge, United States|
Duration: Sep 7 2014 → Sep 10 2014