Evaluation of Multiscale Frequency Approach for Visualizing Rotors in Patients with Atrial Fibrillation

Atrial Fibrillation (AF) is most common cardiac arrhythmia. It is associated with increased risk of stroke, heart failure and sudden cardiac death. Catheter ablation is a treatment used to control AF and has had suboptimal success for patients with persistent AF, which is primarily maintained by rotors outside of the pulmonary veins (PV) region. The pivot point (core) of the rotor is considered an efficient target for ablation. Currently available electro-anatomical mapping systems cannot accurately predict the exact location of the pivot point of rotors outside of the PV region, so there is a need for novel approaches to accurately identify and distinguish sites for ablation. Recently, a multiscale frequency (MSF) technique was developed for accurate identification of the pivot point of rotors and validated using optical mapping experiments in exvivo rabbit hearts, where electrical activity can be directly visualized. However, the nature of optical signals and its spatial resolution are very different from clinical intracardiac electrograms (iEGM). Here we extend the MSF approach to 3D iEGM and compare its prediction with the traditional dominant frequency (DF) approach, using Pearson's correlation and earth mover's distance methods. Our results demonstrate that the similarity between MSF and DF are high in some regions, but very low in other spatial regions of the human atria. This indicates the inconsistency in the traditional DF approach in identifying pivot points and identifying such low similarity regions can be used to find sites for successful ablation.