Mapping the Dynamics of Atrial Fibrillation with Spatiotemporal Graph Neural Networks

Alexander Jenkins, Andrea Cini, Joseph Barker, Danilo Mandic, Cesare Alippi, Fu Siong Ng

Keywords: Atrial Fibrillation, electrophysiological mapping, catheter ablation, graph neural network, recurrent neural network, spatiotemporal, imputation

TL;DR: A graph recurrent neural network models atrial fibrillation dynamics, transforming sparse sequential contact measurements into complete global maps via spatiotemporal imputation, providing a pathway toward personalised ablation.

Abstract: Catheter ablation of persistent Atrial Fibrillation (AF) consists of a one-size-fits-all treatment with limited success. This may be due to our inability to map AF electrical dynamics on the atrial surface with the limited resolution and coverage provided by sequential contact mapping catheters, preventing effective patient phenotyping for personalised, targeted ablation. In this proof-of-concept study, we introduce FibMap, a graph recurrent neural network model that reconstructs global AF dynamics from sparse measurements. Trained and validated on 51 non-contact whole atria recordings, FibMap reconstructs whole atria dynamics from 10% surface coverage, achieving a 47% lower mean absolute error and a 109% higher performance in recovering phase singularities compared to baseline methods. Clinical utility of FibMap is demonstrated on real-world contact mapping recordings, achieving reconstruction fidelity comparable to non-contact mapping. Integrating FibMap into clinical practice could enable personalised AF care and improve outcomes.

Submission Number: 121