Persistent atrial fibrillation remains one of cardiology's most stubborn therapeutic puzzles, and the gap between promising technology and clinical outcomes has rarely felt wider. A mechanistic framework that personalizes ablation planning — rather than applying blanket tissue destruction — could meaningfully shift the risk-benefit calculus for the millions of patients in whom current catheter techniques repeatedly fail.
The review centers on applying personalized cardiac digital twins (DTs) to persistent AF (PsAF), the fibrotic, structurally remodeled form of the arrhythmia that defies the pulmonary vein isolation strategy effective in paroxysmal AF. Digital twins reconstruct patient-specific atrial electrophysiology from clinical imaging and electroanatomic data, enabling computational simulation of reentrant wavefront behavior and identification of locations of reentry (LRs) — the arrhythmogenic substrate sites believed to sustain fibrillatory conduction. The approach aims to predict arrhythmia inducibility pre-procedurally and guide targeted, minimal-footprint ablation rather than extensive substrate modification, which carries risks including scar-related atrial tachycardia and degraded atrial mechanical function.
Digital twin modeling in cardiac electrophysiology has accelerated considerably over the past decade, with groups including those at King's College London and IHU Liryc demonstrating feasibility in small cohorts. The concept addresses a genuine mechanistic blind spot: conventional intracardiac electrogram mapping approaches — CFAE, dominant frequency, and rotor mapping — have produced inconsistent clinical trial results partly because they cannot fully account for the three-dimensional, patient-specific architecture of fibrotic remodeling. The digital twin framework shifts the paradigm from intraoperative empirical mapping toward pre-procedural computational prescription. Key limitations remain substantial, however: computational models require validation in prospective randomized trials, processing pipelines are not yet clinically scalable, and the field lacks consensus on which biophysical model parameters best represent human atrial fibrosis. This work is best characterized as a thoughtful mechanistic roadmap rather than practice-changing evidence — incremental but directionally important for precision electrophysiology.