Artificial Intelligence-Based Evaluation of Post-Procedural Electrocardiographic Parameters to Identify Patients at Risk of Atrial Fibrillation Recurrence After Transcatheter Ablation

<b>Background/Objectives:</b> Arrhythmic recurrence is a common issue affecting a significant percentage of patients undergoing transcatheter ablation (TCA) of Atrial Fibrillation (AF). The use of artificial intelligence (AI) for the identification of electrocardiographic predictors of post-ablation recurrence may offer a valuable and cost-effective approach to improve risk stratification and optimize follow-up. This study aims to investigate the relationship between post-procedural electrocardiographic (ECG) P-wave parameters, measured using AI, and AF recurrence in patients undergoing transcatheter ablation (TCA). <b>Methods:</b> Seventy-four patients (age 62.36 ± 10.4 years) with a diagnosis of AF were retrospectively analyzed. ECGs were processed using AI software to analyze P-wave-related variables. All patients had either an implantable loop recorder (ILR) or another form of cardiac implantable electronic device (CIED). <b>Results:</b> Post-procedural P-wave amplitude in lead II (PwA in lead II) showed a significant association with AF recurrence, defined as an average arrhythmic burden >6% at one-year follow-up. <b>Conclusions:</b> These findings underscore the potential of PwA in lead II as a biomarker for the follow-up of patients undergoing TCA and highlight the contribution of AI in the analysis of electrocardiographic parameters predictive of AF recurrence. Together, these results may contribute to the development of early risk-stratification strategies following catheter ablation.