Atrial fibrillation ablation using three-dimensional artificial intelligence module integration with intracardiac...

Area of Science:

• Cardiology
• Medical Technology
• Artificial Intelligence in Medicine

Background:

• Intracardiac echocardiography (ICE)-based electroanatomical mapping (EAM) is standard for atrial fibrillation (AF) ablation.
• The CARTOSOUND FAM (AI FAM) module automates left atrial reconstruction using deep learning, eliminating manual contouring.

Purpose of the Study:

• To evaluate the one-year outcomes of AI FAM compared to standard EAM in AF ablation.
• Assess procedural efficiency and long-term efficacy of AI FAM.

Main Methods:

• A comparative study of 298 patients undergoing radiofrequency AF ablation (Jan 2021-Dec 2023).
• Patients before Jan 2023 used standard EAM; 2023 patients used AI FAM-based reconstruction.
• Analysis included demographics, procedural data, acute success, complications, and one-year AF recurrence.

Main Results:

• AI FAM (115 patients) vs. EAM (183 patients) showed comparable baseline characteristics.
• AI FAM significantly reduced mean total procedure time (122.5 vs. 129.0 min) and left atrial dwell time (78.3 vs. 87.5 min).
• Acute success rates were similar (98.3% vs. 98.9%), with fewer complications in the AI FAM group (1 vs. 4). One-year AF recurrence was comparable (80.0% vs. 81.4%).

Conclusions:

• AI FAM offers significant procedural advantages, reducing procedure and LA dwell times.
• AI FAM does not compromise acute or long-term safety and rhythm control efficacy.
• Integration of AI FAM with PFA promises more streamlined AF ablation procedures.