Project R-15061

Integrating Artificial Intelligence in electrophysiology: Determining optimal treatment paths in atrial fibrillation ablation. (Research)

The integration of artificial intelligence (AI) in electrophysiology, specifically in managing atrial fibrillation (AF), marks a significant advancement in precision medicine. This PhD thesis investigates the potential of advanced AI technologies, such as convolutional neural networks, to improve diagnostic accuracy and refine therapeutic strategies for AF treatment. The study aims to evaluate AI's role in enhancing AF diagnosis by analyzing electrocardiographic (ECG) data and correlating it with voltage mapping from electrophysiological studies. The goal is to accurately identify areas of low voltage, such as atrial fibrosis, associated with AF risk. Additionally, the research explores the correlation between AI-detected atrial scar tissue or fibrosis and the choice of ablation strategies, comparing standard pulmonary vein isolation (PVI) with a more comprehensive PVI+ approach. The thesis also considers how AI algorithms can predict AF recurrence postoperatively, using less commonly analyzed variables like atrial extrasystoles, to identify patients at higher risk. These results may be used in the management of anticoagulation therapy, particularly in younger patients with a low CHA2DS2-VASC score, to inform clinical decisions on therapy continuation or discontinuation.

16 April 2024 - 15 April 2028