Digital-twin-based Online Parameter Personalization for Implantable Cardiac Defibrillators

摘要

This paper proposes a reinforcement-learning framework for online personalization of implantable cardioverter defibrillator parameters. A patient-specific digital twin inferred from ECG data is used as the training environment, allowing the system to adapt therapy settings to evolving heart conditions and outperform default parameter choices in virtual-patient experiments.