Robustness evaluation of computer-aided clinical trials for medical devices

摘要: Medical cyber-physical systems, such as the implantable cardioverter defibrillator (ICD), require evaluation of safety and efficacy in context a patient population clinical trial. Advances computer modeling simulation allow for generation simulated cohort or virtual which mimics can be used source prior information. A major obstacle to acceptance results information is lack framework explicitly sources uncertainty quantifying effect on trial outcomes. In this work, we formulate Computer-Aided Clinical Trial (CACT) within Bayesian statistical allowing explicit assumptions utilization at all stages To quantify robustness CACT outcome with respect assumption, define δ-robustness minimum perturbation base distribution resulting change provide method estimate δ-robustness. We demonstrate utility how utilized various through an application Rhythm ID Goes Head-to-head (RIGHT), was comparative specific software algorithms across different cardiac devices. Finally, introduce hardware interface that allows direct interaction physical device order validate confirm