Bridging Regulatory Science and Medical Imaging Evaluation
While there is a push for regulatory organizations to establish standards and guidelines for new artificial intelligence (AL) technologies (e.g., generative Al, autonomous Al, continual learning), formulating effective regulations is challenging without rigorous evaluation of these innovations; the principle "if we cannot evaluate it, we cannot regulate it" highlights this dependency. Thus, regulation and evaluation are intricately connected.The lack of robust evaluation frameworks introduces both challenges and opportunities within regulatory science, a field dedicated to developing tools and methodologies to assess and monitor the performance of Al-enabled or Al-powered medical devices.
This workshop aims to introduce Al regulatory science to the MICCAI community and provide a platform for researchers, clinicians, and regulators to discuss algorithmic, technical, and policy challenges and collaboratively develop rigorous solutions for the safe and effective deployment of Al in healthcare. In addition, this workshop will provide an educational platform to highlight the often-overlooked importance of regulatory aspects and evaluation in AI research.