ACR Provides Input to FDA on AI Measurement and Evaluation

ACR® filed comments with the FDA in response to a request for comment regarding measuring and evaluating the performance of AI-enabled medical devices in real-world settings. This input will help inform FDA’s ongoing efforts to gather feedback on how to strengthen oversight of emerging AI technologies, including generative AI-enabled devices.  

In its comments, ACR highlighted several ACR Data Science Institute® programs designed to support safe and effective implementation of AI in clinical settings, including: 

• Assess-AI - A national quality registry that enables clinical sites to continuously monitor AI, reduce its risk, and optimize its performance. 
  
• ARCH-AI - The ACR Recognized Center for Healthcare-AI program, a national quality assurance initiative that recognizes best practices for AI use in imaging interpretation. 

ACR stressed the importance of local acceptance testing, governance and continuous monitoring. ACR also provided FDA with information on imaging AI performance metrics, evaluation methods and infrastructure, postmarket data sources and quality management, monitoring triggers and response protocols, and the importance of human-AI interaction and user experience.

For questions about ACR’s comments to the FDA, contact Michael Peters, ACR Senior Director, Government Affairs, or Lindsay Robbins, ACR Regulatory Policy Specialist.