US FDA's AI Framework for Medical Devices

Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing the healthcare landscape, driving innovation in medical device development. The ability of AI/ML software to learn from real-world feedback and enhance its performance has caught the attention of the FDA, leading to a paradigm shift in regulatory expectations and device design standards.

The FDA acknowledges the immense potential and risks associated with AI in medical devices. It has taken proactive steps to define industry expectations, addressing crucial aspects like device design, programming intricacies, unique validation challenges, documentation requirements, and submission protocols.

While many AI technologies authorized by the FDA are "locked" algorithms with static functionalities, the agency is increasingly focusing on "adaptive" or "continuously learning" algorithms. These advanced algorithms evolve and learn from real-time user data, enhancing their capabilities over time.

The FDA is exploring frameworks to facilitate real-world learning and adaptation in AI algorithms while ensuring the safety and effectiveness of Software as a Medical Device (SaMD). This includes evaluating algorithm performance, verifying and validating AI/ML software, outlining manufacturers' modification plans, and managing associated risks through predetermined change control strategies.

WHY YOU SHOULD ATTEND:

The US FDA is advancing a fresh regulatory framework aimed at fostering safe and efficient medical devices leveraging advanced artificial intelligence (AI) and machine learning (ML) algorithms. These algorithms, capable of learning from and responding to data, are already revolutionizing disease screening and treatment recommendations. Notably, the FDA has authorized AI-based devices for detecting diabetic retinopathy and alerting providers to potential strokes, marking significant progress in device safety and performance enhancement.

As more medical devices integrate advanced AI algorithms, the FDA is implementing new AI software validation requirements. This proactive approach aligns with the agency's commitment to adapting regulatory strategies, ensuring device safety amidst rapid innovation in the healthcare sector. Join us to explore the transformative impact of AI in medical device development and navigate the evolving regulatory landscape to stay ahead in healthcare innovation.

AREAS COVERED IN THE SESSION:

• The FDA's AI "Framework" for AI in Medical Devices
• Roles of Verification and Validation
• IEC 62304 and an FDA Software Guidance
• FDA AI device submission requirements
• A Typical Software V&V Protocol / Test Report; "Black" and "White" box
• "Locked" vs ML algorithms
• Predetermined Change Control in AI
• Expected Regulatory Submission Deliverables
• The Future of AI in Medical Devices

WHO SHOULD ATTEND: 

• Quality Assurance Departments
• Engineering Departments
• Regulatory Affairs Departments
• Research and Development Departments
• Manufacturing Departments
• Operations Departments
• Production Departments
• Risk Management Professionals

Course Director: JOHN E. LINCOLN  

John E. Lincoln, is Principal of J. E. Lincoln and Associates LLC, a consulting company with over 36 years experience in U.S. FDA-regulated industries, 22 of which are as an independent consultant. John has worked with companies from start-up to Fortune 100, in the U.S., Mexico, Canada, France, Germany, Sweden, China and Taiwan. He specializes in quality assurance, regulatory affairs, QMS problem remediation and FDA responses, new / changed product 510(k)s, process / product / equipment QMS and software validations, ISO 14971 product risk management files / reports, Design Control / Design History Files, Technical Files, CAPA systems and analysis. He’s held positions in Manufacturing Engineering, QA, QAE, Regulatory Affairs, to the level of Director and VP (R&D). In addition, John has prior experience in military, government, electronics, and aerospace. He has published numerous articles in peer reviewed journals, conducted workshops and webinars worldwide on CAPA, 510(k)s, risk analysis / management, FDA / GMP audits, validation, root cause analysis, and others. He writes a recurring column for the Journal of Validation Technology. John is a graduate of UCLA.