AI, ML, V and V and FDA Submission Framework

The artificial intelligence technologies granted FDA marketing authorization and cleared by the agencypreviously are generally called “locked” algorithms that don’t continually adapt or learn every time the algorithm is used. However, the FDA is looking beyond these elemental devices, to those capable of true AI, -  machine learning algorithms that continually evolve, often called “adaptive” or “continuously learning” algorithms. Adaptive algorithms can learn from new user data presented to the algorithm through real-world use. The FDA is exploring a framework to allow modifications to algorithms to be made from real-world learning and adaptation, while still ensuring safety and effectiveness of the software as a medical device (SaMD)  is maintained, without requiring frequent updates or submissions.

This webinar will discuss information specific to devices that include artificial intelligence algorithms that make real-world modifications that the Agency would require for premarket review. They include the algorithm’s performance, the added concerns for AI / ML software verification and Validation,  the manufacturer’s plan for modifications and the ability of the manufacturer to manage and control risks of the modifications, including the software’s "predetermined change control plan" (PCCP).

• 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

The ability of artificial intelligence / machine learning software to learn from real-world feedback and improve its performance is spurring innovation and leading to the development of novel medical devices. The FDA recognizes AI's advantages and dangers and has started to outline its expectations of industry and device design, programming, unique validation issues, documentation and submissions.

The US FDA has defined the steps needed for the regulatory framework for the development of safe and effective medical devices that use advanced artificial intelligence / machine learning algorithms. Artificial intelligence algorithms are software that can learn from and act on data. These types of algorithms are already being used to aid in screening for diseases and to provide treatment recommendations. The authorization of these technologies was a harbinger of progress that the FDA expects to see as more medical devices incorporate advanced artificial intelligence algorithms to improve their performance and safety.

Since AI changes over time, how will the Agency control these changes without requiring newmarketing submissions. AI software validation has some new requirements as well. The Agency plans to apply their current authorities in new ways to keep up with the rapid pace of innovation and ensure the safety of these devices.

This webinar will provide valuable assistance to all regulated companies in evaluating their existing or anticipated AI medical device R&D and production activities and documentation requirements, as currently anticipated by the US FDA. This information applies to personnel / companies in the  Medical Device and some Combination products fields. The employees who will benefit include: 

• Software engineering
• Senior management
• Regulatory Affairs
• Quality Assurance / QAE
• Production
• Engineering, R&D, and software development and testing teams

All personnel involved in a U.S. FDA-regulated environment.  Especially those involved in new product development with products and productioin / test equipment that are software driven, companies with quality management / MRP / ERP systems moving toward "paperless" systems, or other regulated medical products / CGMP and ISO QMS activities.