FDA Releases Draft Guidance on Lifecycle Management for AI-Enabled Device Software Functions

The U.S. Food and Drug Administration (FDA) has issued a draft guidance aimed at strengthening the regulatory framework for Artificial Intelligence-Enabled Device Software Functions (AI-DSFs). This guidance provides comprehensive recommendations for manufacturers on lifecycle management and marketing submissions for AI-enabled devices, ensuring that they meet regulatory requirements while maintaining high standards of safety and effectiveness.

The draft guidance emphasizes a Total Product Life Cycle (TPLC) approach, highlighting the need for manufacturers to address transparency and mitigate bias throughout the development and marketing phases. This approach ensures that AI-enabled devices remain effective for diverse patient populations, meeting the needs of all demographic groups, including race, ethnicity, sex, and age. Additionally, the guidance encourages the adoption of FDA-recognized consensus standards to improve the quality and consistency of submissions.

Key recommendations include providing detailed documentation for FDA review, implementing strategies to address AI-related risks, and incorporating machine learning practices that support high-quality, safe devices. The FDA urges early engagement with the agency for manufacturers developing combination products that incorporate AI-DSFs, ensuring that all relevant regulatory considerations are addressed early in the process.

Manufacturers are advised to incorporate these lifecycle management practices throughout the development and submission stages to align with FDA expectations and promote the creation of safe, effective, and high-quality AI-enabled medical devices.

For more information on the draft guidelines, read the full document below.