Marketing submission recommendations for prescribed change control plans for artificial intelligence/machine learning (AI/ML) enabled device software features

Machine Learning

who submitted the comment

You may comment on any guidance at any time (see 21 CFR 10.115(g)(5)), but please ensure that your comments on the draft guidance are considered before FDA begins work on the final version of the guidance. Please submit online to do so. or written comments on the draft guidance by the deadline.

If you are unable to submit your comments online, please mail your written comments to:

docket management
Food and Drug Administration
5630 Fishers Lane, Rm 1061
Rockville, Maryland 20852

All written comments should be identified with this document’s docket number: FDA-2022-D-2628

Docket number:
issued by:

Guidance issuing office

Medical Equipment and Radiation Health Center

Center for Biologics Evaluation and Research

Pharmaceutical Evaluation Research Center

Commissioner’s Office, Clinical Policy and Programs Office, Combination Products Office

FDA is issuing this draft guidance to further develop regulatory approaches tailored to artificial intelligence/machine learning (AI/ML)-enabled devices and increase patient access to safe and effective AI/ML-enabled devices. and aims to protect and promote the public. health. This draft guidance continues to ensure the safety and effectiveness of machine learning-enabled device software functions (ML-DSFs), as described in the FDA’s 2019 AI/ML Discussion Papers and 2021 AI/ML, while continuing to It describes the least burdensome approach to supporting iterative improvement. Action plan. The Draft Guidance provides recommendations regarding information to be included in a Predetermined Change Control Plan (PCCP) that may be provided in marketing submissions for ML-DSFs. The PCCP mechanism includes planned ML-DSF changes, associated methodologies for implementing and validating those changes, and assessment of the impact of those changes. Since the introduction of the PCCP concept, there has been great interest in using this mechanism for AI/ML-enabled medical devices. The FDA continues to increase the number of marketing submissions and pre-submissions for AI/ML-enabled medical devices that have the potential to have a significant positive impact on healthcare, and the agency expects this to increase over time. This guidance builds on the FDA’s longstanding commitment to developing and applying innovative approaches to regulation to ensure the safety and efficacy of medical device software and other digital health technologies. The recommendations in this guidance apply to device components of combination products, such as drug-medical device and biological device combination products, when the medical device component contains an ML-DSF.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *