The US Food and Drug Administration wants drug sponsors to follow a seven-step framework for assessing and establishing the credibility of artificial intelligence models when their outputs are used to make regulatory decisions.
Artificial Intelligence: US FDA Outlines 7 Steps To Establishing Model Credibility
The risk-based framework described in a new draft guidance starts with defining the question of interest and context of use and includes development and execution of a credibility assessment plan. The guidance is limited to AI models used to support regulatory decisions about drug safety, effectiveness or quality.
More from AI
More from Advanced Technologies
• By
A new chapter that will be published in the European Pharmacopoeia soon is expected to provide a comprehensive framework for the production and quality control of cell-based preparations. Meanwhile, an existing chapter has been revised to facilitate the implementation of rapid microbiological testing methods.
• By
India is pressing ahead with “big shifts” in its regulatory approach for cell and gene therapies. Reduced “layers of review”, tighter internal timelines, upcoming guidelines to facilitate CGT development and reforms in the constitution of expert committees are some of the key changes.
• By
Cellares’ fully automated cell therapy manufacturing platform is the first system to receive an Advanced Manufacturing Technology designation from US FDA