Blog Post on Device AI Lifecycle Model
A new blog post from FDA’s Digital Health Center of Excellence (DHCoE) outlines key activities that should be taking place during the lifecycle of artificial intelligence-enabled medical devices (AI lifecycle).
The post provides an AI lifecycle (AILC) concept diagram that highlights systematic methods related to data and model evaluation during data collection and management, and model building and tuning phases. “This diagram also illustrates monitoring AI software post-deployment in operation and monitoring and real-world performance evaluation phases,” it says. “One possible use of this AILC model is as a guide, or playbook, to help assess standards, tools, metrics, and best practices for the considerations identified” in each phase.
For example, DHCoE says, under its “data suitability” element “one could work to identify the relevant standards and applicable metrics, like data quality, population coverage, and provenance. Additionally, one could explore operational tools for tasks such as data preprocessing, augmentation, bias detection.”
Additionally, DHCoE says standards play a role in the AI lifecycle by “helping to ensure quality, facilitate interoperability, and promote ethical practices. They also help guide development, enhance transparency, support compliance, encourage innovation, and build trust.” It says that its AI lifecycle concept will help “spur development of other activities,” such as:
- Creating a comprehensive checklist to aid developers using AI as a medical device and in healthcare solutions.
- Establishing a robust foundation for developing AI models rooted in high-quality, reliable, and ethically sound data and AI practices.
- Developing a systematic approach for evaluating relevant standards, tools, metrics, and best practices
- Adopting a harmonized approach to unify development strategies, techniques, and discipline.