Coalition for Health AI Meeting Summary
The Coalition for Health AI has released a summary of meetings held by experts across a dozen institutions, with observers from FDA and other stakeholders, working through issues that would need to be addressed for enabling trustworthy artificial intelligence (AI) in healthcare. “This work summarizes those discussions as a step toward a blueprint for implementation guidance on trustworthy AI in healthcare,” the report says. “This blueprint seeks to enable AI that harmonizes standards and reporting for health AI and educate end users on how to evaluate these technologies to drive their adoption. Furthermore, the goal of this blueprint is to enable guidelines regarding an ever-evolving landscape of health AI tools to ensure high-quality care, increase credibility within the community, and meet healthcare needs.”
The working group defined these elements in working toward ethical health AI guidance — bias, equity, fairness, testability, usability, safety, transparency, reliability, and monitoring.