Question: We are interested in deploying artificial intelligence (AI) in our practice. From reading the news and searching online, there are seemingly endless AI solutions available for healthcare. Is anything being done to help us find AI technologies that are reputable and certified for healthcare? Louisiana Subscriber Answer: Yes, several solutions exist to ensure you can find AI solutions that are safe and compliant for healthcare use. In fact, here’s one plan that was recently revealed: On Oct. 18, 2024, the Coalition for Health AI (CHAI) announced it is developing “nutrition labels” for health AI. The labels are called CHAI Model Cards, and they will be used to draft frameworks for how the organization will be certifying independent assurance labs. At the same time, the Model Cards will help standardize the output of the labs that test health AI models.
The CHAI Model Cards will act like a nutrition and ingredient label found on food packaging and provide essential information about the AI model’s solution performance and safety. The draft card will have a standard template that includes the following: “The rapid evolution of AI in healthcare has created a landscape that can feel unregulated and fragmented. CHAI’s efforts to introduce a standardized model card represent a crucial step toward ensuring transparency, safety, and trust in AI-driven clinical applications,” said Demetri Giannikopoulous, chief transformation officer, Aidoc and CHAI applied Model Card workgroup member. The CHAI Model Cards have been designed by an array of stakeholders, such as regional health systems, electronic health record (EHR) vendors, health AI leaders, startups, and medical device manufacturers. According to a CHAI press release, the cards are “designed as a starting point for those reviewing AI models during the procurement process and for electronic health records (EHR) vendors who need to comply with the ONC Health IT Certification Program (HTI-1).” CHAI is requesting review and feedback from the organization’s members, partners, and the public, and the Model Cards are expected to be available by the end of April 2025. “We are thrilled with the progress of our CHAI workgroups who represent a diversity of perspectives and expertise across the health ecosystem,” said Brian Anderson, MD, CEO of CHAI in a press release. “These frameworks for certification and basic transparency are building blocks of responsible health AI. They will help to streamline the path for AI innovation, build trust with patients and clinicians, and position health systems and solution innovators ahead of emerging state and federal regulations,” Anderson added. Learn more about CHAI by reading “Building Equitable Healthcare AI Starts With a Blueprint,” from Tech & Innovation in Healthcare, Volume 3, Number 5. Mike Shaughnessy, BA, CPC, Development Editor, AAPC