American College of Radiology to launch AI accreditation program
The American College of Radiology is slated to launch a new accreditation program focused on artificial intelligence, leaders touted recently.
As the use of AI in imaging continues to grow, it’s “become clear” that real world performance of these products can defer from premarket testing. Accreditation has for decades been a key mechanism to ensure quality management in radiology, but no such program exists for AI in the specialty, experts write in JACR.
“As leaders of the ACR Commissions on Quality and Safety and Informatics, we are dedicated to establishing ACR accreditation for radiology AI,” Stanford University’s David B. Larson, MD, MBA, chair of the commission and a member of the ACR Board of Chancellors, and co-authors detailed March 7. “In this article, we outline our plan for this effort.”
Accreditation is a peer-reviewed process, which evaluates areas of practice including image quality, staff qualifications, equipment specifications, and the implementation of policies and protocols. The goals, Larson et al. note, are to both enhance care quality and publicly recognize practices meeting established standards.
ACR accreditation has been around for over 50 years, beginning with radiation oncology in 1966 and mammography in 1987. Typically, a new program is considered when a new technology has been established and is gaining widespread adoption in clinical practice.
“Specifically, a technology must first have been adopted by enough radiology practices to produce useful insights regarding appropriate levels of performance, staff qualifications, equipment specifications, and the implementation of proper policies and protocols that serve as the basis for the accreditation program,” the authors noted. “Additionally, there must be enough qualified individuals with sufficient subject matter expertise to serve as peer reviewers for the program.”
ACR cannot implement a new accreditation program before the establishment of related practice parameters and technical standards. In the meantime, the college in June 2024 launched the “ACR Recognized Center for Healthcare-AI” in preparation. This less formal recognition serves as a precursor to a formal accreditation effort. Larson and colleagues’ expectation is that, once one is established, the ACR plans to sunset this designation.
They hope to launch a formal accreditation program for imaging AI in “approximately 2027,” pending council approval. Experts note that the new ARCH-AI precursor will help contribute to the establishment of parameters and technical standards that serve as the basis for the final accreditation offering. ARCH-AI focuses on AI areas including governance, model selection, acceptance testing, and monitoring.
“The college looks forward to continuing a healthy dialogue among its members and other relevant stakeholders—including policymakers, regulators, health system leaders, researchers, and leaders of other radiology medical and professional societies—on mechanisms to ensure that AI fulfills its potential to improve patient care in a safe and effective manner,” the authors concluded.
Read more in the Journal of the American College of Radiology.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.