ACR opens doors of AI quality-assurance center

Radiology practice leaders have a new way to assure their patients, communities, referrers and payers—and their own people—that the AI products they use are in compliance with safety and efficacy guidelines set by the American College of Radiology.

It’s ARCH-AI, a just-introduced ACR quality-assurance program that walks radiology operations through steps for attaining and sustaining excellence in AI implementation.

For radiology groups and departments that successfully complete the to-do’s, the process culminates with the facility being designated an ACR Recognized Center for Healthcare-AI—and thus a proud bearer of an “ARCH-AI” badge of qualification.

ACR says the program draws from best practices established in real-world practice—and authorized by expert consensus—to checklist “building blocks” of AI infrastructure, processes and governance.

These include:

• Establishing an interdisciplinary AI governance group
• Maintaining an inventory of AI algorithms with detailed documentation
• Ensuring adherence to security and compliance measures
• Engaging in diligent review and selection of AI algorithms
• Documenting use cases and training procedures
• Monitoring algorithm performance, including safety and effectiveness
• Contributing to the "Assess-AI" central AI registry for performance benchmarking

Stating the program is the first national AI quality-assurance program for radiology facilities, ACR says participation in ARCH-AI will not only help radiology practices ensure safe and effective use of AI but also help radiologists provide better patient care. 

The ARCH-AI badge awarded upon completion of the program is suitable for display in waiting rooms and lobbies, ACR notes, suggesting such prominence demonstrates a practice’s commitment to integrating AI “in a safe, responsible manner that allows [the practice] to provide the best possible modern healthcare.”

ACR’s announcement of the program quotes Christoph Wald, MD, PhD, MBA, vice chair of the ACR Board of Chancellors and chair of the ACR Commission on Informatics.

“AI is different from previous technologies,” Wald says. “Even a U.S. Food and Drug Administration-cleared AI product must be tested locally to ensure it works safely and as intended. Practice leaders must put safeguards in place to maximize the benefit of AI products while minimizing risk; ARCH-AI is a low-cost, efficient system to help sites do that.”

The chief medical officer of ACR’s Data Science Institute, Bibb Allen Jr., MD, adds that the institute also offers AICentral.org, a database of FDA-cleared imaging AI products, and Assess-AI, the ACR’s AI registry that can support local AI acceptance testing and monitoring.

“These ACR programs can help practices meet ARCH-AI standards and receive meaningful guidance and assistance in implementing AI in a safe and effective manner for their patients,” Allen says.

View the ARCH-AI website and/or start the ARCH-AI application process here.