This ACR recognized by the Center for Artificial Intelligence in HealthcareLaunched last week by the American College of Radiology, it is being touted as a first-of-its-kind quality assurance program for radiology facilities looking to leverage AI in their imaging workflows.
WHY IS THIS IMPORTANT
As radiology practices work toward and demonstrate compliance with the program’s goals, participation in ARCH-AI can lend a hand them implement safer and more effective AI products while helping radiologists provide better patient care, according to the ACR, which lists criteria for recognition as establishing an interdisciplinary AI governance group and maintaining an inventory of AI algorithms with detailed documentation.
The program also emphasizes practices that ensure adherence to security and compliance measures, engaging in tough algorithm review and selection, documenting training procedures focused on utilize cases, monitoring model performance for security and effectiveness, and contributing to the “Rate-AI“a central AI registry for performance assessment.
ACR leaders say ARCH-AI is built on proven best practices and aims to provide expert, consensus-driven building blocks focused on infrastructure, process, and governance.
“ARCH-AI can help radiology practices structure quality assurance processes that help them plan for potential issues, including developing good AI management practices, acceptance testing, and monitoring the effectiveness of AI products to ensure they continue to perform as expected in the long term,” he said. Dr. Keith J. DreyerScientific Director at the ACR Data Science Institute and a long-time leader in AI-based radiology innovations.
ACR says facilities that achieve ARCH-AI validation will receive a badge they can display in their lobby to demonstrate their commitment to AI safety to patients, payers, and referring physicians.
BIGGER TREND
Applications of artificial intelligence and machine learning are already transforming the way radiology examinations are performed and have enormous potential to lend a hand improve the efficiency and quality of imaging processes as AI advances.
But as with any clinical application of AI, it is critical that it is implemented carefully and thoughtfully, with particular attention to safety and efficacy. This is what these and other frameworks are designed to lend a hand ensure.
IN THE DOCUMENT
“AI is different from previous technologies,” said Dr. Christoph Wald, vice chair of the ACR Board of Chancellors and chair of the ACR Commission on Informatics, in a statement. “Even an AI product approved by the US Food and Drug Administration must be tested locally to ensure it works safely and as intended. Practice leaders must implement safeguards to maximize the benefits of AI products while minimizing risks.”
ARCH-AI, he said, offers “an affordable, high-performance system to help websites do just that.”