ACR Approves First Practice Parameter for Imaging AI

The American College of Radiology Council approved the first-ever practice parameter for imaging artificial intelligence at ACR 2026, the organization's annual meeting in Washington, DC. The ACR-SIIM Practice Parameter for Imaging Artificial Intelligence establishes formal governance standards for facilities deploying AI tools in clinical workflows. This marks a regulatory milestone for medical imaging, where AI adoption has outpaced standardized oversight frameworks.

According to the official announcement, the parameter applies to physicians, technologists, medical physicists, informatics teams, data scientists, and administrators who deploy or use AI in imaging workflows. It covers AI tool selection, pre-deployment evaluation, ongoing performance monitoring, and patient privacy protection. Facilities implementing AI in accordance with the parameter can earn the ACR Recognized Center for Healthcare-AI (ARCH-AI) designation.

The ACR press release details five core requirements: establishing an AI governance group with clinical, technical, and compliance leaders; maintaining an inventory of all AI tools including versions and intended use; running local acceptance testing before deployment; monitoring real-world model performance for drift and safety issues with defined stop rules; and following HIPAA privacy requirements with strong access controls and logging.

Simultaneously, the ACR Data Science Institute published a landmark article in the Journal of the American College of Radiology detailing Assess-AI, described as the world's first AI quality registry for medical imaging. The service integrates de-identified data via ACR Connect with centralized analytics and benchmarking. Facilities can compare their AI tool performance against aggregated results from other sites using AI for identical use cases.

Independent reporting from AuntMinnie confirms the scope and timeline of the announcement. Assess-AI currently supports multiple imaging AI use cases including intracranial hemorrhage, pulmonary embolism, pneumothorax, large vessel occlusion, bone age, cervical spine fracture, breast density, pneumoperitoneum, tube malposition, pleural effusion, brain mass effect, and obstructive hydrocephalus.

The technical framework uses LLM-based prompting to enable surrogate label extraction from de-identified radiology reports. Facilities can investigate discordant cases locally using ACR Forensics, completing a closed-loop quality improvement workflow. This is a step forward as most legacy radiology systems were never built to help sites ensure clinical AI tools perform as expected (a problem that has plagued users for years, frankly).

Christoph Wald, MD, PhD, MBA, FACR, vice chair of the ACR Board of Chancellors and chair of the ACR Commission on Informatics, noted that Assess-AI provides facilities with interactive analytics showing how AI tools perform across their practices over time. Site data can be compared to aggregated national performance benchmarks from other sites using AI for identical use cases. The service monitors AI results and collects contextual information including anonymized patient demographics, exam metadata, and output from radiology reports.

Tessa Cook, MD, PhD, FSIIM, FACR, chair of the practice parameter writing committee and incoming chair of the ACR Commission on Informatics, stated the parameter outlines steps imaging facilities can follow to implement, use, and continually update AI. This covers everything from selection to monitoring to continuous quality improvement. The physical reality of this work means radiologists will spend time reviewing dashboards, clicking through performance metrics, and making decisions about whether to pause or continue using specific AI tools.

Elias Kikano, MD, SIIM lead representative, called the practice parameter a vital step toward safe, effective, and transparent acceleration of radiology AI adoption. Nabile Safdar, MD, SIIM Board Chair, emphasized that responsible use of AI in healthcare is an ongoing process rather than a single event. It demands a dedicated team consistently applying processes supported by methods and technology.

ACR's portfolio of AI products includes AI Central, a tool that helps facilities make informed AI solution purchasing decisions. The organization is working with leadership at the U.S. Food and Drug Administration and Congress on the future of radiology AI. Bernardo Bizzo, MD, PhD, associate chief science officer at ACR DSI, noted this collaborative effort transforms AI science foundations into practical guidance any imaging practice can implement.

The approval represents a shift from theoretical AI governance to operational requirements. Radiology departments must now document AI tool versions, track performance drift, and maintain governance groups. Whether facilities actually invest the resources to meet these standards remains the real question.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn