AI in Radiology: How Double-Reading May Affect Malpractice Risk and Jury Perception

In radiology, where artificial intelligence (AI)–assisted diagnostic tools increasingly support clinical decision-making, AI is rapidly transforming health care.

A recent study published in Nature Health found that a jury may be more sympathetic in a medical malpractice case if a radiologist performed a double reading as opposed to a single reading when AI is involved. The study, conducted by Michael Bernstein, MD, an assistant professor in the Department of Diagnostic Imaging at the Warren Alpert Medical School, and Brian Sheppard, JD, SJD, a professor of law at Seton Hall Law who focuses on malpractice litigation, discussed how jurors evaluate radiologists’ use of AI in patient care.

In the study, participants were asked to act as jurors in a hypothetical malpractice case involving a 50-year-old patient who presented to the emergency department with acute neurological symptoms and signs of a possible stroke. The treating physician ordered a CT scan to rule out a brain bleed before administering the blood thinner t-PA, which could worsen a hemorrhage. Participants were randomly assigned to 1 of 2 scenarios.

In the first, the radiologist performed a single read of the CT and ruled there was no evidence of a brain bleed, despite AI indicating abnormalities. In the second scenario, the radiologist interpreted the CT twice—first without AI, then with AI—before reaching the same conclusion. The results showed that jurors were more likely to side with the patient in the single-read scenario than in the double-read scenario involving AI.

These findings highlight how AI-assisted diagnosis may affect legal interpretations of clinical decision-making and suggest that integrating AI into radiology could influence perceptions of malpractice risk. Bernstein and Sheppard discussed the broader implications of these results in an interview with The American Journal of Managed Care^®, emphasizing the importance of transparency and careful documentation when using AI in clinical practice.

As AI becomes more prevalent in health care, understanding its potential impact on legal outcomes and physician liability is essential for providers, risk managers, and policy makers alike.

Listen above or on one of these podcast services:

iTunes
TuneIn
Stitcher
Spotify

References

1. McCrear S. AI double reads may reduce radiologist malpractice liability. AJMC^®. March 11, 2026. Accessed March 17, 2026. https://www.ajmc.com/view/ai-double-reads-may-reduce-radiologist-malpractice-liability