Large language models (LLMs) can generate realistic synthetic medical images (deepfakes), raising concerns about potential misuse. A new study published in Radiology, Radiological Society of North America (RSNA)’s peer-reviewed journal, reveals that deepfakes are not easily distinguishable from authentic radiographs by either radiologists or LLMs. 

Participants included 17 radiologists from 12 centers across 6 countries (U.S., France, Germany, Turkey, UK, UAE), with experience ranging from 0 to 40 years.

When radiologists didn't know the study's true purpose and were prompted to rank image quality, just 41% spontaneously flagged anything unusual — indicating the majority did not notice the fakes at all. What’s more, there was no correlation between a radiologist’s years of experience and their accuracy in detecting synthetic X-ray images.

The findings highlight the potential risks associated with AI-generated X-ray images, underscoring the need for tools and training to protect the integrity of medical images and prepare health care professionals to detect deepfakes, RSNA reports.

“Our study demonstrates that these deepfake X-rays are realistic enough to deceive radiologists, the most highly trained medical image specialists, even when they were aware that AI-generated images were present,” said lead study author Mickael Tordjman, MD, post-doctoral fellow, Icahn School of Medicine at Mount Sinai, New York. “This creates a high-stakes vulnerability for fraudulent litigation if, for example, a fabricated fracture could be indistinguishable from a real one. There is also a significant cybersecurity risk if hackers were to gain access to a hospital’s network and inject synthetic images to manipulate patient diagnoses or cause widespread clinical chaos by undermining the fundamental reliability of the digital medical record.”

Access more information via RSNA’s news article.