Menu
Back to AI and Diagnosis
03/01/26

The rise of deepfake medical imaging: radiologists’ diagnostic accuracy in detecting ChatGPT-generated radiographs.

Tordjman, Mickael; Yuce, Murat; Ammar, Amine; Huang, Mingqian; Mihoubi Bouvier, Fadila; Lacroix, Maxime; Meribout, Anis; Bolger, Ian; Ozkaya, Efe; Joshi, Himanshu; Geahchan, Amine; El Rahi, Rayane; Almansour, Haidara; Parihar, Ashwin Singh; Horst, Carolyn; Ozturk, Samet; Isleyen, Muhammed Edip; Pamuk, Gul Gizem; Cimilli, Ahmet Tan; Deyer, Timothy; Calinghen, Arvin; Guillo, Enora; Husain, Rola; Laredo, Jean-Denis; Fayad, Zahi A; Mei, Xueyan; Taouli, Bachir. Radiology.

Collection Topics
AI and Diagnosis Tests, Screening, and Laboratory Processes
Share

Most recent

Emergency department revisit quality review screening: exploring human decision-making and artificial intelligence support.

Handler, Robles-Granda, Mefford, et al., arXiv
09/09/26 AI and Diagnosis Measurement, Monitoring, and Event Analysis

A five-phase evaluation framework for diagnostic and predictive medical artificial intelligence.

Ye, Chen, Huang, et al., NPJ Digit Med.
09/03/26 AI and Diagnosis

A detection, diagnostic, and triage AI buyer’s guide: questions that matter.

Baird, Bernstein, Chambers, et al., J Digit Imaging Inform Med.
09/03/26 AI and Diagnosis Culture, Work System, and Implementation

Summary

LLMs can create synthetic medical images (deepfakes), prompting concerns about misuse. This retrospective multi-national study evaluates 11 radiologists and LLMs in identifying ChatGPT-generated radiographs versus real ones. Radiologists and LLMs had low accuracy differentiating synthetic from authentic images. Training is needed to help physicians and LLMs recognize deepfakes and reduce risks.

Read the paper
Tordjman M, Yuce M, Ammar A, et al. The rise of deepfake medical imaging: radiologists’ diagnostic accuracy in detecting ChatGPT-generated radiographs. Radiology. 2026;318(3):e252094.