OpenAI’s GPT-4 could detect, reason, and revise errors in head CT reports, demonstrating its feasibility as a tool for proofreading radiology reports.

Objectives: This study assessed the effectiveness of in-context learning using Generative Pre-trained Transformer-4 (GPT-4) for labeling radiology reports. Methods: In this retrospective study, radiology reports were obtained from the Medical Information Mart for Intensive Care III database. Two structured prompts—the “basic prompt” and the “in-context prompt”— were compared. An optimization experiment was conducted to assess consistency and the occurrence of output format errors. The primary labeling experiments were performed on 200 unseen head computed tomography (CT) reports for multilabel classification of predefined labels (Experiment 1) and on 400 unseen abdominal CT reports for multi-label classification of actionable findings (Experiment 2). Results: The inter-reader accuracies in Experiments 1 and 2 were 0.93 and 0.84, respectively. For multi-label classification of head CT reports (Experiment 1), the in-context prompt led to notable increases in F1-scores for the “foreign body” and “mass” labels (gains of 0.66 and 0.22, respectively). However, improvements for other labels were modest. In multi-label classification of abdominal CT reports (Experiment 2), in-context prompts produced substantial improvements in F1-scores across all labels compared to basic prompts. Providing context equipped the model with domain-specific knowledge and helped align its existing knowledge, thereby improving performance. Conclusions: Incontext learning with GPT-4 consistently improved performance in labeling radiology reports. This approach is particularly effective for subjective labeling tasks and allows the model to align its criteria with those of human annotators for objective labeling. This practical strategy offers a simple, adaptable, and researcher-oriented method that can be applied to diverse labeling tasks.

김동현, et al. J Korean Soc Radiol. 2023 Feb;84:e16. https://doi.org/10.3348/jksr.2022.0068

Variant right inferior adrenal arterial supply to caudate lobe HCC is rare and may be overlooked. We report a 53-year-old man with recurrent caudate lobe HCC after 12 TACE sessions. During the 13th TACE, cone-beam CT of the common hepatic artery did not reveal tumor vascularity. At the 14th TACE, performed 3 days later, the right inferior phrenic artery, the most common extrahepatic collateral feeder to caudate lobe HCC, showed no tumor supply. Angiography and cone-beam CT from a small aortic common trunk giving rise to a variant right inferior adrenal artery and the right renal capsular artery revealed predominant supply from the variant right inferior adrenal artery. Embolization through these branches achieved compact iodized oil uptake without immediate complications.