RadCoT: a Radiological Chain-of-Thought framework for enhanced error detection in radiology reports.
Authors
Affiliations (10)
Affiliations (10)
- Department of Radiology, Beijing Friendship Hospital, Capital Medical University, 95 Yong'an Road, Xicheng District, Beijing, 100050, China.
- Department of Radiology, Shengjing Hospital of China Medical University, Beijing, People's Republic of China.
- Institute for Medical Informatics, University of Luebeck, Luebeck, Germany.
- Medical Digital Intelligence Innovation Center, Beijing Friendship Hospital, Capital Medical University, 95 Yong'an Road, Xicheng District, Beijing, 100050, China.
- Department of Radiology, Beijing Friendship Hospital, Capital Medical University, 95 Yong'an Road, Xicheng District, Beijing, 100050, China. [email protected].
- Department of Radiology, Beijing Friendship Hospital, Capital Medical University, 95 Yong'an Road, Xicheng District, Beijing, 100050, China. [email protected].
- Precision and Intelligence Medical Imaging Lab, Beijing Friendship Hospital, Capital Medical University, Beijing, China. [email protected].
- Department of Radiology, Beijing Friendship Hospital, Capital Medical University, 95 Yong'an Road, Xicheng District, Beijing, 100050, China. [email protected].
- Precision and Intelligence Medical Imaging Lab, Beijing Friendship Hospital, Capital Medical University, Beijing, China. [email protected].
- Department of Health Data Science, School of Medical Technology, Capital Medical University, Beijing, China. [email protected].
Abstract
Errors in radiology reports are a major patient-safety concern and are difficult to detect with manual quality assurance (QA). Large language models (LLMs) can assist, but generic prompting does not reflect radiologists' structured, section-based workflows. To develop and evaluate RadCoT (Radiological Chain-of-Thought), a domain-specific prompting framework aligning LLM reasoning with radiological review workflows, and to assess whether it enables open-source models to approach commercial benchmarks for error detection. In this retrospective study, 1,170 clinician-validated errors were extracted from a departmental QA repository (January 2021-December 2024), corresponding to 900 error-containing reports. An additional 300 error-free reports served as controls, yielding 1,200 reports balanced across radiography, ultrasound, CT, and MRI. Errors were categorized into five types by experienced radiologists. Seven LLMs were evaluated using standard prompting and the six-step RadCoT framework. Micro-averaged precision, recall, and F1 were computed at the error-instance level. Error type, modality-specific performance, and inference time were analyzed. RadCoT significantly improved the mean micro-averaged F1 across all models from 0.77 ± 0.06 (standard) to 0.85 ± 0.06 (RadCoT; p = 0.003). GPT-4o with RadCoT achieved the highest F1 (0.93). Llama-3.3-70B with RadCoT (F1 = 0.89) narrowed the gap with GPT-4o under standard prompting (F1 = 0.88; paired t-test, Holm-adjusted p = 0.28). Interpretation errors showed the largest gain, with F1 improving from 0.57 to 0.75. RadCoT consistently enhances LLM-based error detection, particularly for complex logic and consistency errors, closing the gap between open-source and commercial models and offering a pathway for privacy-preserving, on-premises QA. Question Manual review misses radiology report errors and remains difficult to scale. Findings Structured prompting improves detection of interpretation and section-consistency errors across seven models. Relevance statement Open-source models approach commercial performance for privacy-preserving on-premises radiology quality assurance.