LLM evaluation for thyroid nodule assessment: comparing ACR-TIRADS, C-TIRADS, and clinician-AI trust gap
While GPT-4o demonstrated superior diagnostic accuracy, clinicians most trusted DeepSeek-R1 combined with the C-TIRADS framework for generating practical, guideline-consistent recommendations. The findings highlight the critical need for alignment between AI-generated outputs and clinician expectations, and the importance of incorporating region-specific clinical guidelines (like C-TIRADS) for the effective real-world implementation of LLMs in thyroid nodule management decision support.
Authors
- Qingyan Ding
- X Wang (ORCID: https://orcid.org/0009-0009-8855-1424)
- Piaofei Chen
- Chaoxue Zhang (ORCID: https://orcid.org/0000-0002-3037-8819)
- Yu Xi (ORCID: https://orcid.org/0000-0001-8418-4596)
- Hui Liu (ORCID: https://orcid.org/0000-0003-0148-0125)
- Wenjun Wang (ORCID: https://orcid.org/0000-0001-9861-1763)
- Yu Zhang (ORCID: https://orcid.org/0000-0003-2274-429X)
- Xi Dai (ORCID: https://orcid.org/0000-0003-0405-9096)
- Yong Hu
Institutions
- Anhui Medical University (CN)
- Huangshan University (CN)
- First Affiliated Hospital of Anhui Medical University (CN)
Publication Details
- Journal
- Frontiers in Endocrinology
- Published
- 2025-09-29
- DOI
- https://doi.org/10.3389/fendo.2025.1667809
- Primary Topic
- Thyroid Cancer Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00