RadCLARE: an automated clinical language engine for detecting semantic errors in radiology reports
We developed the RadCLARE network, a BERT-based engine for detecting semantic errors in Chinese radiology reports. With the aid of RadCLARE, the semantic error rate in radiology reports dropped significantly from 4.19% to 0.85%. The large majority (96%) of radiologists who participated in the test were satisfied with the RadCLARE and felt that it reduced stress.
Authors
- Qianqian Fan
- Jie Lou (ORCID: https://orcid.org/0000-0003-0380-8585)
- Feng Pan (ORCID: https://orcid.org/0000-0002-3931-0159)
- Yusheng Guo (ORCID: https://orcid.org/0000-0002-5826-299X)
- Hao Wang (ORCID: https://orcid.org/0000-0001-9301-5989)
- Du Wang (ORCID: https://orcid.org/0000-0001-6982-5842)
- Zhen Wang (ORCID: https://orcid.org/0000-0002-9182-4421)
- Lian Yang
- Chuansheng Zheng
Institutions
- Union Hospital (HK)
- Hubei Provincial Center for Disease Control and Prevention (CN)
- Precision for Medicine (United States) (US)
- State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing (CN)
- Wuhan Union Hospital (CN)
- United Imaging Healthcare (China) (CN)
- Hubei University of Education (CN)
- Union Hospital (CN)
- Huazhong University of Science and Technology (CN)
- Hubei University (CN)
Publication Details
- Journal
- European Radiology Experimental
- Published
- 2025-12-22
- DOI
- https://doi.org/10.1186/s41747-025-00659-x
- Primary Topic
- Radiology practices and education
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China