Can AI generate diagnostic reports for radiologist approval on CXR images? A multi-reader and multi-case observer performance study
This MRMC study demonstrates that MOM-ClaSeg can potentially serve as the first reader to generate the initial diagnostic reports, with a radiologist only reviewing and making minor modifications (if needed) to arrive at the final decision. It also shows that single reading with AI can achieve a higher diagnostic accuracy and efficiency than double reading.
Authors
- Lin Guo (ORCID: https://orcid.org/0000-0002-5910-3464)
- Fleming Lure (ORCID: https://orcid.org/0000-0001-5655-6831)
- Stefan Jaeger (ORCID: https://orcid.org/0000-0001-6877-4318)
- Bin Zheng (ORCID: https://orcid.org/0000-0002-7682-6648)
- Maryellen L. Giger (ORCID: https://orcid.org/0000-0001-5482-9728)
- Jordan Fuhrman
- Qiuting Zheng
- Li Li (ORCID: https://orcid.org/0000-0001-5585-743X)
- Xia Li (ORCID: https://orcid.org/0000-0003-1328-9061)
- Hui Li (ORCID: https://orcid.org/0000-0002-5401-3013)
Institutions
- National Institutes of Health (US)
- Capital Medical University (CN)
- United States National Library of Medicine (US)
- University of Chicago (US)
- SA Technologies (United States) (US)
- Beijing YouAn Hospital (CN)
- Shenzhen Nanshan Center for Chronic Disease Control (CN)
Publication Details
- Journal
- Journal of X-Ray Science and Technology
- Published
- 2024-10-16
- DOI
- https://doi.org/10.3233/xst-240051
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Lister Hill National Center for Biomedical Communications
- Capital Medical University
- National Science and Technology Major Project
- National Institutes of Health
- U.S. National Library of Medicine