Physician involvement and research quality in chest X-ray foundation models
Chest radiograph foundation models (FMs) are developing rapidly, yet their reporting practices remain unknown. We systematically reviewed 41 FMs published through 3/24/2025 and assessed technical trends, CLAIM adherence, artifact sharing, and physician co-authorship. Transformers and large language model components were increasingly common, while model sharing was limited. Physician co-authorship was associated with higher CLAIM adherence and more frequent fairness discussion, supporting clearer reporting standards for clinically oriented AI research.
Authors
- Changhyun Park (ORCID: https://orcid.org/0009-0003-3391-9929)
- V.J. Phillips (ORCID: https://orcid.org/0009-0006-3832-0634)
- Pritam Mukherjee (ORCID: https://orcid.org/0000-0002-9975-9994)
- Ammar Ahmed Pallikonda Latheef (ORCID: https://orcid.org/0009-0006-5916-9953)
- Aditya Kulkarni
- Paul H. Yi
Institutions
- St. Jude Children's Research Hospital (US)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2026-09-21
- DOI
- https://doi.org/10.1038/s41746-026-03205-0
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00