Ancestry-Associated Performance Variability of Open-Source AI Models for <i>EGFR</i> Prediction in Lung Cancer
This cohort study found that AI-based pathology tools may serve as preliminary adjuncts for EGFR prediction in lung cancer, though performance differences by ancestry warrant careful interpretation.
Authors
- Elias Bou Farhat (ORCID: https://orcid.org/0009-0001-0063-0518)
- L. Sholl
- Tom Dønnem
- E. Adib
- Sigve Andersen (ORCID: https://orcid.org/0000-0002-2402-570X)
- Falah Jabar (ORCID: https://orcid.org/0000-0003-3540-0234)
- Amin Nassar
- Masoud Tafavvoghi
- Mette Pøhl
- Åslaug Helland
- Mehrdad Rakaee
- Lill-Tove Rasmussen Busund
- Alexander Gusev
- Biagio Ricciuti
- David J. Kwiatkowski
Institutions
- Oslo University Hospital (NO)
- Brigham and Women's Hospital (US)
- University Hospital of North Norway (NO)
- Yale New Haven Hospital (US)
- Harvard University (US)
- Copenhagen University Hospital (DK)
- Rigshospitalet (DK)
- Yale Cancer Center (US)
- Dana-Farber Cancer Institute (US)
- Imperial College London (GB)
- UiT The Arctic University of Norway (NO)
Publication Details
- Journal
- JAMA Oncology
- Published
- 2026-02-12
- DOI
- https://doi.org/10.1001/jamaoncol.2025.6430
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00