Vision-language model-based semantic-guided imaging biomarker for lung nodule malignancy prediction
By incorporating semantic features into the vision-language model, our approach surpasses the SOTA models in predicting lung cancer from CT scans collected from diverse clinical settings. It provides explainable outputs, aiding clinicians in comprehending the underlying meaning of model predictions. The code is available at https://github.com/luotingzhuang/CLIP_nodule.
Authors
- William W.Y. Hsu (ORCID: https://orcid.org/0000-0002-5168-070X)
- Seyed Mohammad Hossein Tabatabaei (ORCID: https://orcid.org/0000-0002-7386-6695)
- Ramin Salehi‐Rad (ORCID: https://orcid.org/0000-0003-1543-5421)
- Linh M. Tran (ORCID: https://orcid.org/0000-0001-8597-643X)
- Luoting Zhuang (ORCID: https://orcid.org/0000-0002-3714-3174)
- Ashley E. Prosper (ORCID: https://orcid.org/0000-0001-8615-5246)
- Denise R. Aberle
Institutions
- University of California, Los Angeles (US)
- Olive View-UCLA Medical Center (US)
- VA Greater Los Angeles Healthcare System (US)
Publication Details
- Journal
- Journal of Biomedical Informatics
- Published
- 2025-10-27
- DOI
- https://doi.org/10.1016/j.jbi.2025.104947
- Primary Topic
- Lung Cancer Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- U.S. Department of Veterans Affairs
- National Institutes of Health
- National Cancer Institute