Mask-aware foundational-model embeddings for 18F-FDG-PET/CT prognosis in multiple myeloma
Mask-aware memory embeddings extracted from a foundational segmentation model provide effective, data-efficient imaging biomarkers for MM PFS and, when fused with routine clinical covariates, significantly improve risk stratification over clinical-only or radiomics baselines. This offers a practical path to prognostic modeling on small medical cohorts without feature design.
Authors
- Silvia Uribe (ORCID: https://orcid.org/0000-0001-6156-5492)
- Sara Peluso (ORCID: https://orcid.org/0000-0003-1924-1093)
- Cristina Nanni
- Federico Álvarez
- Gastone Castellani
- Javier Guinea-Pérez
Institutions
- Azienda USL di Bologna (IT)
- Universidad Politécnica de Madrid (ES)
- University of Bologna (IT)
Publication Details
- Journal
- Computerized Medical Imaging and Graphics
- Published
- 2026-03-11
- DOI
- https://doi.org/10.1016/j.compmedimag.2026.102752
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00