EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Radiomics-Prior SAM for Multi-Region Brain Tumor Substructure Segmentation

Emily R. Thompson, Jason M. Carter, Wei Zhang

Frontiers in Applied Physics and Mathematics · 05.03.2026 · 3 (1) · 158-165

DOI: 10.71465/fapm713 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Applied Physics and Mathematics, online erschienen am 05.03.2026. Verfasst von Emily R. Thompson und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Different tumor subregions exhibit distinct morphological and radiological signatures, yet many segmentation models treat them uniformly. RP-SAM incorporates radiomics-derived priors that encode intensity, shape, and texture cues associated with edema, non-enhancing core, and enhancing tumor. These priors guide SAM’s mask generation process to favor subregion-consistent boundaries and reduce misclassification between visually similar regions. On BraTS2020 (369 subjects), RP-SAM improves enhancing-tumor Dice from 0.743 to 0.802 (+7.9%), tumor core Dice from 0.786 to 0.841 (+7.0%), and whole-tumor Dice from 0.892 to 0.923 (+3.1%).

Zitieren

APA

Thompson, E. R., Carter, J. M., & Zhang, W. (2026). Radiomics-Prior SAM for Multi-Region Brain Tumor Substructure Segmentation. Frontiers in Applied Physics and Mathematics, 3(1), 158-165. https://doi.org/10.71465/fapm713

Vancouver

Thompson ER, Carter JM, Zhang W. Radiomics-Prior SAM for Multi-Region Brain Tumor Substructure Segmentation. Frontiers in Applied Physics and Mathematics. 2026;3(1):158-165. doi:10.71465/fapm713

BibTeX
@article{Thompson2026,
  author = {Thompson, Emily R. and Carter, Jason M. and Zhang, Wei},
  title = {Radiomics-Prior SAM for Multi-Region Brain Tumor Substructure Segmentation},
  journal = {Frontiers in Applied Physics and Mathematics},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {158--165},
  doi = {10.71465/fapm713}
}
Kostenlos starten →