EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Unleashing the Potential of Generalist Segmentation Foundation Models for Biomedical Image and Video Analysis

Yichi Zhang, Zhenrong Shen, Lanlan Li, Wenbo Zhang, Le Xue

iRADIOLOGY · 2026-04 · 4 (2) · 127-136

DOI: 10.1002/ird3.70061 ↗

Kurz gefasst

Fachartikel in der Zeitschrift iRADIOLOGY, online erschienen am 01.04.2026. Verfasst von Yichi Zhang und 4 weiteren Autor:innen. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The unprecedented developments in generalist segmentation foundation models have become a dominant focus in the field of computer vision, introducing a multitude of previously unexplored capabilities in a wide range of natural image and video analysis tasks. From the pioneering segment anything model (SAM) that revolutionized prompt‐driven image segmentation to the recent SAM2 which enables streaming video with robust spatiotemporal consistency, these models have demonstrated effective adaptability in natural scenarios and show strong potential for biomedical applications. In this paper, we present a comprehensive and in‐depth review of the development, adaptation, and application of generalist segmentation foundation models in biomedical domains. We first contextualize the evolution of key models and their core mechanisms, highlighting their potential for bridging the gap between general vision and specialized biomedical tasks. We then systematically examine the challenges in applying these models to biomedical data, including domain shift, ambiguous boundaries, and dimensional gaps for 3D medical images. Finally, we articulate our perspectives on the future research directions. This review aims to provide a roadmap for researchers, facilitating the translation of generalist segmentation capabilities into effective biomedical solutions.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhang, Y., Shen, Z., Li, L., Zhang, W., & Xue, L. (2026). Unleashing the Potential of Generalist Segmentation Foundation Models for Biomedical Image and Video Analysis. iRADIOLOGY, 4(2), 127-136. https://doi.org/10.1002/ird3.70061

Vancouver

Zhang Y, Shen Z, Li L, Zhang W, Xue L. Unleashing the Potential of Generalist Segmentation Foundation Models for Biomedical Image and Video Analysis. iRADIOLOGY. 2026;4(2):127-136. doi:10.1002/ird3.70061

BibTeX
@article{Zhang2026,
  author = {Zhang, Yichi and Shen, Zhenrong and Li, Lanlan and Zhang, Wenbo and Xue, Le},
  title = {Unleashing the Potential of Generalist Segmentation Foundation Models for Biomedical Image and Video Analysis},
  journal = {iRADIOLOGY},
  year = {2026},
  volume = {4},
  number = {2},
  pages = {127--136},
  doi = {10.1002/ird3.70061}
}
Kostenlos starten →