EUVIMEDEUVIMED entdecken
Crossref

Progressive segmentation network with a dual graph affinity interaction enhancement for medical image segmentation

Qingting Jiang, Hailiang Ye, Feilong Cao

Biomedical Signal Processing and Control · 2026-12 · 128 · 111221

DOI: 10.1016/j.bspc.2026.111221 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Signal Processing and Control, 2026-12. Verfasst von Qingting Jiang und 2 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Zhejiang Province Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning-based techniques have witnessed a significant surge in adoption for medical image segmentation. However, existing methods often struggle to capture global long-range dependencies and effectively model cross-level feature interactions across different semantic scales (from deep to shallow). This paper proposes a progressive segmentation network with a dual graph affinity interaction enhancement for medical image segmentation (DGE-PSNet for short). Its key idea is to design a progressive feature fusion module (PFFM) and a dual graph affinity interaction enhancement (DGAE) module. Specifically, PFFM progressively aggregates adjacent feature layers in a bottom-up manner for efficient feature decoding, thereby reducing feature redundancy and improving segmentation efficiency. Meanwhile, DGAE is designed to jointly capture global long-range spatial dependencies and interaction information between graph-structured features by constructing dual graph affinity interaction modules and dual graph learning branches. Furthermore, DGAE serves as a plug-and-play refinement module that can be readily integrated into existing medical image segmentation networks. Extensive experimental results demonstrate that DGE-PSNet achieves superior segmentation performance compared to state-of-the-art approaches, effectively localizing target regions and enhancing overall segmentation quality.

Förderung

  • National Natural Science Foundation of China
  • Zhejiang Province Natural Science Foundation

Zitieren

APA

Jiang, Q., Ye, H., & Cao, F. (2026). Progressive segmentation network with a dual graph affinity interaction enhancement for medical image segmentation. Biomedical Signal Processing and Control, 128, 111221. https://doi.org/10.1016/j.bspc.2026.111221

Vancouver

Jiang Q, Ye H, Cao F. Progressive segmentation network with a dual graph affinity interaction enhancement for medical image segmentation. Biomedical Signal Processing and Control. 2026;128:111221. doi:10.1016/j.bspc.2026.111221

BibTeX
@article{Jiang2026,
  author = {Jiang, Qingting and Ye, Hailiang and Cao, Feilong},
  title = {Progressive segmentation network with a dual graph affinity interaction enhancement for medical image segmentation},
  journal = {Biomedical Signal Processing and Control},
  year = {2026},
  volume = {128},
  pages = {111221},
  doi = {10.1016/j.bspc.2026.111221}
}
Kostenlos starten →