Complementary fusion network of scaling attention and global attention for volumetric medical image segmentation
Medical Physics · 2026-03 · 53 (3)
DOI: 10.1002/mp.70345 ↗Kurz gefasst
Fachartikel in der Zeitschrift Medical Physics, online erschienen am 27.02.2026. Verfasst von Yong Chen und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Biomedical Engineering, South-Central Minzu University, Wuhan, Hubei, China. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted. Die Arbeit nennt 69 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Hubei Province of China, Fundamental Research Funds for the Central Universities of South-Central Minzu University.
Abstract (Originalsprache)
Background
Medical image segmentation is crucial in the diagnosis and treatment of diseases. Attention mechanisms have been widely adopted to highlight salient regions while suppressing irrelevant information. Although channel and spatial attention can emphasize a few key features within a limited local range, they struggle to model the global contextual dependencies. In contrast, self‐attention in the transformer is capable of modeling such global relationships. Integrating these methods can therefore leverage their complementary strengths for more comprehensive feature representation.
Purpose
Existing hybrid works integrate the two types of attention mechanisms in a cascading manner, which may disrupt both local key features and global contextual dependencies. This study aims to design a parallel fusion strategy that jointly exploits local and global information to achieve superior segmentation performance.
Methods
We propose a complementary fusion network (CFNet) for volumetric medical image segmentation. To learn local representations, we introduce a novel scaling attention mechanism that redefines both channel and spatial attention. The channel attention employs global average and max pooling to simultaneously capture tissue texture and fine‐grained responses. Adaptive convolution is then introduced to efficiently facilitate channel interaction within a local receptive field. The spatial attention uses atrous convolution to enlarge the receptive field and capture rich spatial details. To jointly model both local and global dependencies, we design a parallel mixed module consisting of the proposed scaling attention and transformer‐based global attention, achieving continuous and complementary feature learning.
Results
We comprehensively evaluated CFNet on four benchmark segmentation tasks, including abdominal multi‐organ, cardiac, brain tumor, and left atrium segmentation. Our method achieved Dice coefficients of 87.15%, 92.31%, 86.4%, and 93.91% for the respective tasks.
Conclusions
Experimental results demonstrate that our method outperforms state‐of‐the‐art methods. These superior results highlight the potential of CFNet to support clinical decision‐making and treatment planning.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Biomedical Engineering, South-Central Minzu University, Wuhan, Hubei, China.
- Experimental Teaching and Engineering Training Center, South-Central Minzu University, Wuhan, Hubei, China.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Hubei Province of China
- Fundamental Research Funds for the Central Universities of South-Central Minzu University
Zitieren
APA
Chen, Y., Lu, X., Chen, H., Chen, T., & Xie, Q. (2026). Complementary fusion network of scaling attention and global attention for volumetric medical image segmentation. Medical Physics, 53(3). https://doi.org/10.1002/mp.70345
Vancouver
Chen Y, Lu X, Chen H, Chen T, Xie Q. Complementary fusion network of scaling attention and global attention for volumetric medical image segmentation. Medical Physics. 2026;53(3). doi:10.1002/mp.70345
BibTeX
@article{Chen2026,
author = {Chen, Yong and Lu, Xuesong and Chen, Hong and Chen, Tong and Xie, Qinlan},
title = {Complementary fusion network of scaling attention and global attention for volumetric medical image segmentation},
journal = {Medical Physics},
year = {2026},
volume = {53},
number = {3},
doi = {10.1002/mp.70345}
}
EUVIMED