EUVIMEDEUVIMED entdecken
Crossref

EA‐Net: Edge Attention Network for Brain Tumour Segmentation in MRI

Meng Fansheng iD, Duan Xingguang, Song Xinya, Fang Fengxinyun, Tian Jiexi, Wang Xujia, Chen Yu, Zhao Yuanli, Li Changsheng iD

The International Journal of Medical Robotics and Computer Assisted Surgery · 2026-04 · 22 (2)

DOI: 10.1002/rcs.70163 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The International Journal of Medical Robotics and Computer Assisted Surgery, online erschienen am 11.04.2026. Verfasst von Meng Fansheng und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechatronical Engineering, Beijing Institute of Technology, Beijing, China. Hauptthemen laut MEDLINE: Brain Neoplasms, Magnetic Resonance Imaging, Image Processing, Computer-Assisted. Die Arbeit nennt 40 Literaturangaben. Gefördert von: National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate brain tumour segmentation is crucial for clinical diagnosis and treatment planning, yet remains challenging due to the scale diversity of tumour regions, ambiguous boundary structures, and highly irregular shapes.

Methods

We propose a novel Edge Attention Network that integrates two key components: a Multi‐Scale Context Fusion Module to dynamically adjust receptive fields and capture multi‐scale contextual information, and an Edge Segmentation Module that explicitly extracts tumour boundaries and injects them into the backbone as spatial attention to refine segmentation details, particularly at edges.

Results

Experiments show that our model achieves Dice coefficients of 90.37% for Tumour Core (TC) and 88.91% for Whole Tumour (WT) on the BraTS2021 dataset. In cross‐dataset generalisation tests on BTM‐PVS, it maintains strong performance with 75.20% TC and 74.20% WT.

Conclusions

The proposed method demonstrates superior segmentation accuracy and robust generalisation capability, highlighting its clinical potential and offering new insights for medical image segmentation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Brain NeoplasmsMagnetic Resonance ImagingImage Processing, Computer-AssistedHumansAlgorithmsImage Interpretation, Computer-AssistedNeural Networks, Computer

Stichwörter der Autor:innen

brain tumour segmentationcontextual informationedge attention

Autor:innen und Institutionen

  • School of Mechatronical Engineering, Beijing Institute of Technology, Beijing, China.
  • Key Laboratory of Biomimetic Robots and Systems, Beijing Institute of Technology, Ministry of Education, Beijing, China.
  • School of Medical Technology, Beijing Institute of Technology, Beijing, China.
  • Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China.
  • Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.

ORCID-Profile: Meng Fansheng, Li Changsheng

Förderung

  • National Key Research and Development Program of China

Zitieren

APA

Fansheng, M., Xingguang, D., Xinya, S., Fengxinyun, F., Jiexi, T., Xujia, W., Yu, C., Yuanli, Z., & Changsheng, L. (2026). EA‐Net: Edge Attention Network for Brain Tumour Segmentation in MRI. The International Journal of Medical Robotics and Computer Assisted Surgery, 22(2). https://doi.org/10.1002/rcs.70163

Vancouver

Fansheng M, Xingguang D, Xinya S, Fengxinyun F, Jiexi T, Xujia W, et al. EA‐Net: Edge Attention Network for Brain Tumour Segmentation in MRI. The International Journal of Medical Robotics and Computer Assisted Surgery. 2026;22(2). doi:10.1002/rcs.70163

BibTeX
@article{Fansheng2026,
  author = {Fansheng, Meng and Xingguang, Duan and Xinya, Song and Fengxinyun, Fang and Jiexi, Tian and Xujia, Wang and Yu, Chen and Yuanli, Zhao and Changsheng, Li},
  title = {EA‐Net: Edge Attention Network for Brain Tumour Segmentation in MRI},
  journal = {The International Journal of Medical Robotics and Computer Assisted Surgery},
  year = {2026},
  volume = {22},
  number = {2},
  doi = {10.1002/rcs.70163}
}
Kostenlos starten →