EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Nested Attention Network for Robust Medical Image Segmentation Under Digital Watermarking

Mohammad J M Zedan iD, Ahmed A Mohammed iD, Mohammed A M Abdullah iD, Ersin Elbasi iD, Wai Lok Woo iD, Mohd Asyraf Zulkifley iD

Biomimetics · 08.07.2026 · 11 (7) · 475

DOI: 10.3390/biomimetics11070475 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomimetics, online erschienen am 08.07.2026. Verfasst von Mohammad J M Zedan und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer and Information Engineering, Ninevah University, Mosul 41002, Iraq. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Digital watermarking is widely used to protect medical images in terms of ownership, authenticity, and traceability; however, the embedding process may introduce subtle modifications that can affect the reliability of deep-learning-based clinical analysis. Existing studies have shown that watermarking has a negligible effect on medical image classification; nevertheless, its impact on segmentation performance remains insufficiently explored. Therefore, this paper aims to investigate the effects of segmentation model enhancement on watermarked medical image analysis. In this context, three representative watermarking approaches were employed, and five baseline segmentation models, namely U-Net, ResUNet++, SegNet, FCDenseNet, and TernausNet, were evaluated on two benchmark datasets: LIDC-IDRI and BRISC. Additionally, a novel deep learning model with nested attention mechanisms was specifically designed to improve feature extraction and increase sensitivity to subtle pixel-level variations in watermarked images. Segmentation performance was assessed using five standard evaluation metrics, including mean Intersection over Union (mIoU), Dice Similarity Coefficient (DSC), and the 95th percentile Hausdorff Distance (HD95). The experimental results indicate consistently minor performance degradation across both datasets. For the BRISC dataset, the reduction in mIoU ranges from 0.15% to 0.44%, while for the LIDC-IDRI dataset, it ranges from 0.19% to 0.29% compared with the no-watermarking baseline. These findings provide quantitative insight into the compatibility of watermarking techniques for medical image protection with AI-based medical image segmentation systems, highlighting their potential for broader clinical application.

Schlagwörter

Stichwörter der Autor:innen

deep learningdigital watermarkingdiscrete wavelet transformleast significant bitmedical imagingsemantic segmentationsingular value decomposition

Autor:innen und Institutionen

  • Department of Computer and Information Engineering, Ninevah University, Mosul 41002, Iraq.
  • College of Engineering and Technology, American University of the Middle East, Egaila 54200, Kuwait.
  • Department of Computer and Information Sciences, Northumbria University, Newcastle upon Tyne NE7 7XA, UK.
  • Department of Electrical, Electronic and Systems Engineering, Universiti Kebangsaan Malaysia, Bangi 43600, Malaysia.

ORCID-Profile: Mohammad J M Zedan, Ahmed A Mohammed, Mohammed A M Abdullah, Ersin Elbasi, Wai Lok Woo, Mohd Asyraf Zulkifley

Zitieren

APA

Zedan, M. J. M., Mohammed, A. A., Abdullah, M. A. M., Elbasi, E., Woo, W. L., & Zulkifley, M. A. (2026). Nested Attention Network for Robust Medical Image Segmentation Under Digital Watermarking. Biomimetics, 11(7), 475. https://doi.org/10.3390/biomimetics11070475

Vancouver

Zedan MJM, Mohammed AA, Abdullah MAM, Elbasi E, Woo WL, Zulkifley MA. Nested Attention Network for Robust Medical Image Segmentation Under Digital Watermarking. Biomimetics. 2026;11(7):475. doi:10.3390/biomimetics11070475

BibTeX
@article{Zedan2026,
  author = {Zedan, Mohammad J M and Mohammed, Ahmed A and Abdullah, Mohammed A M and Elbasi, Ersin and Woo, Wai Lok and Zulkifley, Mohd Asyraf},
  title = {Nested Attention Network for Robust Medical Image Segmentation Under Digital Watermarking},
  journal = {Biomimetics},
  year = {2026},
  volume = {11},
  number = {7},
  pages = {475},
  doi = {10.3390/biomimetics11070475}
}
Kostenlos starten →