Cascaded attention-driven dynamic multi-scale feature fusion network for gastrointestinal polyp segmentation
Journal of King Saud University Computer and Information Sciences · 2026-09 · 38 (7)
DOI: 10.1007/s44443-026-01104-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of King Saud University Computer and Information Sciences, online erschienen am 18.08.2026. Verfasst von Hao Zhang und 3 weiteren Autor:innen. Die Arbeit nennt 85 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Henan Provincial Key Discipline of Traditional Chinese Medicine – TCM Proctology, Henan Provincial Science and Technology Research Project, Henan Provincial Joint Fund for Science and Technology Research and Development (Discipline Category) Cultivation Project.
Abstract (Originalsprache)
Intestinal endoscopic images are frequently degraded by uneven illumination, motion blur, and specular reflections, which tend to cause diffuse lesion boundaries and obscure micro-lesions. Moreover, conventional multi-scale feature fusion methods struggle to accommodate the drastic scale variations of lesions induced by fluctuating shooting distances. To address these challenges, this paper proposes a Cascaded Attention-Driven Dynamic Multi-Scale Feature Fusion Network(CADM-Net). The network first constructs a Multi-Kernel Residual Block to perform parallel multi-scale feature extraction, thereby effectively preserving the fine-grained details of micro-lesions. Subsequently, a Dynamic Multi-Scale Attention mechanism is introduced, which leverages learnable context aggregation and serial dual-attention calibration to adaptively perceive and respond to severe lesion scale variations. Finally, a dedicated Feature Fusion Module is designed to alleviate cross-level semantic misalignment and further refine diffuse lesion boundaries. These modules operate synergistically to significantly enhance the network’s robustness against endoscopic imaging degradation and ensure precise segmentation of lesion regions across varying scales. Extensive experimental results on four public polyp benchmark datasets demonstrate that CADM-Net consistently surpasses existing state-of-the-art methods, achieving superior segmentation accuracy and boundary delineation under diverse imaging conditions.
Förderung
- Henan Provincial Key Discipline of Traditional Chinese Medicine – TCM Proctology
- Henan Provincial Science and Technology Research Project
- Henan Provincial Joint Fund for Science and Technology Research and Development (Discipline Category) Cultivation Project
Zitieren
APA
Zhang, H., Zhang, X., Shi, M., & Tang, M. (2026). Cascaded attention-driven dynamic multi-scale feature fusion network for gastrointestinal polyp segmentation. Journal of King Saud University Computer and Information Sciences, 38(7). https://doi.org/10.1007/s44443-026-01104-8
Vancouver
Zhang H, Zhang X, Shi M, Tang M. Cascaded attention-driven dynamic multi-scale feature fusion network for gastrointestinal polyp segmentation. Journal of King Saud University Computer and Information Sciences. 2026;38(7). doi:10.1007/s44443-026-01104-8
BibTeX
@article{Zhang2026,
author = {Zhang, Hao and Zhang, Xiangan and Shi, Menghua and Tang, Mengmeng},
title = {Cascaded attention-driven dynamic multi-scale feature fusion network for gastrointestinal polyp segmentation},
journal = {Journal of King Saud University Computer and Information Sciences},
year = {2026},
volume = {38},
number = {7},
doi = {10.1007/s44443-026-01104-8}
}
EUVIMED