D-Net: Dynamic large kernel with dynamic feature fusion for volumetric medical image segmentation
Biomedical Signal Processing and Control · 2026-03 · 113 · 108837
DOI: 10.1016/j.bspc.2025.108837 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biomedical Signal Processing and Control, 2026-03. Verfasst von Jin Yang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Institutes of Health National Cancer Institute, National Institutes of Health.
Abstract (Originalsprache)
Hierarchical Vision Transformers (ViTs) have achieved significant success in medical image segmentation due to their large receptive field and ability to leverage long-range contextual information. Convolutional neural networks (CNNs) may also deliver a large receptive field by using large convolutional kernels. However, because they use fixed-sized kernels, CNNs with large kernels remain limited in their ability to adaptively capture multi-scale features from organs that vary greatly in shape and size. They are also unable to utilize global contextual information efficiently. To address these limitations, we propose lightweight Dynamic Large Kernel (DLK) and Dynamic Feature Fusion (DFF) modules. The DLK employs multiple large kernels with varying kernel sizes and dilation rates to capture multi-scale features. Subsequently, DLK utilizes a dynamic selection mechanism to adaptively highlight the most important channel and spatial features based on global information. The DFF is proposed to adaptively fuse multi-scale local feature maps based on their global information. We incorporated DLK and DFF into a hierarchical ViT architecture to leverage their scaling behavior, but they struggle to extract low-level features effectively due to feature embedding constraints in ViT architectures. To tackle this limitation, we propose a Salience layer to extract low-level features from images at their original dimensions without feature embedding. This Salience layer employs a Channel Mixer to capture global representations effectively. We further incorporated the Salience layer into the hierarchical ViT architecture to develop a novel network, termed D-Net. D-Net effectively utilizes a multi-scale large receptive field and adaptively harnesses global contextual information. Extensive experimental results demonstrate its superior segmentation performance compared to state-of-the-art models, with comparably lower computational complexity. The code is made available at https://github.com/sotiraslab/DLK .
Förderung
- National Institutes of Health National Cancer Institute
- National Institutes of Health
Zitieren
APA
Yang, J., Qiu, P., Zhang, Y., Marcus, D. S., & Sotiras, A. (2026). D-Net: Dynamic large kernel with dynamic feature fusion for volumetric medical image segmentation. Biomedical Signal Processing and Control, 113, 108837. https://doi.org/10.1016/j.bspc.2025.108837
Vancouver
Yang J, Qiu P, Zhang Y, Marcus DS, Sotiras A. D-Net: Dynamic large kernel with dynamic feature fusion for volumetric medical image segmentation. Biomedical Signal Processing and Control. 2026;113:108837. doi:10.1016/j.bspc.2025.108837
BibTeX
@article{Yang2026,
author = {Yang, Jin and Qiu, Peijie and Zhang, Yichi and Marcus, Daniel S. and Sotiras, Aristeidis},
title = {D-Net: Dynamic large kernel with dynamic feature fusion for volumetric medical image segmentation},
journal = {Biomedical Signal Processing and Control},
year = {2026},
volume = {113},
pages = {108837},
doi = {10.1016/j.bspc.2025.108837}
}
EUVIMED