DCLA-UNet: dynamic cross-layer compressive attention and large axial separable convolution for multi-modal MRI brain tumor segmentation
PeerJ Computer Science · 05.08.2026 · 12 · e4055
DOI: 10.7717/peerj-cs.4055 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 05.08.2026. Verfasst von Yunyan Wang und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Accurate brain tumor segmentation from multi-modal Magnetic Resonance Imaging (MRI) is critical for clinical prognosis, yet existing 3D architectures face a persistent dilemma. Traditional Convolutional Neural Network (CNN) suffer from restricted receptive fields, while emerging Vision Transformers incur prohibitive quadratic computational complexity and excessive parameterization, severely hindering their deployment on resource-constrained clinical devices.
Methods
To address this fundamental trade-off between global contextual modeling and computational efficiency, this study proposes DCLA-UNet, a highly parameter-efficient 3D segmentation framework driven by two core architectural innovations. First, we design a lightweight encoder-decoder structure integrating a Slim Large-Kernel Module (SLKM) that utilizes depthwise axial convolutions to enlarge the effective receptive field, coupled with a Multi-Scale Fusion Module (MSFM) employing parallel dilated branches for robust semantic reconstruction. Second, we introduce a Dynamic Cross-Layer Compressive Attention (DCLA) mechanism that leverages extreme single-channel compression and multi-kernel spatial alignment to adaptively modulate bottleneck features, effectively bridging the semantic gap and suppressing background noise.
Results
Extensive evaluations across the BraTS (2019, 2020, 2021) and MSD BrainTumour bench- marks systematically validate the efficacy and generalization capability of DCLA-UNet. The proposed architecture requires only 0.53M trainable parameters—a 94.76% reduction compared to the 3D U-Net baseline—and consumes just 39.23 Giga Floating Point Operations (GFLOPs). Despite this extreme compression, DCLA-UNet achieves a competitive mean Dice Similarity Coefficient (DSC) of 87.5 ± 0.6% on BraTS2021, outperforming emerging lightweight architectures such as SegFormer3D and Mamba3D. While it matches the DSC of heavier models like MogaNet using only 9.2% of its parameters, a trade-off is observed in the 95th percentile Hausdorff Distance (HD_95) (7.6 ± 1.8), objectively reflecting the inherent limits of boundary precision under strict parameter constraints.
Conclusion
DCLA-UNet successfully breaks the conventional bottleneck of 3D global context modeling by achieving a competitive balance between parameter efficiency and segmentation accuracy. It provides a robust, parameter-efficient solution for 3D medical image segmentation in storage-constrained clinical scenarios. (The source code is available at https://github.com/Helium-327/DCLA-UNet-3D ).
Zitieren
APA
Wang, Y., Xiong, J., & Xia, C. (2026). DCLA-UNet: dynamic cross-layer compressive attention and large axial separable convolution for multi-modal MRI brain tumor segmentation. PeerJ Computer Science, 12, e4055. https://doi.org/10.7717/peerj-cs.4055
Vancouver
Wang Y, Xiong J, Xia C. DCLA-UNet: dynamic cross-layer compressive attention and large axial separable convolution for multi-modal MRI brain tumor segmentation. PeerJ Computer Science. 2026;12:e4055. doi:10.7717/peerj-cs.4055
BibTeX
@article{Wang2026,
author = {Wang, Yunyan and Xiong, Junyin and Xia, Congling},
title = {DCLA-UNet: dynamic cross-layer compressive attention and large axial separable convolution for multi-modal MRI brain tumor segmentation},
journal = {PeerJ Computer Science},
year = {2026},
volume = {12},
pages = {e4055},
doi = {10.7717/peerj-cs.4055}
}
EUVIMED