DCSAU-Net: A deeper and more compact split-attention U-Net for medical image segmentation

2023/02/01 by Qing Xu, Zhicheng Ma, Na HE +2 · 444 citations
Computer Science · Mathematics · Medicine · #AI in cancer detection #Advanced Neural Network Applications #Artificial intelligence #Block (permutation group theory) #COVID-19 diagnosis using AI #Cartography #Code (set theory) #Computer science #Convolutional neural network #Deep learning #Encoder #Feature (linguistics) #Field (mathematics) #Image (mathematics) #Image segmentation #Intersection (aeronautics) #Mathematics #Net (polyhedron) #Pattern recognition (psychology) #Range (aeronautics) #Segmentation #Sørensen–Dice coefficient #Upsampling

paper · doi:10.1016/j.compbiomed.2023.106626

published in Computers in Biology and Medicine 154, 106626 (Elsevier BV)

openalex publication_date 2023/02/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

Deep learning architecture with convolutional neural network achieves outstanding success in the field of computer vision. Where U-Net has made a great breakthrough in biomedical image segmentation and has been widely applied in a wide range of practical scenarios. However, the equal design of every downsampling layer in the encoder part and simply stacked convolutions do not allow U-Net to extract sufficient information of features from different depths. The increasing complexity of medical images brings new challenges to the existing methods. In this paper, we propose a deeper and more compact split-attention u-shape network, which efficiently utilises low-level and high-level semantic information based on two frameworks: primary feature conservation and compact split-attention block. We evaluate the proposed model on CVC-ClinicDB, 2018 Data Science Bowl, ISIC-2018, SegPC-2021 and BraTS-2021 datasets. As a result, our proposed model displays better performance than other state-of-the-art methods in terms of the mean intersection over union and dice coefficient. More significantly, the proposed model demonstrates excellent segmentation performance on challenging images. The code for our work and more technical details can be found at https://github.com/xq141839/DCSAU-Net.

Citations

• A Threshold Selection Method from Gray-Level Histograms

Cited by

• DTAN: Diffusion-based Text Attention Network for medical image segmentation
• Guided-attention and gated-aggregation network for medical image segmentation
• Co-Seg++: Mutual Prompt-Guided Collaborative Learning for Versatile Medical Segmentation
• MM-UNet: Morph Mamba U-shaped Convolutional Networks for Retinal Vessel Segmentation
• Segmentation of Multiple Sclerosis Lesions in MRI Using U‐Net With a Wavelet Transformer‐Based Attention Mechanism
• DTEA: Dynamic Topology Weaving and Instability-Driven Entropic Attenuation for Medical Image Segmentation
• MSM-Seg: A Modality-and-Slice Memory Framework with Category-Agnostic Prompting for Multi-Modal Brain Tumor Segmentation
• DyGLNet: Hybrid Global-Local Feature Fusion with Dynamic Upsampling for Medical Image Segmentation
• Co-Seg: Mutual Prompt-Guided Collaborative Learning for Tissue and Nuclei Segmentation
• TCSAFormer: Efficient Vision Transformer with Token Compression and Sparse Attention for Medical Image Segmentation
• MambaVesselNet++: A Hybrid CNN-Mamba Architecture for Medical Image Segmentation
• DistMedVL: Distributional Vision-Language Alignment for Uncertainty-Aware Medical Image Segmentation
• MHAU-Net: Skin Lesion Segmentation Based on Multi-Scale Hybrid Residual Attention Network. [europepmc]
• Enhancing skin lesion segmentation with a fusion of convolutional neural networks and transformer models. [europepmc]
• Hybrid deep spatial and statistical feature fusion for accurate MRI brain tumor classification. [europepmc]
• ELA-Net: An Efficient Lightweight Attention Network for Skin Lesion Segmentation. [europepmc]
• Attention-based multi-residual network for lung segmentation in diseased lungs with custom data augmentation. [europepmc]
• An Efficient Retinal Fluid Segmentation Network Based on Large Receptive Field Context Capture for Optical Coherence Tomography Images. [europepmc]
• Advances in Deep Learning for Semantic Segmentation of Low-Contrast Images: A Systematic Review of Methods, Challenges, and Future Directions. [europepmc]
• DEFIF-Net: A lightweight dual-encoding feature interaction fusion network for medical image segmentation. [europepmc]
• DCM-Net: dual-encoder CNN-Mamba network with cross-branch fusion for robust medical image segmentation. [europepmc]
• Multi-interactive feature embedding learning for medical image segmentation. [europepmc]
• Image segmentation network based on enhanced dual encoder. [europepmc]
• Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction. [europepmc]
• [A multi-scale feature capturing and spatial position attention model for colorectal polyp image segmentation]. [europepmc]
• Gated Attention-Augmented Double U-Net for White Blood Cell Segmentation. [europepmc]
• MSDTCN-Net: A Multi-Scale Dual-Encoder Network for Skin Lesion Segmentation. [europepmc]
• DMFF-Net: a multi-scale feature fusion network based on DeepLabV3 for skin lesion segmentation. [europepmc]
• Enhanced feature dynamic fusion gated UNet for robust retinal vessel segmentation. [europepmc]
• Dual-branch attention network with deep split convolution and multi-dimensional transformers for medical image segmentation. [europepmc]
• Multi-level semi-supervised GAN enables accurate segmentation of breast ultrasound images. [europepmc]
• Benchmarking MedViT and hybrid CNN-ViT architectures for multi-label thoracic disease classification. [europepmc]
• Efficient Medical Image Segmentation in Multisensor Imaging: A Survey in the Era of Mamba and Foundation Models. [europepmc]
• DSMUNet: A Lightweight Model for Road Crack Segmentation. [europepmc]
• Dynamic U-shaped convolutional network for mouse cardiac image segmentation and quantification. [europepmc]
• Early Diagnosis of Parkinson’s Disease Through Lite HGWA-Net Model: A Hybrid CNN Based on Wavelet Transform and Attention Mechanism [europepmc]
• LTM-UNet: Linear Transformer–Mamba with Attention-Based U-Net for Context-Aware Breast Ultrasound Image Segmentation [europepmc]

Related

• DCSAU-Net: A Deeper and More Compact Split-Attention U-Net for Medical Image Segmentation
• A hybrid approach for improving U-Net variants in medical image segmentation
• Multi-level Context Gating of Embedded Collective Knowledge for Medical Image Segmentation
• DoubleU-Net: A Deep Convolutional Neural Network for Medical Image Segmentation
• U-Net Using Stacked Dilated Convolutions for Medical Image Segmentation