Abstract:The understanding of long-range pixel–pixel dependencies plays a vital role in image segmentation. The use of a CNN plus an attention mechanism still has room for improvement, since existing transformer-based architectures require many thousands of annotated training samples to model long-range spatial dependencies. This paper presents a smooth attention branch (SAB), a novel architecture that simplifies the understanding of long-range pixel–pixel dependencies for biomedical image segmentation in small datasets. The SAB is essentially a modified attention operation that implements a subnetwork via reshaped feature maps instead of directly calculating a softmax value over the attention score for each input. The SAB fuses multilayer attentive feature maps to learn visual attention in multilevel features. We also introduce position blurring and inner cropping specifically for small-scale datasets to prevent overfitting. Furthermore, we redesign the skip pathway for the reduction of the semantic gap between every captured feature of the contracting and expansive path. We evaluate the architecture of U-Net with the SAB (SAB-Net) by comparing it with the original U-Net and widely used transformer-based models across multiple biomedical image segmentation tasks related to the Brain MRI, Heart MRI, Liver CT, Spleen CT, and Colonoscopy datasets. Our training set was made of random 100 images of the original training set, since our goal was to adopt attention mechanisms for biomedical image segmentation tasks with small-scale labeled data. An ablation study conducted on the brain MRI test set demonstrated that every proposed method achieved an improvement in biomedical image segmentation. Integrating the proposed methods helped the resulting models consistently achieve outstanding performance on the above five biomedical segmentation tasks. In particular, the proposed method with U-Net improved its segmentation performance over that of the original U-Net by 13.76% on the Brain MRI dataset. We proposed several novel methods to address the need for modeling long-range pixel–pixel dependencies in small-scale biomedical image segmentation. The experimental results illustrated that each method could improve the medical image segmentation accuracy to various degrees. Moreover, SAB-Net, which integrated all proposed methods, consistently achieved outstanding performance on the five biomedical segmentation tasks.

Nuclei Segmentation Using Cascaded Bilateral Attention U-Net.

AL-Net: Attention Learning Network based on Multi-Task Learning for Cervical Nucleus Segmentation

Research on the Nucleus Segmentation Method Based on Improved U-Net Network

Dual U-Net for the Segmentation of Overlapping Glioma Nuclei.

Multi-task Generative Adversarial Learning for Nuclei Segmentation with Dual Attention and Recurrent Convolution

MIU-Net: MIX-Attention and Inception U-Net for Histopathology Image Nuclei Segmentation

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

Convolutional Blur Attention Network for Cell Nuclei Segmentation

DSCA-Net: Double-stage Codec Attention Network for automatic nuclear segmentation

DAN-NucNet: A Dual Attention Based Framework for Nuclei Segmentation in Cancer Histology Images under Wild Clinical Conditions

Attention Based Multi-Scale Nested Network for Biomedical Image Segmentation

A Hybrid-Attention Nested UNet for Nuclear Segmentation in Histopathological Images

Multi-Scale TransUnet Combined with CBAM for Nuclear Image Segmentation.

Cervical Cell Segmentation Method Based on Global Dependency and Local Attention

An Integrative Segmentation Framework for Cell Nucleus of Fluorescence Microscopy

Cervical cell’s nucleus segmentation through an improved UNet architecture

Enhancing Nucleus Segmentation with HARU-Net: A Hybrid Attention Based Residual U-Blocks Network

Scalable model for segmenting Cells' Nuclei using the U-NET architecture

Hard-Boundary Attention Network for Nuclei Instance Segmentation

Attention Mechanism Trained with Small Datasets for Biomedical Image Segmentation

Improved Unet brain tumor image segmentation based on GSConv module and ECA attention mechanism