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.

More than Encoder: Introducing Transformer Decoder to Upsample

[Multi-scale medical image segmentation based on pixel encoding and spatial attention mechanism]

CSWin-UNet: Transformer UNet with Cross-Shaped Windows for Medical Image Segmentation

DW-SCA Unet: medical image segmentation based on depth-wise separable convolutional attention U-shaped network

More complex encoder is not all you need

DGFAU-Net: Global feature attention upsampling network for medical image segmentation

TransAttUnet: Multi-level Attention-guided U-Net with Transformer for Medical Image Segmentation.

HAD-Net: an Attention U-based Network with Hyper-Scale Shifted Aggregating and Max-Diagonal Sampling for Medical Image Segmentation

Transformer and group parallel axial attention co-encoder for medical image segmentation

Enhanced Transformer Encoder and Hybrid Cascaded Upsampler for Medical Image Segmentation.

DAU-Net: A medical image segmentation network combining the Hadamard product and dual scale attention gate

Narrowing the semantic gaps in U-Net with learnable skip connections: The case of medical image segmentation

A pixel and channel enhanced up-sampling module for biomedical image segmentation

Optimizing Medical Image Segmentation with Advanced Decoder Design

A Improved Spatial-Reduction Attention Transformer for Medical Image Segmentation

TSCA-Net: Transformer based spatial-channel attention segmentation network for medical images

MoAFormer: Aggregating Adjacent Window Features into Local Vision Transformer Using Overlapped Attention Mechanism for Volumetric Medical Segmentation.

ASYMMETRIC ATTENTION UPSAMPLING: RETHINKING UPSAMPLING FOR BIOLOGICAL IMAGE SEGMENTATION

A Multi-Scale Cross-Fusion Medical Image Segmentation Network Based on Dual-Attention Mechanism Transformer

Attention Mechanism Trained with Small Datasets for Biomedical Image Segmentation

GAUNet: Gated Attention U-Net for Medical Image Segmentation