Attention Swin U-Net: Cross-Contextual Attention Mechanism for Skin Lesion Segmentation

TL;DR

This paper introduces Att-SwinU-Net, an advanced skin lesion segmentation model that integrates cross-contextual attention into the Swin U-Net architecture to improve long-range dependency capture and feature re-usability.

Contribution

It proposes an attention-based extension to Swin U-Net that enhances skip connections with attention mechanisms, improving segmentation performance in medical images.

Findings

Attention mechanism improves segmentation accuracy

Enhanced feature re-usability in skip connections

Effective across multiple skin lesion datasets

Abstract

Melanoma is caused by the abnormal growth of melanocytes in human skin. Like other cancers, this life-threatening skin cancer can be treated with early diagnosis. To support a diagnosis by automatic skin lesion segmentation, several Fully Convolutional Network (FCN) approaches, specifically the U-Net architecture, have been proposed. The U-Net model with a symmetrical architecture has exhibited superior performance in the segmentation task. However, the locality restriction of the convolutional operation incorporated in the U-Net architecture limits its performance in capturing long-range dependency, which is crucial for the segmentation task in medical images. To address this limitation, recently a Transformer based U-Net architecture that replaces the CNN blocks with the Swin Transformer module has been proposed to capture both local and global representation. In this paper, we propose Att-SwinU-Net, an attention-based Swin U-Net extension, for medical image segmentation. In our design, we seek to enhance the feature re-usability of the network by carefully designing the skip connection path. We argue that the classical concatenation operation utilized in the skip connection path can be further improved by incorporating an attention mechanism. By performing a comprehensive ablation study on several skin lesion segmentation datasets, we demonstrate the effectiveness of our proposed attention mechanism.