SCRNet: Spatial-Channel Regulation Network for Medical Ultrasound Image Segmentation

TL;DR

SCRNet is a novel neural network architecture that combines convolution and cross-attention mechanisms to improve medical ultrasound image segmentation by effectively capturing both local and long-range features.

Contribution

The paper introduces SCRNet, integrating a new Feature Aggregation Module and Spatial-Channel Regulation Module into UNet, enhancing segmentation accuracy over existing methods.

Findings

SCRNet achieves state-of-the-art segmentation performance.

The proposed modules improve focus on salient regions.

Extensive experiments validate the effectiveness of SCRNet.

Abstract

Medical ultrasound image segmentation presents a formidable challenge in the realm of computer vision. Traditional approaches rely on Convolutional Neural Networks (CNNs) and Transformer-based methods to address the intricacies of medical image segmentation. Nevertheless, inherent limitations persist, as CNN-based methods tend to disregard long-range dependencies, while Transformer-based methods may overlook local contextual information. To address these deficiencies, we propose a novel Feature Aggregation Module (FAM) designed to process two input features from the preceding layer. These features are seamlessly directed into two branches of the Convolution and Cross-Attention Parallel Module (CCAPM) to endow them with different roles in each of the two branches to help establish a strong connection between the two input features. This strategy enables our module to focus concurrently on both long-range dependencies and local contextual information by judiciously merging convolution operations with cross-attention mechanisms. Moreover, by integrating FAM within our proposed Spatial-Channel Regulation Module (SCRM), the ability to discern salient regions and informative features warranting increased attention is enhanced. Furthermore, by incorporating the SCRM into the encoder block of the UNet architecture, we introduce a novel framework dubbed Spatial-Channel Regulation Network (SCRNet). The results of our extensive experiments demonstrate the superiority of SCRNet, which consistently achieves state-of-the-art (SOTA) performance compared to existing methods.