Dual Attention Residual U-Net for Accurate Brain Ultrasound Segmentation in IVH Detection

TL;DR

This paper introduces a novel dual attention residual U-Net architecture that significantly improves brain ultrasound segmentation accuracy for IVH detection by combining spatial and channel attention mechanisms with sparse attention filtering.

Contribution

The work presents a new Residual U-Net model with integrated CBAM and SAL modules, enhancing feature refinement and noise suppression for better segmentation performance.

Findings

Achieved a Dice score of 89.04% on Brain US dataset

Outperformed existing methods in IVH segmentation accuracy

Demonstrated robustness in noisy ultrasound images

Abstract

Intraventricular hemorrhage (IVH) is a severe neurological complication among premature infants, necessitating early and accurate detection from brain ultrasound (US) images to improve clinical outcomes. While recent deep learning methods offer promise for computer-aided diagnosis, challenges remain in capturing both local spatial details and global contextual dependencies critical for segmenting brain anatomies. In this work, we propose an enhanced Residual U-Net architecture incorporating two complementary attention mechanisms: the Convolutional Block Attention Module (CBAM) and a Sparse Attention Layer (SAL). The CBAM improves the model's ability to refine spatial and channel-wise features, while the SAL introduces a dual-branch design, sparse attention filters out low-confidence query-key pairs to suppress noise, and dense attention ensures comprehensive information propagation. Extensive experiments on the Brain US dataset demonstrate that our method achieves state-of-the-art segmentation performance, with a Dice score of 89.04% and IoU of 81.84% for ventricle region segmentation. These results highlight the effectiveness of integrating spatial refinement and attention sparsity for robust brain anatomy detection. Code is available at: https://github.com/DanYuan001/BrainImgSegment.