WDFFU-Mamba: A Wavelet-guided Dual-attention Feature Fusion Mamba for Breast Tumor Segmentation in Ultrasound Images

TL;DR

The paper introduces WDFFU-Mamba, a novel wavelet-guided dual-attention network that significantly improves breast tumor segmentation accuracy in ultrasound images by effectively handling noise and irregular lesion boundaries.

Contribution

It proposes a new segmentation architecture integrating wavelet-based noise suppression and dual-attention feature fusion, enhancing robustness and accuracy over existing methods.

Findings

Achieves higher Dice coefficient and lower Hausdorff Distance than existing methods.

Demonstrates strong generalization across different datasets.

Maintains computational efficiency while improving segmentation quality.

Abstract

Breast ultrasound (BUS) image segmentation plays a vital role in assisting clinical diagnosis and early tumor screening. However, challenges such as speckle noise, imaging artifacts, irregular lesion morphology, and blurred boundaries severely hinder accurate segmentation. To address these challenges, this work aims to design a robust and efficient model capable of automatically segmenting breast tumors in BUS images.We propose a novel segmentation network named WDFFU-Mamba, which integrates wavelet-guided enhancement and dual-attention feature fusion within a U-shaped Mamba architecture. A Wavelet-denoised High-Frequency-guided Feature (WHF) module is employed to enhance low-level representations through noise-suppressed high-frequency cues. A Dual Attention Feature Fusion (DAFF) module is also introduced to effectively merge skip-connected and semantic features, improving contextual consistency.Extensive experiments on two public BUS datasets demonstrate that WDFFU-Mamba achieves superior segmentation accuracy, significantly outperforming existing methods in terms of Dice coefficient and 95th percentile Hausdorff Distance (HD95).The combination of wavelet-domain enhancement and attention-based fusion greatly improves both the accuracy and robustness of BUS image segmentation, while maintaining computational efficiency.The proposed WDFFU-Mamba model not only delivers strong segmentation performance but also exhibits desirable generalization ability across datasets, making it a promising solution for real-world clinical applications in breast tumor ultrasound analysis.