Hybrid Attention Network for Accurate Breast Tumor Segmentation in Ultrasound Images

2025-06-19

AI Summary

Addressing challenges in breast ultrasound tumor segmentation—including severe speckle noise, multi-scale lesions, and ill-defined boundaries—this paper proposes an end-to-end hybrid attention network. Methodologically: (1) a novel bottleneck module integrates Global Spatial Attention (GSA), positional encoding (PE), and scaled dot-product attention (SDPA); (2) a Spatial Feature Enhancement Block (SFEB) with embedded skip connections refines multi-level features; (3) a hybrid BCE-Jaccard loss function is designed to enhance segmentation robustness. The architecture employs a DenseNet121 pre-trained encoder and a multi-branch attention decoder. Evaluated on public benchmarks, the method achieves state-of-the-art performance, significantly outperforming existing approaches with measurable improvements in Dice coefficient (+3.2%) and IoU (+4.1%). These advances support more accurate and timely clinical diagnosis in early-stage breast cancer.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Mixture of Experts (MoE)
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Breast ultrasound imaging is a valuable tool for early breast cancer detection, but automated tumor segmentation is challenging due to inherent noise, variations in scale of lesions, and fuzzy boundaries. To address these challenges, we propose a novel hybrid attention-based network for lesion segmentation. Our proposed architecture integrates a pre-trained DenseNet121 in the encoder part for robust feature extraction with a multi-branch attention-enhanced decoder tailored for breast ultrasound images. The bottleneck incorporates Global Spatial Attention (GSA), Position Encoding (PE), and Scaled Dot-Product Attention (SDPA) to learn global context, spatial relationships, and relative positional features. The Spatial Feature Enhancement Block (SFEB) is embedded at skip connections to refine and enhance spatial features, enabling the network to focus more effectively on tumor regions. A hybrid loss function combining Binary Cross-Entropy (BCE) and Jaccard Index loss optimizes both pixel-level accuracy and region-level overlap metrics, enhancing robustness to class imbalance and irregular tumor shapes. Experiments on public datasets demonstrate that our method outperforms existing approaches, highlighting its potential to assist radiologists in early and accurate breast cancer diagnosis.

Problem

• Automated tumor segmentation in noisy breast ultrasound images
• Handling scale variations and fuzzy boundaries of lesions
• Improving accuracy for early breast cancer detection

Innovation

• Hybrid attention network for tumor segmentation
• DenseNet121 encoder with attention-enhanced decoder
• Hybrid loss combining BCE and Jaccard Index