BreAST-U²Net:基于注意力肿瘤融合的双流U²Net用于自动乳腺超声二维肿瘤分割
该研究提出BreAST-U²Net双流神经网络,利用注意力融合机制整合单个三维自动乳腺超声(ABUS)容积内相邻二维轴位层面的肿瘤特征,在TDSC数据集上分割质量可接受(Dice≥0.1)的病例中Dice达66.4%,在UMMC数据集上达65.2%,全测试集总体Dice分别为38.1%和23.8%。
要点
- 在分割质量可接受(Dice≥0.1)的病例中,模型Dice在TDSC数据集为66.4%、UMMC数据集为65.2%。
- 全测试集总体Dice较低:TDSC为38.1%、UMMC为23.8%。
- 双流编码与通道注意力融合使模型对声影伪影和域偏移表现稳健。
核心结论
- BreAST-U²Net在源数据集和未见过的ABUS数据集上均表现出域稳健的分割性能,在预测质量可接受时Dice最高达66.4%,并通过双流编码和高效通道注意力特征融合对声影伪影表现出韧性。
研究设计与样本
- 研究为方法学论文,提出并评估BreAST-U²Net双流U²Net编码器结合基于高效通道注意力的融合模块和共享解码器,用于三维ABUS中的乳腺肿瘤分割,采用Dice评分进行消融研究并与DSGMFFN和DEMAC-Net等先进方法基准比较。
诊断标准与关键数值
- 分割质量可接受的判定标准为Dice≥0.1,在此条件下TDSC数据集肿瘤检测、分割与分类任务的Dice为66.4%。
- 分割质量可接受的判定标准为Dice≥0.1,在此条件下UMMC数据集Dice为65.2%。
- 全测试集总体Dice在TDSC数据集为38.1%。
- 全测试集总体Dice在UMMC数据集为23.8%。
临床应用与阅片要点
- 模型在部分声影阈值下持续位列前茅,对声学伪影和域偏移表现稳健。
- 某些肿瘤复杂程度可能仍需进一步调参,作者据此解释全测试集总体Dice较温和的原因。
局限性
- 某些肿瘤复杂程度可能仍需进一步调参。
- 全测试集总体Dice较温和,TDSC为38.1%、UMMC为23.8%。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND: Automated breast ultrasound (ABUS) shows potential for breast cancer diagnosis but faces tumor segmentation challenges due to limited annotated data and the tendency of deep convolution neural networks to introduce irrelevant features, hindering performance. In this paper we introduce BreAST-U²Net, a collaborative twin-stream neural network featuring an attention-based fusion mechanism. This hybrid architecture enables the model to collaboratively enrich tumor features from neighboring 2-D axial slices within a single 3-D ABUS volume, allowing for minimal short-term spatial continuity between ABUS slices while maintaining competitive and robust performance.
METHODS: BreAST-U²Net combines a twin-stream U²Net encoder with an efficient channel attention-based fusion module and a shared decoder tailored for breast tumor segmentation in 3-D ABUS. We evaluated its performance using Dice score, as demonstrated in ablation studies comparing BreAST-U²Net with its single-stream and twin-stream variants. Additionally, the model was benchmarked against state-of-the-art methods, including DSGMFFN and DEMAC-Net.
RESULTS: BreAST-U²Net achieved strong performance in cases with acceptable segmentation quality (Dice ≥0.1), reaching 66.4% on Tumor Detection, Segmentation, and Classification (TDSC) and 65.2% on Universiti Malaya Medical Center (UMMC). It consistently ranked among the top models under certain shadow thresholds, demonstrating robustness to acoustic artifacts and domain shifts. However, certain levels of tumor complexity may still require further tuning, which helps to explain why overall Dice scores on the full test sets remained more modest-38.1% for TDSC and 23.8% for UMMC.
CONCLUSION: BreAST-U²Net demonstrates domain-robust segmentation performance across both source and unseen ABUS datasets, achieving Dice scores of up to 66.4% with acceptable prediction quality and showing resilience to shadow artifacts through its dual-stream encoding and efficient channel attention-based feature fusion.
引用
- Haqkiem N, Mohd Faizal Abdullah ER, Saw SN, et al. BreAST-U²Net: A Twin-Stream U2Net with Attention-based Tumor Fusion for 2-D Tumor Segmentation in Automated Breast Ultrasound. Ultrasound Med Biol. 2025. PMID: 41177730. doi:10.1016/j.ultrasmedbio.2025.09.011
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。