基于动态多尺度融合Transformer的DMFormer实现高效超声乳腺癌检测
研究提出基于Transformer的DMFormer模型,采用动态多尺度特征融合机制,结合窗口注意力与网格注意力,用于超声乳腺良恶性肿块鉴别。在两个独立数据集上AUC分别为90.48%和86.57%,均优于卷积神经网络、Transformer架构及混合模型等对比方法。
要点
- DMFormer在乳腺超声良恶性鉴别中两个数据集AUC分别为90.48%和86.57%。
- 模型结合窗口注意力处理局部特征与网格注意力混合全局上下文。
- 在两个独立数据集上均优于所对比的卷积神经网络、Transformer及混合模型。
核心结论
- DMFormer通过双注意力方法在超声乳腺癌检测中表现出优越性能,其平衡局部与全局特征处理并保持计算效率的能力,验证了其在临床环境中提升乳腺癌筛查准确性和可靠性的潜力。
研究设计与样本
- 研究提出基于Transformer的新型架构DMFormer,在两个独立数据集上评估,并与卷积神经网络、Transformer架构及混合模型等先进方法比较。
诊断标准与关键数值
- DMFormer在两个独立数据集上的曲线下面积分别为90.48%和86.57%,均持续优于所有对比模型。
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
OBJECTIVE: To develop an advanced deep learning model for accurate differentiation between benign and malignant masses in ultrasound breast cancer screening, addressing the challenges of noise, blur, and complex tissue structures in ultrasound imaging.
METHODS: We propose Dynamic Multiscale Fusion Transformer (DMFormer), a novel Transformer-based architecture featuring a dynamic multiscale feature fusion mechanism. The model integrates window attention for local feature interaction with grid attention for global context mixing, enabling comprehensive capture of both fine-grained tissue details and broader anatomical contexts.
RESULTS: DMFormer was evaluated on two independent datasets and compared against state-of-the-art approaches, including convolutional neural networks, Transformer-based architectures, and hybrid models. The model achieved areas under the curve of 90.48% and 86.57% on the respective datasets, consistently outperforming all comparison models.
CONCLUSION: DMFormer demonstrates superior performance in ultrasound breast cancer detection through its innovative dual-attention approach. The model's ability to effectively balance local and global feature processing while maintaining computational efficiency represents a significant advancement in medical image analysis. These results validate DMFormer's potential for enhancing the accuracy and reliability of breast cancer screening in clinical settings.
引用
- Guo L, Zhang H, Ma C, et al. Efficient Ultrasound Breast Cancer Detection with DMFormer: A Dynamic Multiscale Fusion Transformer. Ultrasound Med Biol. 2025. PMID: 40628603. doi:10.1016/j.ultrasmedbio.2025.05.017
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。