对比卷积神经网络与多模态模型在乳腺密度评估中的表现
Comparison of ConvNeXt and Vision-Language Models for Breast Density Assessment in Screening Mammography
- 用ConvNeXt和BioMedCLIP三种学习方式对比评估
- 微调后的ConvNeXt在分类上优于BioMedCLIP线性探测
- 适合医学影像自动化分析研究者参考
乳腺密度分类对癌症风险评估至关重要，但受主观判断和观察者差异影响，仍具挑战。本研究比较了基于多模态与卷积神经网络的方法，使用BI-RADS系统，在零样本分类、文本描述线性探测和数值标签微调三种学习场景下评估BioMedCLIP与ConvNeXt。结果表明，零样本分类表现一般，而微调后的ConvNeXt模型优于BioMedCLIP线性探测。尽管线性探测利用预训练嵌入展现潜力，但仍弱于完整微调。研究提示，尽管多模态学习前景可观，但针对特定医学影像任务，端到端微调的CNN模型表现更优。未来需改进文本表征细节与领域适配。
原文摘要 · Abstract (English)
Mammographic breast density classification is essential for cancer risk assessment but remains challenging due to subjective interpretation and inter-observer variability. This study compares multimodal and CNN-based methods for automated classification using the BI-RADS system, evaluating BioMedCLIP and ConvNeXt across three learning scenarios: zero-shot classification, linear probing with textual descriptions, and fine-tuning with numerical labels. Results show that zero-shot classification achieved modest performance, while the fine-tuned ConvNeXt model outperformed the BioMedCLIP linear probe. Although linear probing demonstrated potential with pretrained embeddings, it was less effective than full fine-tuning. These findings suggest that despite the promise of multimodal learning, CNN-based models with end-to-end fine-tuning provide stronger performance for specialized medical imaging. The study underscores the need for more detailed textual representations and domain-specific adaptations in future radiology applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。