用知识蒸馏让模型在弱监督下准确定位胸片结核病灶
Weakly Supervised Tuberculosis Localization in Chest X-rays through Knowledge Distillation
- 用教师-学生框架训练模型，无需框选标注即可定位病灶
- 在TBX11k数据集上达到0.2428的mIOU，学生模型优于教师
- 适合资源有限地区，可提升基层医疗诊断能力
结核病仍是全球主要致死病因之一，尤其在资源匮乏地区。胸部X光（CXR）是低成本、易获取的诊断工具，但需专家解读，常难以获得。尽管机器学习在结核分类上表现优异，却常依赖表面相关性且泛化能力差。构建高质量医学图像标注数据集需大量人力与经费，通常需多名专家达成一致，成本高昂。本研究采用知识蒸馏技术，训练基于ResNet50架构的卷积神经网络，在不依赖边界框标注的情况下减少虚假相关性，并实现结核异常区域定位。在TBX11k数据集上，该方法取得0.2428的mIOU得分。实验表明，学生模型性能持续优于教师模型，展现出更强鲁棒性，具备在多样化临床环境中推广的潜力。
原文摘要 · Abstract (English)
Tuberculosis (TB) remains one of the leading causes of mortality worldwide, particularly in resource-limited countries. Chest X-ray (CXR) imaging serves as an accessible and cost-effective diagnostic tool but requires expert interpretation, which is often unavailable. Although machine learning models have shown high performance in TB classification, they often depend on spurious correlations and fail to generalize. Besides, building large datasets featuring high-quality annotations for medical images demands substantial resources and input from domain specialists, and typically involves several annotators reaching agreement, which results in enormous financial and logistical expenses. This study repurposes knowledge distillation technique to train CNN models reducing spurious correlations and localize TB-related abnormalities without requiring bounding-box annotations. By leveraging a teacher-student framework with ResNet50 architecture, the proposed method trained on TBX11k dataset achieve impressive 0.2428 mIOU score. Experimental results further reveal that the student model consistently outperforms the teacher, underscoring improved robustness and potential for broader clinical deployment in diverse settings.
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。