通过更精确的分区校正和渐进的硬增强学习来打击医疗标签噪音
Sanyan Zhang1, Surong Chu1, Yan Qiang2
1Imaging & Intelligence Lab, Taiyuan University of Technology, China.
Computer methods and programs in biomedicine
|April 1, 2025
概括
这项研究引入了一个新的框架,通过解决噪音标签来提高医疗图像分类准确性. 该方法有效地纠正了标签噪声,提高了对具有挑战性的数据集的诊断模型性能.
科学领域:
- 医疗成像医学成像
- 人工智能的人工智能
- 计算机辅助诊断 计算机辅助诊断
背景情况:
- 医疗诊断的深度神经网络需要高质量的标记数据.
- 由于复杂性和所需的专业知识,医疗图像的手动注释可能会引入标签噪声.
- 标签噪声可能会对分类模型的训练和性能产生负面影响.
研究的目的:
- 为医疗图像分类开发一个耐噪框架.
- 为了有效地解决和减轻医疗数据集中的标签噪声带来的挑战.
- 提高计算机辅助诊断系统的准确性和稳定性.
主要方法:
- 一个两阶段的框架:预训练校正和渐进的硬样本增强学习.
- 双分支样本分区可以检测干净,硬和杂的实例.
- 硬样标签的精细化和联合校正提高了数据质量.
- 渐进强化学习可以改善特征表示学习.
主要成果:
- 在肺结核病数据集上获得了82.39%的准确性.
- 在五类皮肤病数据集上表现出强大的性能,噪音水平各不相同 (高达40%).
- 在二元聚分类中表现出高准确度,即使有显著的标签噪声 (高达40%).
结论:
- 拟议的框架有效地处理医疗图像分类中的标签噪声.
- 在各种数据集和噪音水平中证明了稳定性和有效性.
- 验证了改进计算机辅助诊断系统与噪音标签的潜力.
相关概念视频
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...