[论文解读] Robust Screening of COVID-19 from Chest X-ray via Discriminative Cost-Sensitive Learning

Tianyang Li, Zhongyi Han+4 more|arXiv (Cornell University)|Apr 27, 2020

COVID-19 diagnosis using AI参考文献 56被引用 48

一句话总结

本文提出 discriminative cost-sensitive learning (DCSL) 用于从胸部 X 光筛查 COVID-19，结合条件中心损失和分数级成本敏感学习，提升细粒度分类并降低误诊成本，在三类数据集上实现 97.01% 的准确率。

ABSTRACT

This paper addresses the new problem of automated screening of coronavirus disease 2019 (COVID-19) based on chest X-rays, which is urgently demanded toward fast stopping the pandemic. However, robust and accurate screening of COVID-19 from chest X-rays is still a globally recognized challenge because of two bottlenecks: 1) imaging features of COVID-19 share some similarities with other pneumonia on chest X-rays, and 2) the misdiagnosis rate of COVID-19 is very high, and the misdiagnosis cost is expensive. While a few pioneering works have made much progress, they underestimate both crucial bottlenecks. In this paper, we report our solution, discriminative cost-sensitive learning (DCSL), which should be the choice if the clinical needs the assisted screening of COVID-19 from chest X-rays. DCSL combines both advantages from fine-grained classification and cost-sensitive learning. Firstly, DCSL develops a conditional center loss that learns deep discriminative representation. Secondly, DCSL establishes score-level cost-sensitive learning that can adaptively enlarge the cost of misclassifying COVID-19 examples into other classes. DCSL is so flexible that it can apply in any deep neural network. We collected a large-scale multi-class dataset comprised of 2,239 chest X-ray examples: 239 examples from confirmed COVID-19 cases, 1,000 examples with confirmed bacterial or viral pneumonia cases, and 1,000 examples of healthy people. Extensive experiments on the three-class classification show that our algorithm remarkably outperforms state-of-the-art algorithms. It achieves an accuracy of 97.01%, a precision of 97%, a sensitivity of 97.09%, and an F1-score of 96.98%. These results endow our algorithm as an efficient tool for the fast large-scale screening of COVID-19.

研究动机与目标

• 解决在三类别设置中将 COVID-19 与其他肺炎和健康 X 光片区分开的挑战。
• 开发一个强调 COVID-19 风险的判别性、成本敏感的框架，以降低误诊成本。
• 利用细粒度表示学习和类别感知成本以提升筛查性能。

提出的方法

• 引入条件中心损失，以学习用于细粒度 COVID-19 分类的类平衡、判别表示。
• 提出分数级成本敏感学习，在模型输出后应用一个领域感知的成本矩阵，以在适当情况下偏向 COVID-19 的预测。
• 将这两个模块整合到可端到端优化的 Discriminative Cost-Sensitive Learning (DCSL) 框架中。
• 在一个多中心胸部 X 光数据集上进行评估，包含 239 例 COVID-19、1,000 例细菌/病毒性肺炎和 1,000 例健康图像。
• 以 VGG16 作为主干，进行 ImageNet 迁移学习、数据增强，以及 5 折交叉验证。

实验结果

研究问题

• RQ1 细粒度、成本敏感的方法能否在胸部 X 光筛查中优于标准损失函数来提升 COVID-19 的识别？
• RQ2 将类别条件信息引入中心损失以及领域感知的分数级成本矩阵是否能减少 COVID-19 的误分类？
• RQ3 在三类胸部 X 光任务中，DCSL 相对于已经建立的 CNN 架构和针对 COVID 的模型表现如何？

主要发现

• DCSL 在三类任务上实现 97.01% 的准确率、97.00% 的精确度、97.09% 的敏感性，以及 96.98% 的 F1-score。
• DCSL 在所有报告的指标上均超越 COVID-Net 及其他基线。
• 消融研究表明，条件中心损失和分数级成本敏感学习对性能提升贡献显著。
• 三类设置（COVID-19、健康、其他肺炎）显示出与所提出方法的强辨别能力。