A critic evaluation of methods for COVID-19 automatic detection from X-ray images.

A critic evaluation of methods for COVID-19 automatic detection from X-ray images.
复制标题

DOI:
10.1016/j.inffus.2021.04.008
复制
发表时间:
2021-12
期刊:
An international journal on information fusion
影响因子:
--
通讯作者:
Nanni L
Nanni L
中科院分区:
其他
文献类型:
--
作者:
Maguolo G;Nanni L

文献摘要

参考文献

被引文献

相似文献

在本文中,我们比较和评估了最近文献中用于通过 X 射线图像自动诊断 COVID-19 的不同测试协议。我们表明,使用不包含大部分肺部的 X 射线图像可以获得类似的结果。通过将 X 射线扫描的中心变成黑色并仅在图像的外部训练我们的分类器,我们能够从图像中去除肺部。因此,我们推断出几种用于识别的测试协议是不公平的,并且神经网络正在数据集中学习与 COVID-19 的存在无关的模式。最后,我们表明创建公平的测试协议是一项具有挑战性的任务,并且我们提供了一种方法来衡量特定测试协议的公平程度。在未来的研究中,我们建议使用我们的工具检查测试协议的公平性,并鼓励研究人员寻找比我们提出的技术更好的技术。
In this paper, we compare and evaluate different testing protocols used for automatic COVID-19 diagnosis from X-Ray images in the recent literature. We show that similar results can be obtained using X-Ray images that do not contain most of the lungs. We are able to remove the lungs from the images by turning to black the center of the X-Ray scan and training our classifiers only on the outer part of the images. Hence, we deduce that several testing protocols for the recognition are not fair and that the neural networks are learning patterns in the dataset that are not correlated to the presence of COVID-19. Finally, we show that creating a fair testing protocol is a challenging task, and we provide a method to measure how fair a specific testing protocol is. In the future research we suggest to check the fairness of a testing protocol using our tools and we encourage researchers to look for better techniques than the ones that we propose.
DOI: 10.1007/s10044-021-00984-y
发表时间: 2021
期刊: Pattern analysis and applications : PAA
影响因子: --
作者:
Narin A;Kaya C;Pamuk Z
通讯作者: Pamuk Z
DOI: 10.1016/s0140-6736(20)30183-5
发表时间: 2020-02-15
期刊: LANCET
影响因子: 168.9
作者:
Huang, Chaolin;Wang, Yeming;Cao, Bin
通讯作者: Cao, Bin
DOI: 10.1038/s41598-020-76550-z
发表时间: 2020-11-11
期刊: Scientific reports
影响因子: 4.6
作者:
Wang L;Lin ZQ;Wong A
通讯作者: Wong A
DOI: 10.15585/mmwr.mm6912e2
发表时间: 2020-03-27
期刊: MMWR. Morbidity and mortality weekly report
影响因子: --
作者:
CDC COVID-19 Response Team
通讯作者: CDC COVID-19 Response Team
DOI: 10.1109/jbhi.2020.3037127
发表时间: 2020-12-01
影响因子: 7.7
作者:
Tabik, S.;Gomez-Rios, A.;Herrera, F.
通讯作者: Herrera, F.