COVIDGR Dataset and COVID-SDNet Methodology for Predicting COVID-19 Based on Chest X-Ray Images
COVIDGR Dataset and COVID-SDNet Methodology for Predicting COVID-19 Based on Chest X-Ray Images
复制标题
DOI：
10.1109/jbhi.2020.3037127
复制
发表时间：
2020-12-01
影响因子：
7.7
通讯作者：
Herrera, F.
中科院分区：
文献类型：
--
作者：
Tabik, S.;Gomez-Rios, A.;Herrera, F.
目前，冠状病毒病 (COVID-19) 是 21 世纪最具传染性的疾病之一，可通过 RT-PCR 检测、CT 扫描和/或胸部 X 光 (CXR) 图像进行诊断。大多数医疗中心都没有 CT（计算机断层扫描）扫描仪和 RT-PCR 检测，因此在许多情况下 CXR 图像成为协助临床医生做出决策的最省时/最具成本效益的工具。深度学习神经网络在构建 COVID-19 分诊系统和检测 COVID-19 患者，特别是严重程度较低的患者方面具有巨大潜力。不幸的是，当前的数据库不允许构建这样的系统，因为它们高度异构并且偏向严重的情况。本文分为三部分：(i) 我们揭开了最新 COVID-19 分类模型实现的高灵敏度的神秘面纱，(ii) 在与西班牙格拉纳达圣塞西利奥大学医院的密切合作下，我们建立了 COVIDGR-1.0，这是一个同质且平衡的数据库，其中包括所有严重程度级别，从正常到阳性 RT-PCR、轻度、中度到重度。 COVIDGR-1.0 包含 426 个正向和 426 个负向 PA（PosteroAnterior）CXR 视图，并且 (iii) 我们提出了基于 COVID 智能数据的网络 (COVID-SDNet) 方法来提高 COVID 分类模型的泛化能力。我们的方法取得了良好且稳定的结果，在严重、中度和轻度 COVID-19 严重程度级别的准确度分别为 97.72% +/- 0.95%、86.90% +/- 3.20%、61.80% +/- 5.49%。我们的方法有助于及早发现 COVID-19。科学界可通过此链接 https://dasci.es/es/transferencia/open-data/covidgr/ 获取 COVIDGR-1.0 以及严重级别标签。
Currently, Coronavirus disease (COVID-19), one of the most infectious diseases in the 21st century, is diagnosed using RT-PCR testing, CT scans and/or Chest X-Ray (CXR) images. CT (Computed Tomography) scanners and RT-PCR testing are not available in most medical centers and hence in many cases CXR images become the most time/cost effective tool for assisting clinicians in making decisions. Deep learning neural networks have a great potential for building COVID-19 triage systems and detecting COVID-19 patients, especially patients with low severity. Unfortunately, current databases do not allow building such systems as they are highly heterogeneous and biased towards severe cases. This article is threefold: (i) we demystify the high sensitivities achieved by most recent COVID-19 classification models, (ii) under a close collaboration with Hospital Universitario Clinico San Cecilio, Granada, Spain, we built COVIDGR-1.0, a homogeneous and balanced database that includes all levels of severity, from normal with Positive RT-PCR, Mild, Moderate to Severe. COVIDGR-1.0 contains 426 positive and 426 negative PA (PosteroAnterior) CXR views and (iii) we propose COVID Smart Data based Network (COVID-SDNet) methodology for improving the generalization capacity of COVID-classification models. Our approach reaches good and stable results with an accuracy of 97.72% +/- 0.95%, 86.90% +/- 3.20%, 61.80% +/- 5.49% in severe, moderate and mild COVID-19 severity levels. Our approach could help in the early detection of COVID-19. COVIDGR-1.0 along with the severity level labels are available to the scientific community through this link https://dasci.es/es/transferencia/open-data/covidgr/.