Assisting Scalable Diagnosis Automatically via CT Images in the Combat against COVID-19

  1. Bohan Liu
  2. Pan Liu
  3. Lutao Dai
  4. Yanlin Yang
  5. Peng Xie
  6. Yiqing Tan
  7. Jicheng Du
  8. Wei Shan
  9. Chenghui Zhao
  10. Qin Zhong
  11. Xixiang Lin
  12. Xizhou Guan
  13. Ning Xing
  14. Yuhui Sun
  15. Wenjun Wang
  16. Zhibing Zhang
  17. Xia Fu
  18. Yanqing Fan
  19. Meifang Li
  20. Na Zhang
  21. Lin Li
  22. Yaou Liu
  23. Lin Xu
  24. Jingbo Du
  25. Zhenhua Zhao
  26. Xuelong Hu
  27. Weipeng Fan
  28. Rongpin Wang
  29. Chongchong Wu
  30. Yongkang Nie
  31. Liuquan Cheng
  32. Lin Ma
  33. Zongren Li
  34. Qian Jia
  35. Minchao Liu
  36. Huayuan Guo
  37. Gao Huang
  38. Haipeng Shen
  39. Weimin An
  40. Hao Li
  41. Jianxin Zhou
  42. Kunlun He
1 evaluations Published on
Read the full article Related papers
This article on Sciety

Abstract

The pandemic of coronavirus Disease 2019 (COVID-19) caused enormous loss of life globally. 1-3 Case identification is critical. The reference method is using real-time reverse transcription PCR (rRT-PCR) assays, with limitations that may curb its prompt large-scale application. COVID-19 manifests with chest computed tomography (CT) abnormalities, some even before the onset of symptoms. We tested the hypothesis that application of deep learning (DL) to the 3D CT images could help identify COVID-19 infections. Using the data from 920 COVID-19 and 1,073 non-COVID-19 pneumonia patients, we developed a modified DenseNet-264 model, COVIDNet, to classify CT images to either class. When tested on an independent set of 233 COVID-19 and 289 non-COVID-19 patients. COVIDNet achieved an accuracy rate of 94.3% and an area under the curve (AUC) of 0.98. Application of DL to CT images may improve both the efficiency and capacity of case detection and long-term surveillance.

Related articles

Related articles are currently not available for this article.