Deep Learning-Based Recognizing COVID-19 and other Common Infectious Diseases of the Lung by Chest CT Scan Images

Min Fu; Shuang-Lian Yi; Yuanfeng Zeng; Feng Ye; Yuxuan Li; Xuan Dong; Yan-Dan Ren; Linkai Luo; Jin-Shui Pan; Qi Zhang

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi. org/10.1101 org/10. /2020 Deep Learning-Based Recognizing COVID-19 7 may have been degraded by contaminated RNAase, which at least partly explains the 107 low positive rate of nucleic acid assay from nasal and pharyngeal swab. Another 108 constraint in practice is that the supply of assay kits of nucleic acid detection may be 109 seriously inadequate in case of a large-scale outbreak of disease. In contrast, 110 COVID-19 has relatively unique imaging features in CT manifestations. In early stage 111 (less than 1 week after symptom onset), the predominant pattern was unilateral or 112 bilateral ground-glass opacities. Within 1-3 weeks, ground-glass opacities will 113 progress to or co-existed with consolidations. (3) According to the investigation by 114 Guan et al, (4) at the time of admission, 86.2% revealed abnormal CT scans whereas 115 radiographic or CT abnormality was found in 97.1% of the patients with severe type 116 of COVID-19. What cause the characteristic abnormality found by CT scans? 117