Predicting COVID-19 malignant progression with AI techniques

Xiang Bai; Cong Fang; Yu Zhou; Song Bai; Zaiyi Liu; Qianlan Chen; Yongchao Xu; Tian Xia; Shi Gong; Xudong Xie; Dejia Song; Ronghui Du; Chunhua Zhou; Chengyang Chen; Dianer Nie; Dandan Tu; Changzheng Zhang; Xiaowu Liu; Lixin Qin; Weiwei Chen

The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.20.20037325 doi: medRxiv preprint parameters were selected by logistic regression analysis in this study. Elder age and comorbidity with hypertension, reflecting the basic physical condition of susceptible people, are the most important risk factors for malignant progression 19 . Pathological studies about COVID-19 reported extensive cellular fibromyxoid exudates and interstitial lymphocytes inflammatory infiltrates 20, 21 . Thus, the decrease of albumin and lymphocytes, the following two risk factors, probably suggest they migrate from the circulation into affected alveoli and also resulting in an increase of consolidation on follow-up CT imaging. Elevated HCRP is an indicator of anti-virus inflammation, also served as a risk factor for malignant progression. The risk factors mentioned above are all in accordance with the results from observation studies [22] [23] [24] [25] [26] . Fibrosis on CT imaging is reported to be closely associated with the late outcome of COVID-19 pneumonia 27,28 . However, our study depicted the early presence of fibrosis on CT imaging is a protective factor for malignant progression, which might reflect early absorption of inflammation and the occurrence of the repair stage.