ABSTRACT
The pandemic of Coronavirus Disease 2019 (COVID-19) is causing enormous loss of life globally. Prompt case identification is critical. The reference method is the real-time reverse transcription PCR (RT-PCR) assay, whose limitations 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 the application of deep learning (DL) to 3D CT images could help identify COVID-19 infections. Using 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 pneumonia patients, COVIDNet achieved an accuracy rate of 94.3% and an area under the curve of 0.98. As of March 23, 2020, the COVIDNet system had been used 11,966 times with a sensitivity of 91.12% and a specificity of 88.50% in six hospitals with PCR confirmation. Application of DL to CT images may improve both efficiency and capacity of case detection and long-term surveillance.
Subject(s)
COVID-19/diagnostic imaging , COVID-19/diagnosis , Tomography, X-Ray Computed/methods , COVID-19/epidemiology , COVID-19/metabolism , China/epidemiology , Data Accuracy , Deep Learning , Humans , Lung/pathology , Pneumonia/diagnostic imaging , Retrospective Studies , SARS-CoV-2/isolation & purification , Sensitivity and SpecificityABSTRACT
It is widely recognized that hypertension is one of the major risk factor for disease severity and mortality in patients with coronavirus disease 2019 (COVID-19). However, type 2 diabetes mellitus (T2DM) and hypertension are frequent comorbid conditions, complicating the assessment of hypertension's individual contribution to the risk. The aims of this study were to evaluate the contributions of hypertension alone, T2DM alone, or their combination to the risk of death, acute respiratory distress syndrome (ARDS)/respiratory failure, and severe COVID-19 infection. Additionally, we assessed risks associated with elevated blood pressure and fasting blood glucose on the same three clinical outcomes. Multivariate logistic models were used for these analyses. Among the 3400 patients, 3327(97.9%) survived and 73(2.1%) died. Compared to patients having neither hypertension nor T2DM (n = 1392), the risk of mortality was significantly higher in patients with T2DM alone (n = 226, OR 5.26 [95% CI: 2.39-11.58]) or with T2DM in combination with hypertension (n = 507, OR 3.02, [95% CI: 1.48-6.15]). Similarly, T2DM was a risk factor for development of ARDS/respiratory failure and severe infection. Hypertension alone (n = 1275) only conferred additional risk for the development of severe infection (OR 1.22 [95% CI: 1.00-1.51]). In conclusion, neither hypertension nor elevated blood pressure was independent risk factors for death or ARDS/respiratory failure but hypertension marginally increased the risk of severe COVID-19 infection. The risk associated with hypertension is accentuated through its confounding effect on T2DM.