Analysis of COVID-19 chest ct in asymptomatic infection transferred to diagnosis patients

Objective: To analyze the clinical classification and chest CT manifestations of coronavirus disease 2019(COVID-19)in asymptomatic infection transferred to diagnosis, and improve image understanding of asymptomatic infection COVID-19.

A Systematic Analysis on COVID-19 Patients in Inner Mongolia Based on Dynamic Monitoring.

COVID-19 has spread globally with over 90,000,000 incidences and 1,930,000 deaths by Jan 11, 2021, which poses a big threat to public health. It is urgent to distinguish COVID-19 from common pneumonia. In this study, we reported multiple clinical feature analyses on COVID-19 in Inner Mongolia for the first time. We dynamically monitored multiple clinical features of all 75 confirmed COVID-19 patients, 219 pneumonia patients, and 68 matched healthy people in Inner Mongolia. Then, we studied the association between COVID-19 and clinical characteristics, based on which to construct a novel logistic regression model for predicting COVID-19. As a result, among the tested clinical characteristics, WBC, hemoglobin, C-reactive protein (CRP), ALT, and Cr were significantly different between COVID-19 patients and patients in other groups. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was 0.869 for the logistic regression model using multiple factors associated with COVID-19. Furthermore, the CRP reaction showed five different time-series patterns with one-peak and double-peak modes. In conclusion, our study identified a few clinical characteristics significantly different between COVID-19 patients and others in Inner Mongolia. The features can be used to establish a reliable logistic regression model for predicting COVID-19.

A deep learning integrated radiomics model for identification of coronavirus disease 2019 using computed tomography.

Since its first outbreak, Coronavirus Disease 2019 (COVID-19) has been rapidly spreading worldwide and caused a global pandemic. Rapid and early detection is essential to contain COVID-19. Here, we first developed a deep learning (DL) integrated radiomics model for end-to-end identification of COVID-19 using CT scans and then validated its clinical feasibility. We retrospectively collected CT images of 386 patients (129 with COVID-19 and 257 with other community-acquired pneumonia) from three medical centers to train and externally validate the developed models. A pre-trained DL algorithm was utilized to automatically segment infected lesions (ROIs) on CT images which were used for feature extraction. Five feature selection methods and four machine learning algorithms were utilized to develop radiomics models. Trained with features selected by L1 regularized logistic regression, classifier multi-layer perceptron (MLP) demonstrated the optimal performance with AUC of 0.922 (95% CI 0.856-0.988) and 0.959 (95% CI 0.910-1.000), the same sensitivity of 0.879, and specificity of 0.900 and 0.887 on internal and external testing datasets, which was equivalent to the senior radiologist in a reader study. Additionally, diagnostic time of DL-MLP was more efficient than radiologists (38 s vs 5.15 min). With an adequate performance for identifying COVID-19, DL-MLP may help in screening of suspected cases.