Objectives: To develop and test machine learning-based CT radiomics models for predicting hospital stay in patients with pneumonia associated with SARS-CoV-2 infection. Design: Cross-sectional Setting: Multicenter Participants: A total of 52 patients with laboratory-confirmed SARS-CoV-2 infection and their initial CT images were enrolled from 5 designated hospitals in Ankang, Lishui, Zhenjiang, Lanzhou, and Linxia between January 23, 2020 and February 8, 2020. As of February 20, patients remained in hospital or with non-findings in CT were excluded. Therefore, 31 patients with 72 lesion segments were included in the final analysis. Intervention: CT radiomics models based on logistic regression (LR) and random forest (RF) were developed on features extracted from pneumonia lesions in training and inter-validation datasets. The predictive performance was further evaluated in test dataset on lung lobe- and patients-level. Main outcomes: Short-term hospital stay (≤10 days) and long-term hospital stay (>10 days). Results: The CT radiomics models based on 6 second-order features were effective in discriminating short- and long-term hospital stay in patients with pneumonia associated with SARS-CoV-2 infection, with areas under the curves of 0.97 (95%CI 0.83-1.0) and 0.92 (95%CI 0.67-1.0) by LR and RF, respectively, in the test dataset. The LR model showed a sensitivity and specificity of 1.0 and 0.89, and the RF model showed similar performance with sensitivity and specificity of 0.75 and 1.0 in test dataset. Conclusions: The machine learning-based CT radiomics models showed feasibility and accuracy for predicting hospital stay in patients with pneumonia associated with SARS-CoV-2 infection.

Purpose To study the extent of pulmonary involvement in COVID-19 with quantitative CT (QCT) and to assess the impact of disease burden on opacity visibility on chest radiographs. Materials and Meth...

Background Lung ultrasound (LUS) and computed tomography (CT) can both be used for diagnosis of interstitial pneumonia caused by coronavirus disease 2019 (COVID-19), but the agreement between LUS and CT is unknown. Purpose to compare the agreement of LUS and CT in the diagnosis of interstitial pneumonia caused by COVID-19. Materials and Methods We searched PubMed, Cochrane library, Embase, Chinese Biomedicine Literature, and WHO COVID-19 databases to identify studies that compared LUS with CT in the diagnosis of interstitial pneumonia caused by COVID-19. We calculated the pooled overall, positive and negative percent agreements, diagnostic odds ratio (DOR) and the area under the standard receiver operating curve (SROC) for LUS in the diagnosis of COVID-19 compared with CT. Results We identified 1896 records, of which nine studies involving 531 patients were finally included. The pooled overall, positive and negative percentage agreements of LUS for the diagnosis of interstitial pneumonia caused by COVID-19 compared with CT were 81% (95% confidence interval [CI] 43–99%), 96% (95% CI, 80–99%, I² = 92.15%) and 80% (95%CI, 60–92%, I² = 92.85%), respectively. DOR was 37.41 (95% CI, 9.43–148.49, I² = 63.9%), and the area under the SROC curve was 0.94 (95% CI, 0.92–0.96). The quality of evidence for both specificity and sensitivity was low because of heterogeneity and risk of bias. Conclusion The level of diagnostic agreement between LUS and CT in the diagnosis of interstitial pneumonia caused by COVID-19 is high. LUS can be therefore considered as an equally accurate alternative for CT in situations where molecular tests are not available.