To characterize CT-ndings of COVID-19 pneumonia and their value in diagnosis and outcome prediction. METHODS Chest CTs of 182 patients with a con rmed diagnosis of COVID-19 infection by RT-PCR were evaluated for the presence of CT-abnormalities and their frequency. Regarding the patient outcome each patient was categorized in 5 progressive stages and the duration of hospitalization was determined. Regression analysis was performed to nd which CT ndings are predictive for patient outcome and to assess prognostic factors for the hospitalization duration. RESULTS Multivariate statistical analysis con rmed a higher age (OR= 1.023, p= 0.025), a higher total visual severity score (OR= 1.038, p= 0.002) and the presence of crazy paving (OR= 2.160, p= 0.034) as predictive parameters for patient outcome. A higher total visual severity score (+ 0.134 days; p= 0.012) and the presence of pleural effusion (+ 13.985 days, p= 0.005) were predictive parameters for a longer hospitalization duration. CONCLUSIONS An increasing percentage of lung opacity as well as the presence of crazy paving and a higher age are associated with a worse patient outcome. The presence of a higher total visual severity score and pleural effusion are signi cant predictors for a longer hospitalization duration.

The Coronavirus disease 2019 (COVID-19) presents open questions in how we clinically diagnose and assess disease course. Recently, chest computed tomography (CT) has shown utility for COVID-19 diagnosis. In this study, we developed Deep COVID DeteCT (DCD), a deep learning convolutional neural network (CNN) that uses the entire chest CT volume to automatically predict COVID-19 (COVID+) from non-COVID-19 (COVID−) pneumonia and normal controls. We discuss training strategies and differences in performance across 13 international institutions and 8 countries. The inclusion of non-China sites in training significantly improved classification performance with area under the curve (AUCs) and accuracies above 0.8 on most test sites. Furthermore, using available follow-up scans, we investigate methods to track patient disease course and predict prognosis.

No abstract

Objective To determine the type and frequency of incidental findings detected on preoperative computed tomography (CT) imaging obtained for robotic-assisted joint replacements and their effect on the planned arthroplasty. Materials and methods All preoperative CT examinations performed for a robotic-assisted knee or total hip arthroplasty were obtained. This resulted in 1432 examinations performed between September 2016 and February 2020 at our institution. These examinations were initially interpreted by 1 of 9 fellowship-trained musculoskeletal radiologists. Using a diagnosis search, the examination reports were then reviewed to catalog all incidental findings and further classify as significant or non-significant findings. Demographic information was obtained. In those with significant findings, a chart review was performed to record the relevant workup, outcomes, and if the planned arthroplasty was affected. Results Incidental findings were diagnosed in 740 (51.7%) patients. Of those with incidental findings, 41 (5.5%) were considered significant. A significant finding was more likely to be detected in males (P = 0.007) and on the hip protocol CT (P = 0.014). In 8 patients, these diagnoses resulted in either delay or cancelation of the arthroplasty. A planned total hip arthroplasty was more likely to be altered as compared to a knee arthroplasty (P = 0.018). Conclusion Incidental findings are commonly detected by radiologists on preoperative CT imaging obtained for robotic-assisted joint replacement. Several were valuable findings and resulted in a delay or even cancelation of the planned arthroplasty after the detection of critical diagnoses, which if not identified may have resulted in devastating outcomes.