Cannonball appearance, a tricky yet underrated imaging manifestation of COVID-19.

Spreading swiftly across the borders and over the seas, severe acute respiratory syndrome-related coronavirus-2 (SARS-COV-2), as causative pathogen of coronavirus disease 2019 (COVID-19), is currently the main global public health concern. "Cannonball appearance," as a rare and yet underrated CT feature of COVID-19 pneumonia, has been typically linked to certain hematogenous pulmonary metastases and some inflammatory/infection conditions, including tuberculosis, but no other viral or atypical pneumonia. Cannonball appearance can bring diagnostic dilemmas and difficulties in monitoring treatment response in patients with or suspicious for hematogenous pulmonary metastasis. Hereby, we report two cases of COVID-19 delta variant-induced pneumonia manifesting unusually in chest CT scan with cannonball appearance.

CT Features and Short-term Prognosis of COVID-19 Pneumonia: A Single-Center Study from Kashan, Iran.

PURPOSE: To assess whether certain CT chest features of patients with confirmed coronavirus disease 2019 (COVID-19) may have short-term prognostic value. MATERIALS AND METHODS: One hundred-twenty consecutive symptomatic patients with COVID-19 infection who had undergone chest CT were enrolled in this retrospective study. Patients were categorized into three groups: routine inward hospitalization, intensive care unit admission, and deceased based on a short-term follow-up. Detailed initial CT features and distributional evaluation were recorded. RESULTS: The mean age in the deceased group was 70.7 years, significantly higher than the other two groups (P < .05). Ninety-four percent (113/120) of the patients had ground-glass opacities (GGO). Peripheral and lower zone predilection was present in most patients. Subpleural sparing and pleural effusion were seen in approximately 23% (28/120) and 17% (20/120) of the patients, respectively. The combined intensive care unit group and deceased patients had significantly more consolidation, air bronchograms, crazy paving, and central involvement of the lungs compared with routinely hospitalized patients (all P < .05). CONCLUSION: This study supports the previously described typical CT appearance of COVID-19 pneumonia with bilateral GGO, in peripheral distribution and lower lung zone predilection. Subpleural sparing and pleural effusion were seen approximately in one-fifth and one-sixth of the patients with COVID-19, respectively. Consolidation, air bronchograms, central lung involvement, crazy paving and pleural effusion on initial CT chest have potential prognostic values, the features more commonly observed in critically ill patients.© RSNA, 2020.

A multi-center study of COVID-19 patient prognosis using deep learning-based CT image analysis and electronic health records.

PURPOSE: As of August 30th, there were in total 25.1 million confirmed cases and 845 thousand deaths caused by coronavirus disease of 2019 (COVID-19) worldwide. With overwhelming demands on medical resources, patient stratification based on their risks is essential. In this multi-center study, we built prognosis models to predict severity outcomes, combining patients' electronic health records (EHR), which included vital signs and laboratory data, with deep learning- and CT-based severity prediction. METHOD: We first developed a CT segmentation network using datasets from multiple institutions worldwide. Two biomarkers were extracted from the CT images: total opacity ratio (TOR) and consolidation ratio (CR). After obtaining TOR and CR, further prognosis analysis was conducted on datasets from INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3. For each data cohort, generalized linear model (GLM) was applied for prognosis prediction. RESULTS: For the deep learning model, the correlation coefficient of the network prediction and manual segmentation was 0.755, 0.919, and 0.824 for the three cohorts, respectively. The AUC (95 % CI) of the final prognosis models was 0.85(0.77,0.92), 0.93(0.87,0.98), and 0.86(0.75,0.94) for INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3 cohorts, respectively. Either TOR or CR exist in all three final prognosis models. Age, white blood cell (WBC), and platelet (PLT) were chosen predictors in two cohorts. Oxygen saturation (SpO2) was a chosen predictor in one cohort. CONCLUSION: The developed deep learning method can segment lung infection regions. Prognosis results indicated that age, SpO2, CT biomarkers, PLT, and WBC were the most important prognostic predictors of COVID-19 in our prognosis model.

Severity and Consolidation Quantification of COVID-19 From CT Images Using Deep Learning Based on Hybrid Weak Labels.

Early and accurate diagnosis of Coronavirus disease (COVID-19) is essential for patient isolation and contact tracing so that the spread of infection can be limited. Computed tomography (CT) can provide important information in COVID-19, especially for patients with moderate to severe disease as well as those with worsening cardiopulmonary status. As an automatic tool, deep learning methods can be utilized to perform semantic segmentation of affected lung regions, which is important to establish disease severity and prognosis prediction. Both the extent and type of pulmonary opacities help assess disease severity. However, manually pixel-level multi-class labelling is time-consuming, subjective, and non-quantitative. In this article, we proposed a hybrid weak label-based deep learning method that utilize both the manually annotated pulmonary opacities from COVID-19 pneumonia and the patient-level disease-type information available from the clinical report. A UNet was firstly trained with semantic labels to segment the total infected region. It was used to initialize another UNet, which was trained to segment the consolidations with patient-level information using the Expectation-Maximization (EM) algorithm. To demonstrate the performance of the proposed method, multi-institutional CT datasets from Iran, Italy, South Korea, and the United States were utilized. Results show that our proposed method can predict the infected regions as well as the consolidation regions with good correlation to human annotation.

Chest CT in COVID-19 pneumonia: what are the findings in mid-term follow-up?

PURPOSE: The purposes of this study are to investigate mid-term chest computed tomography (CT) findings of coronavirus disease 2019 (COVID-19) pneumonia, assess the rate of complete resolution, and determine the individuals at risk for residual abnormalities. METHODS: Fifty-two cases of COVID-19 pneumonia with at least two chest CTs and mean 3-month interval between the initial and follow-up CT were enrolled in this retrospective study. Patients were categorized into two groups: complete resolution and residual disease on follow-up CT. Demographic, clinical, laboratory, and therapeutic data as well as initial and follow-up chest CT scans were compared and analyzed. RESULTS: Thirty patients (57.7%) demonstrate complete resolution of pulmonary findings, and 22 patients (42.3%) had residual disease on follow-up CT. The mean time interval between initial and follow-up CT was 91.3 ± 17.2 and 90.6 ± 14.3 days in the complete resolution and residual groups, respectively. The most common radiologic pattern in residual disease was ground-glass opacities (54.5%), followed by mixed ground-glass and subpleural parenchymal bands (31.8%), and pure parenchymal bands (13.7%). Compared to complete resolution group, patients with residual disease had higher CT severity score on initial exam (10.3 ± 5.4 vs. 7.3 ± 4.6, P value = 0.036), longer duration of hospitalization, higher rate of intensive care unit (ICU) admission, more underlying medical conditions, higher initial WBC count, and higher occurrence rate of leukocytosis in the hospitalization time period (all P values < 0.05). CONCLUSION: Extensive lung involvement on initial CT, ICU admission, long duration of hospitalization, presence of underlying medical conditions, high initial WBC count, and development of leukocytosis during the course of disease are associated with more prevalence of chronic lung sequela of COVID-19.

A low-dose chest CT protocol for the diagnosis of COVID-19 pneumonia: a prospective study.

PURPOSE: The increasing trend of chest CT utilization during the COVID-19 pandemic necessitates novel protocols with reduced dose and maintained diagnostic accuracy. We aimed to investigate the diagnostic accuracy of 30-mAs chest CT protocol in comparison with a 150-mAs standard-dose routine protocol for imaging of COVID-19 pneumonia. METHODS: Upon IRB approval, consecutive laboratory-confirmed positive COVID-19 patients aged 50 years or older who were referred for chest CT scan and had same-day normal CXR were invited to participate in this prospective study. First, a standard-dose chest CT scan (150 mAs) was performed. Only if typical COVID-19 pneumonia features were identified, then a low-dose CT (30 mAs) was done immediately. Diagnostic accuracy of low-dose and standard-dose CT in the detection of typical COVID-19 pneumonia features were compared. RESULTS: Twenty patients with a mean age of 64.20 ± 13.8 were enrolled in the study. There was excellent intrareader agreement in detecting typical findings of COVID-19 pneumonia between low-dose and standard-dose (intraclass correlation coefficient [ICC] = 0.98-0.99, P values < 0.001 all readers). The mean effective dose values in standard- and low-dose groups were 6.60 ± 1.47 and 1.80 ± 0.42 mSv, respectively. Also, absolute cancer risk per mean cumulative effective dose values obtained from the standard- and low-dose CT examinations were 2.71 × 10-4 and 0.74 × 10-4, respectively. CONCLUSIONS: According to our study, it was found that proposed low-dose CT chest protocol is reliable in detecting COVID-19 pneumonia in daily practice with significant reduction in radiation dose and estimated cancer risk.