Lung Segmentation and Characterization in COVID-19 Patients for Assessing Pulmonary Thromboembolism: An Approach Based on Deep Learning and Radiomics

The COVID-19 pandemic is inevitably changing the world in a dramatic way, and the role of computed tomography (CT) scans can be pivotal for the prognosis of COVID-19 patients. Since the start of the pandemic, great care has been given to the relationship between interstitial pneumonia caused by the infection and the onset of thromboembolic phenomena. In this preliminary study, we collected n = 20 CT scans from the Polyclinic of Bari, all from patients positive with COVID-19, nine of which developed pulmonary thromboembolism (PTE). For eight CT scans, we obtained masks of the lesions caused by the infection, annotated by expert radiologists;whereas for the other four CT scans, we obtained masks of the lungs (including both healthy parenchyma and lesions). We developed a deep learning-based segmentation model that utilizes convolutional neural networks (CNNs) in order to accurately segment the lung and lesions. By considering the images from publicly available datasets, we also realized a training set composed of 32 CT scans and a validation set of 10 CT scans. The results obtained from the segmentation task are promising, allowing to reach a Dice coefficient higher than 97%, posing the basis for analysis concerning the assessment of PTE onset. We characterized the segmented region in order to individuate radiomic features that can be useful for the prognosis of PTE. Out of 919 extracted radiomic features, we found that 109 present different distributions according to the Mann–Whitney U test with corrected p-values less than 0.01. Lastly, nine uncorrelated features were retained that can be exploited to realize a prognostic signature.

Modulation of ACE-2 mRNA by inflammatory cytokines in human thyroid cells: a pilot study.

INTRODUCTION: Angiotensin-converting-enzyme-2 (ACE-2) was demonstrated to be the receptor for cellular entry of SARS-CoV-2. ACE-2 mRNA was identified in several human tissues and recently also in thyroid cells in vitro. PURPOSE: Aim of the present study was to investigate the effect of pro-inflammatory cytokines on the ACE-2 mRNA levels in human thyroid cells in primary cultures. METHODS: Primary thyroid cell cultures were treated with IFN-γ and TNF-α alone or in combination for 24 h. ACE-2 mRNA levels were measured by RT-PCR. As a control, the levels of IFN-γ inducible chemokine (CXCL10) were measured in the respective cell culture supernatants. RESULTS: The mean levels of ACE-2 mRNA increased after treatment with IFN-γ and TNF-α in all the thyroid cell preparations, while the combination treatment did not consistently synergically increase ACE-2-mRNA. At difference, CXCL10 was consistently increased by IFN-γ and synergically further increased by the combination treatment with IFN-γ + TNF-α, with respect to IFN-γ alone. CONCLUSIONS: The results of the present study show that IFN-γ and, to a lesser extent TNF-α consistently increase ACE-2 mRNA levels in NHT primary cultures. More interestingly, the combined stimulation (proven to be effective according to the synergic effect registered for CXCL10) produces different responses in terms of ACE-2 mRNA modulation. These results would suggest that elevated levels of pro-inflammatory cytokines could facilitate the entering of the virus in cells by further increasing ACE-2 expression and/or account for the different degree of severity of SARS-COV-2 infection. This hypothesis deserves to be confirmed by further specific studies.

Pulmonary Artery Filling Defects in COVID-19 Patients Revealed Using CT Pulmonary Angiography: A Predictable Complication?

PURPOSE: This study is aimed at assessing the prevalence of pulmonary artery filling defects (PAFDs) consistent with pulmonary artery embolism (PAE) in patients with SARS-CoV-2 infection and at investigating possible radiological or clinical predictors. MATERIALS AND METHODS: Computed Tomography Pulmonary Angiographies (CTPAs) from 43 consecutive patients with a confirmed COVID-19 infection were retrospectively reviewed, taking into consideration the revised Geneva score and the D-dimer value for each patient. Filling defects within the pulmonary arteries were recorded along with pleural and parenchymal findings such as ground glass opacities, consolidation, crazy paving, linear consolidation, and pleural effusion. All these variables were compared between patients with and without PAFD. The predictive performance of statistically different parameters was investigated using the receiver operating characteristics (ROC). RESULTS: Pulmonary embolism was diagnosed in 15/43 patients (35%), whereas CTPA and parenchymal changes related to pulmonary COVID-19 disease were evident in 39/43 patients (91%). The revised Geneva score and the mean D-dimer value obtained using two consecutive measurements were significantly higher in patients with PAFD. The ROC analysis demonstrated that a mean D-dimer value is the parameter with the higher predictivity (AUC 0.831) that is a cut-off value > 1800 µg/l which predicts the probability of PAFD with a sensitivity and specificity of 70% and 78%, respectively. CONCLUSIONS: This single centre retrospective report shows a high prevalence of pulmonary artery filling defects revealed using CTPA in COVID-19 patients and demonstrates that the mean value of multiple D-dimer measurements may represent a predicting factor of this complication.