Myocardial extracellular volume assessment at CT in hospitalized COVID-19 patients with regards to pulmonary embolism.

PURPOSE: To evaluate myocardial status through the assessment of extracellular volume (ECV) calculated at computed tomography (CT) in patients hospitalized for novel coronavirus disease (COVID-19), with regards to the presence of pulmonary embolism (PE) as a risk factor for cardiac dysfunction. METHOD: Hospitalized patients with COVID-19 who underwent contrast-enhanced CT at our institution were retrospectively included in this study and grouped with regards to the presence of PE. Unenhanced and portal venous phase scans were used to calculate ECV by placing regions of interest in the myocardial septum and left ventricular blood pool. ECV values were compared between patients with and without PE, and correlations between ECV values and clinical or technical variables were subsequently appraised. RESULTS: Ninety-four patients were included, 63/94 of whom males (67%), with a median age of 70 (IQR 56-76 years); 28/94 (30%) patients presented with PE. Patients with PE had a higher myocardial ECV than those without (33.5%, IQR 29.4-37.5% versus 29.8%, IQR 25.1-34.0%; p = 0.010). There were no correlations between ECV and patients' age (p = 0.870) or sex (p = 0.122), unenhanced scan voltage (p = 0.822), portal phase scan voltage (p = 0.631), overall radiation dose (p = 0.569), portal phase scan timing (p = 0.460), and contrast agent dose (p = 0.563). CONCLUSIONS: CT-derived ECV could help identify COVID-19 patients at higher risk of cardiac dysfunction, especially when related to PE, to potentially plan a dedicated, patient-tailored clinical approach.

The impact of COVID-19 pandemic on radiology residents in Northern Italy.

OBJECTIVES: To assess changes in working patterns and education experienced by radiology residents in Northwest Italy during the COVID-19 pandemic. METHODS: An online questionnaire was sent to residents of 9 postgraduate schools in Lombardy and Piedmont, investigating demographics, changes in radiological workload, involvement in COVID-19-related activities, research, distance learning, COVID-19 contacts and infection, changes in training profile, and impact on psychological wellbeing. Descriptive and χ2 statistics were used. RESULTS: Among 373 residents invited, 300 (80%) participated. Between March and April 2020, 44% (133/300) of respondents dedicated their full time to radiology; 41% (124/300) engaged in COVID-19-related activities, 73% (90/124) of whom working in COVID-19 wards; 40% (121/300) dedicated > 25% of time to distance learning; and 66% (199/300) were more involved in research activities than before the pandemic. Over half of residents (57%, 171/300) had contacts with COVID-19-positive subjects, 5% (14/300) were infected, and 8% (23/300) lost a loved one due to COVID-19. Only 1% (3/300) of residents stated that, given the implications of this pandemic scenario, they would not have chosen radiology as their specialty, whereas 7% (22/300) would change their subspecialty. The most common concerns were spreading the infection to their loved ones (30%, 91/300), and becoming sick (7%, 21/300). Positive changes were also noted, such as being more willing to cooperate with other colleagues (36%, 109/300). CONCLUSIONS: The COVID-19 pandemic changed radiology residents' training programmes, with distance learning, engaging in COVID-19-related activities, and a greater involvement in research becoming part of their everyday practice. KEY POINTS: • Of 300 participants, 44% were fully dedicated to radiological activity and 41% devoted time to COVID-19-related activities, 73% of whom to COVID-19 wards. • Distance learning was substantial for 40% of residents, and 66% were involved in research activities more than before the COVID-19 pandemic. • Over half of residents were exposed to COVID-19 contacts and less than one in twenty was infected.

Artificial Intelligence Applied to Chest X-ray for Differential Diagnosis of COVID-19 Pneumonia.

We assessed the role of artificial intelligence applied to chest X-rays (CXRs) in supporting the diagnosis of COVID-19. We trained and cross-validated a model with an ensemble of 10 convolutional neural networks with CXRs of 98 COVID-19 patients, 88 community-acquired pneumonia (CAP) patients, and 98 subjects without either COVID-19 or CAP, collected in two Italian hospitals. The system was tested on two independent cohorts, namely, 148 patients (COVID-19, CAP, or negative) collected by one of the two hospitals (independent testing I) and 820 COVID-19 patients collected by a multicenter study (independent testing II). On the training and cross-validation dataset, sensitivity, specificity, and area under the curve (AUC) were 0.91, 0.87, and 0.93 for COVID-19 versus negative subjects, 0.85, 0.82, and 0.94 for COVID-19 versus CAP. On the independent testing I, sensitivity, specificity, and AUC were 0.98, 0.88, and 0.98 for COVID-19 versus negative subjects, 0.97, 0.96, and 0.98 for COVID-19 versus CAP. On the independent testing II, the system correctly diagnosed 652 COVID-19 patients versus negative subjects (0.80 sensitivity) and correctly differentiated 674 COVID-19 versus CAP patients (0.82 sensitivity). This system appears promising for the diagnosis and differential diagnosis of COVID-19, showing its potential as a second opinion tool in conditions of the variable prevalence of different types of infectious pneumonia.