Pericardial Effusion Predicts Clinical Outcomes in Patients with COVID-19: A Nationwide Multicenter Study.

RATIONALE AND OBJECTIVES: The prognostic role of pericardial effusion (PE) in Covid 19 is unclear. The aim of the present study was to estimate the prognostic role of PE in patients with Covid 19 in a large multicentre setting. MATERIALS AND METHODS: This retrospective study is a part of the German multicenter project RACOON (Radiological Cooperative Network of the Covid 19 pandemic). The acquired sample comprises 1197 patients, 363 (30.3%) women and 834 (69.7%) men. In every case, chest computed tomography was analyzed for PE. Data about 30-day mortality, need for mechanical ventilation and need for intensive care unit (ICU) admission were collected. Data were evaluated by means of descriptive statistics. Group differences were calculated with Mann-Whitney test and Fisher exact test. Uni-and multivariable regression analyses were performed. RESULTS: Overall, 46.4% of the patients were admitted to ICU, mechanical lung ventilation was performed in 26.6% and 30-day mortality was 24%. PE was identified in 159 patients (13.3%). The presence of PE was associated with 30-day mortality: HR= 1.54, CI 95% (1.05; 2.23), p = 0.02 (univariable analysis), and HR= 1.60, CI 95% (1.03; 2.48), p = 0.03 (multivariable analysis). Furthermore, density of PE was associated with the need for intubation (OR=1.02, CI 95% (1.003; 1.05), p = 0.03) and the need for ICU admission (OR=1.03, CI 95% (1.005; 1.05), p = 0.01) in univariable regression analysis. The presence of PE was associated with 30-day mortality in male patients, HR= 1.56, CI 95%(1.01-2.43), p = 0.04 (multivariable analysis). In female patients, none of PE values predicted clinical outcomes. CONCLUSION: The prevalence of PE in Covid 19 is 13.3%. PE is an independent predictor of 30-day mortality in male patients with Covid 19. In female patients, PE plays no predictive role.

Metadata-independent classification of MRI sequences using convolutional neural networks: Successful application to prostate MRI.

PURPOSE: The ever-increasing volume of medical imaging data and interest in Big Data research brings challenges to data organization, categorization, and retrieval. Although the radiological value chain is almost entirely digital, data structuring has been widely performed pragmatically, but with insufficient naming and metadata standards for the stringent needs of image analysis. To enable automated data management independent of naming and metadata, this study focused on developing a convolutional neural network (CNN) that classifies medical images based solely on voxel data. METHOD: A 3D CNN (3D-ResNet18) was trained using a dataset of 31,602 prostate MRI volumes with 10 different sequence types of 1243 patients. A five-fold cross-validation approach with patient-based splits was chosen for training and testing. Training was repeated with a gradual reduction in training data assessing classification accuracies to determine the minimum training data required for sufficient performance. The trained model and developed method were tested on three external datasets. RESULTS: The model achieved an overall accuracy of 99.88 % ± 0.13 % in classifying typical prostate MRI sequence types. When being trained with approximately 10 % of the original cohort (112 patients), the CNN still achieved an accuracy of 97.43 % ± 2.10 %. In external testing the model achieved sensitivities of > 90 % for 10/15 tested sequence types. CONCLUSIONS: The herein developed CNN enabled automatic and reliable sequence identification in prostate MRI. Ultimately, such CNN models for voxel-based sequence identification could substantially enhance the management of medical imaging data, improve workflow efficiency and data quality, and allow for robust clinical AI workflows.