Helmut Prosch

The updated version of the EFSUMB guidelines on the application of non-hepatic contrast-enhanced ultrasound (CEUS) deals with the use of microbubble ultrasound contrast outside the liver in the many established and emerging applications.

BACKGROUND: Automated segmentation of anatomical structures is a crucial step in image analysis. For lung segmentation in computed tomography, a variety of approaches exists, involving sophisticated pipelines trained and validated on different datasets. However, the clinical applicability of these approaches across diseases remains limited. METHODS: We compared four generic deep learning approaches trained on various datasets and two readily available lung segmentation algorithms. We performed evaluation on routine imaging data with more than six different disease patterns and three published data sets. RESULTS: Using different deep learning approaches, mean Dice similarity coefficients (DSCs) on test datasets varied not over 0.02. When trained on a diverse routine dataset (n = 36), a standard approach (U-net) yields a higher DSC (0.97 ± 0.05) compared to training on public datasets such as the Lung Tissue Research Consortium (0.94 ± 0.13, p = 0.024) or Anatomy 3 (0.92 ± 0.15, p = 0.001). Trained on routine data (n = 231) covering multiple diseases, U-net compared to reference methods yields a DSC of 0.98 ± 0.03 versus 0.94 ± 0.12 (p = 0.024). CONCLUSIONS: The accuracy and reliability of lung segmentation algorithms on demanding cases primarily relies on the diversity of the training data, highlighting the importance of data diversity compared to model choice. Efforts in developing new datasets and providing trained models to the public are critical. By releasing the trained model under General Public License 3.0, we aim to foster research on lung diseases by providing a readily available tool for segmentation of pathological lungs.

BACKGROUND: After the 2002/2003 severe acute respiratory syndrome outbreak, 30% of survivors exhibited persisting structural pulmonary abnormalities. The long-term pulmonary sequelae of coronavirus disease 2019 (COVID-19) are yet unknown, and comprehensive clinical follow-up data are lacking. METHODS: In this prospective, multicentre, observational study, we systematically evaluated the cardiopulmonary damage in subjects recovering from COVID-19 at 60 and 100 days after confirmed diagnosis. We conducted a detailed questionnaire, clinical examination, laboratory testing, lung function analysis, echocardiography and thoracic low-dose computed tomography (CT). RESULTS: Data from 145 COVID-19 patients were evaluated, and 41% of all subjects exhibited persistent symptoms 100 days after COVID-19 onset, with dyspnoea being most frequent (36%). Accordingly, patients still displayed an impaired lung function, with a reduced diffusing capacity in 21% of the cohort being the most prominent finding. Cardiac impairment, including a reduced left ventricular function or signs of pulmonary hypertension, was only present in a minority of subjects. CT scans unveiled persisting lung pathologies in 63% of patients, mainly consisting of bilateral ground-glass opacities and/or reticulation in the lower lung lobes, without radiological signs of pulmonary fibrosis. Sequential follow-up evaluations at 60 and 100 days after COVID-19 onset demonstrated a vast improvement of symptoms and CT abnormalities over time. CONCLUSION: A relevant percentage of post-COVID-19 patients presented with persisting symptoms and lung function impairment along with radiological pulmonary abnormalities >100 days after the diagnosis of COVID-19. However, our results indicate a significant improvement in symptoms and cardiopulmonary status over time.