Comparison of Deep-Learning and Conventional Machine-Learning Methods for the Automatic Recognition of the Hepatocellular Carcinoma Areas from Ultrasound Images.

The emergence of deep-learning methods in different computer vision tasks has proved to offer increased detection, recognition or segmentation accuracy when large annotated image datasets are available. In the case of medical image processing and computer-aided diagnosis within ultrasound images, where the amount of available annotated data is smaller, a natural question arises: are deep-learning methods better than conventional machine-learning methods? How do the conventional machine-learning methods behave in comparison with deep-learning methods on the same dataset? Based on the study of various deep-learning architectures, a lightweight multi-resolution Convolutional Neural Network (CNN) architecture is proposed. It is suitable for differentiating, within ultrasound images, between the Hepatocellular Carcinoma (HCC), respectively the cirrhotic parenchyma (PAR) on which HCC had evolved. The proposed deep-learning model is compared with other CNN architectures that have been adapted by transfer learning for the ultrasound binary classification task, but also with conventional machine-learning (ML) solutions trained on textural features. The achieved results show that the deep-learning approach overcomes classical machine-learning solutions, by providing a higher classification performance.

Can the Computed Tomographic Aspect of Porto-Systemic Circulation in Cirrhotic Patients be Associated with the Presence of Variceal Hemorrhage?

Background and objectives: Variceal bleeding is a serious complication caused by portal hypertension, frequently encountered among cirrhotic patients. The purpose of this study was to determine whether the aspect of the collateral, porto-systemic circulation, as detected by CT are associated with the presence variceal hemorrhage (VH). Materials and Methods: 81 cirrhotic patients who underwent a contrast-enhanced CT examination were retrospectively included in the study. Patients were divided into two groups: Cirrhotic patients with variceal hemorrhage during the hospital admission concomitant, with the CT examination (n = 33) and group 2-cirrhotic patients, without any variceal hemorrhage in their medical history (n = 48). The diameter of the left gastric vein, the presence or absence and dimensions of oesophageal and gastric varices, paraumbilical veins and splenorenal shunts were the indicators assessed on CT. Results: The univariate analysis showed a significant association between the presence of upper GI bleeding and the diameters of paraoesophageal veins, paragastric veins and left gastric vein respectively, all of these CT parameters being higher in patients with variceal bleeding. In the multivariate logistic regression analysis, only the diameter of the left gastric vein was independently associated with the presence of variceal hemorrhage (OR = 1.6 (95% CI: 1.17-2.19), p = 0.003). We found an optimal cut-off value of 3 mm for the diameter of the left gastric vein useful to discriminate among patients with variceal hemorrhage from the ones without it, with a good diagnostic performance (AUC = 0.78, Se = 97%, Sp = 45.8%, PPV = 55.2%, NPV = 95.7%).Conclusions: Our observations point out that an objective CT quantification of porto-systemic circulation can be correlated with the presence of variceal hemorrhage and the diameter of the left gastric vein can be a reliable parameter associated with this condition.

Point-of-care ultrasound in management of gallstone ileus - a case report.

Point-of-care ultrasound has become of utmost importance in the management of critically ill patients, changing everyday clinical practice. With this case report, we highlight the role of abdominal ultrasound in the management of elderly patientswith gallstone ileus (a rare cause of intestinal obstructions caused by a cholecysto-duodenal fistula) and with contraindications for contrast-enhanced CT scan.