Comparative Study of Deep Transfer Learning Models for Semantic Segmentation of Human Mesenchymal Stem Cell Micrographs.

The aim of this study is to conduct a comparative assessment of the effectiveness of neural network models-U-Net, DeepLabV3+, SegNet and Mask R-CNN-for the semantic segmentation of micrographs of human mesenchymal stem cells (MSCs). A dataset of 320 cell micrographs annotated by cell biology experts was created. The models were trained using a transfer learning method based on ImageNet pre-trained weights.

A Comparative Study of Aortic Valve Neocuspidization Techniques: Formula vs. Template Methods of Neocusp Formation.

Introduction: The template method (TM), pioneered by Ozaki for aortic valve neocuspidization (AVNeo), has been widely adopted for aortic valve replacement, though it requires specialized instruments. This study introduces a novel formula method (FM), which uses the diameter of the aortic valve fibrous ring (AV-D) to determine the dimensions of the neocusps to be trimmed from autologous without the need for templates, potentially reducing costs and complexity. We aimed to compare the clinical outcomes of the FM with the established TM in patients undergoing AVNeo.

[Infective endocarditis in the elderly: The current view of the problem].

The review gives the current views of the etiology and risk factors of infective endocarditis (IE) in the elderly. It reports the specific clinical picture of the disease at this age, including clinically apparent syndromes, comorbidity, and various clinical signs and peculiarities of IE. Particular emphasis is placed on the diagnosis of an endocardial infectious lesion, which is based on the use of the modified Duke criteria.