Attention based multi-scale nested network for biomedical image segmentation.

Convolutional neural network-based methods have significantly enhanced the segmentation performance of biomedical images in recent years. Nevertheless, medical image segmentation presents a challenge marked by layout specificity, with limited variation between samples in medical datasets but significant variation within each individual sample. This aspect has been often overlooked by many models.

Brain tumor feature extraction and edge enhancement algorithm based on U-Net network.

Background: Statistics show that each year more than 100,000 patients pass away from brain tumors. Due to the diverse morphology, hazy boundaries, or unbalanced categories of medical data lesions, segmentation prediction of brain tumors has significant challenges.

Purpose: In this thesis, we highlight EAV-UNet, a system designed to accurately detect lesion regions.

A pair of dinuclear Re(I) enantiomers: synthesis, crystal structures, chiroptical and ferroelectric properties.

The reaction of enantiomeric bis-bidentate bridging ligands (+)/(-)-2,5-bis(4,5-pinene-2-pyridyl)pyrazine (L(S)/L(R)) with [Re(CO)5Cl] yielded a pair of dinuclear Re(I) enantiomers formulated as [Re2(L(S)/L(R))(CO)6Cl2]·4CH2Cl2 (R-1 and S-1, the isomers containing the respective L(R) and L(S) ligands). They were characterized by elemental analyses, IR spectra and X-ray crystallography. Circular dichroism spectra verified their chiroptical activities and enantiomeric nature.

Anion-controlled self-assembly of two NLO-active dinuclear and molecular square Cu(II) enantiomeric pairs: from antiferromagnetic to ferromagnetic coupling.

Two second-order nonlinear optically (NLO)-active dinuclear and square Cu(II) enantiomeric pairs were obtained via the self-assemblies of enantiopure linear bis-bidentate ligands with different copper(II) salts under the identical reaction conditions. Their magnetic properties are switched from antiferromagnetic to ferromagnetic coupling.