Supervised Domain Adaptation for Automated Semantic Segmentation of the Atrial Cavity

Atrial fibrillation (AF) is the most common cardiac arrhythmia. At present, cardiac ablation is the main treatment procedure for AF. To guide and plan this procedure, it is essential for clinicians to obtain patient-specific 3D geometrical models of the atria. For this, there is an interest in automatic image segmentation algorithms, such as deep learning (DL) methods, as opposed to manual segmentation, an error-prone and time-consuming method. However, to optimize DL algorithms, many annotated examples are required, increasing acquisition costs. The aim of this work is to develop automatic and high-performance computational models for left and right atrium (LA and RA) segmentation from a few labelled MRI volumetric images with a 3D Dual U-Net algorithm. For this, a supervised domain adaptation (SDA) method is introduced to infer knowledge from late gadolinium enhanced (LGE) MRI volumetric training samples (80 LA annotated samples) to a network trained with balanced steady-state free precession (bSSFP) MR images of limited number of annotations (19 RA and LA annotated samples). The resulting knowledge-transferred model SDA outperformed the same network trained from scratch in both RA (Dice equals 0.9160) and LA (Dice equals 0.8813) segmentation tasks.     

Structure Theorems for Pure Semisimple Grothendieck Locally PI-Categories

Let Q be a nondegenerate quadratic form over the finite field F _q and O _n (F _q ) be the associated orthogonal group. Let O _n (F _q ) act linearly on the polynomial ring F _q [x ₁, …, x _n ]. We find the invariant subring of O ₄(F _q ) with explicit generators.     

Covering morphisms

Let (R, m) be a two-dimensional regular local ring and let A be a finitely generated torsion-free R-module. If A is a complete module, then Dan Katz and Vijay Kodiyalam show A satisfies five conditions. They ask whether these five conditions are equivalent without assuming A to be complete. In a previous paper we determined all implications among these five conditions with one exception. In this paper, with an additional hypothesis on the units of R, we resolve the remaining case.     

Eyringpy: A program for computing rate constants in the gas phase and in solution

Eyringpy is a modular program for calculating thermochemical properties and rate constants for reactions in the gas phase and in solution. The code is written in Python and it has a user-friendly interface and a simple input format. Unimolecular and bimolecular reactions with one and two products are supported. Thermochemical properties are estimated through canonical ensemble and rate constants are computed according to the transition state theory. One-dimensional Wigner and Eckart tunneling corrections are also available. Rate constants of bimolecular reactions involving the formation of pre-reactive complexes are also estimated. To compute rate constants in solution, Eyringpy uses the Collins–Kimball theory to include the diffusion-limit, the Marcus theory for electron transfer processes, and the molar fractions to account for the solvent pH effect.     

Guest Exchange Mechanisms in Mono‐Metallic PdII/PtII‐Cages Based on a Tetra‐Pyridyl Calix[4]pyrrole Ligand

Planar pyridyl N‐oxides are encapsulated in mono‐metallic Pd^II/Pt^II‐cages based on a tetra‐pyridyl calix[4]pyrrole ligand. The exchange dynamics of the cage complexes are slow on both the NMR chemical shift and EXSY timescales, but encapsulation of the guests by the cages is fast on the human timescale. A “French doors” mechanism, involving the rotation of the meso‐phenyl walls of the cages, allows the passage of the planar guests. The encapsulation of quinuclidine N‐oxide, a sterically more demanding guest, is slower than pyridyl N‐oxides in the Pd^II‐cage, and does not take place in the Pt^II counterpart. A modification of the encapsulation mechanism for the quinuclidine N‐oxide is postulated that requires the partial dissociation of the Pd^II‐cage. The substrate binding selectivity featured by the cages is related to their different guest uptake/release mechanisms.     

Zero‐valent Iron Nanoparticles Modified Screen‐printed Electrode for FIA or HPLC Amperometric Detection of Metronidazole in Pharmaceutical Formulations

A zero‐valent iron nanoparticles modified electrode was employed as an amperometric detector in flow conditions to determine metronidazole and 2‐methyl‐5‐nitroimidazole in pharmaceutical formulations. Flow injection analysis at ?0.6 V (vs. Ag) achieved limits of detection of 2 and 6 μM for metronidazole and 2‐methyl‐5‐nitroimidazole, respectively, but the analysis was not discriminative between the two compounds. When reverse phase high performance liquid chromatography was applied previous to the electrochemical detection both analytes could be analysed simultaneously. However, the limits of detection slightly increased (10 μM) as a consequence of the use of an organic solvent and a lower sample volume. The relative standard deviation of analytical repeatability was <4.0 % in both techniques. The methods were validated by comparing the results obtained from the analysis of commercial samples with those provided by HPLC‐DAD and no significant differences were detected. Results probed that the modified electrode was a successful tool in the FIA and HPLC analysis of nitro compounds.     

Study of nanocarbon black as synergist on improving flame retardancy of ethylene-vinyl acetate/brucite composites

Nanocarbon black (CB) was introduced into ethylene-vinyl acetate/brucite (EM) composites to investigate the synergistic effect of CB and metal hydroxide on improving the flame retardancy of EVA. Flammability properties of the as-prepared EVA composites were investigated by thermogravimetric analysis, limiting oxygen index (LOI), UL-94 test and cone calorimetry test. The results indicated that the optimum mass ratio of CB/brucite was 1/54, at which the EVA composites displayed dramatic improvement on thermal stability and flame retardancy. The LOI value was as high as 35.3%, the UL-94 passed the V-0 rating, and the peak heat release rate reduced 79% in comparison with pure EVA. Based on the morphology and structure analysis for residue chars, the flame-retardant mechanism was attributed mainly to the positive synergistic effect of CB and brucite on promoting the formation of better carbon protective layer during combustion.

     

Thermodynamic study of complexation of Zn(II)/L (L− = acetate,indomethacin and diclofenac anions) by isothermal titration calorimetry

Journal of Thermal Analysis and Calorimetry - In this work, a thermodynamic study of the Zn(II)/L systems (L??=?acetate, AcO? or indomethacin anion, Indo? or...     

Spontaneous trans‐Selective Transfer Hydrogenation of Apolar Boron–Boron Double Bonds

The transfer hydrogenation of N‐heterocyclic carbene (NHC)‐supported diborenes with dimethylamine borane proceeds with high selectivity for the trans‐1,2‐dihydrodiboranes. DFT calculations, supported by kinetic studies and deuteration experiments, suggest a stepwise proton‐first‐hydride‐second reaction mechanism via an intermediate μ‐hydrodiboronium dimethylaminoborate ion pair.     

Carbon Corrosion in Proton-Exchange Membrane Fuel Cells: Spectrometric Evidence for Pt-Catalysed Decarboxylation at Anode-Relevant Potentials

The carbon oxidation reaction (COR) is a critical issue in proton-exchange membrane fuel cells (PEMFCs), as carbon in various forms is the most used electrocatalyst support material. The COR is thermodynamically possible above the C/CO₂ standard potential, but its rate becomes significantly important only at high overpotential (e. g. PEMFC cathode potential). Herein, using on-line differential electrochemical mass spectrometry, we show that oxygen-containing carbon surface groups present on high-surface aera carbon, Vulcan XC72 or reinforced graphite are oxidized at PEMFC anode-relevant potential (E=0.1 V vs. the reversible hydrogen electrode, RHE), but not at E=0.4 V vs. RHE. We rationalized our findings by considering a Pt-catalysed decarboxylation mechanism in which Pt nanoparticles provide adsorbed hydrogen species to the oxygen-containing carbon surface groups, eventually leading to evolution of carbon dioxide and carbon monoxide. These results shed fundamental light on an unexpected degradation mechanism and facilitate the understanding of the long-term stability of PEMFC anode nanocatalysts.