No abstract

Magnetic Resonance Imaging (MRI) offers highresolution in vivo imaging and rich functional and anatomical multimodality tissue contrast. In practice, however, there are challenges associated with considerations of scanning costs, patient comfort, and scanning time that constrain how much data can be acquired in clinical or research studies. In this paper, we explore the possibility of generating high-resolution and multimodal images from low-resolution single-modality imagery. We propose the weakly-supervised joint convolutional sparse coding to simultaneously solve the problems of super-resolution (SR) and cross-modality image synthesis. The learning process requires only a few registered multimodal image pairs as the training set. Additionally, the quality of the joint dictionary learning can be improved using a larger set of unpaired images 1 . To combine unpaired data from different image resolutions/modalities, a hetero-domain image alignment term is proposed. Local image neighborhoods are naturally preserved by operating on the whole image domain (as opposed to image patches) and using joint convolutional sparse coding. The paired images are enhanced in the joint learning process with unpaired data and an additional maximum mean discrepancy term, which minimizes the dissimilarity between their feature distributions. Experiments show that the proposed method outperforms state-of-the-art techniques on both SR reconstruction and simultaneous SR and cross-modality synthesis.

A well-defined thermoresponsive poly(ethylene glycol)-block-poly(N-isopropylacrylamide)-block-poly(ε-caprolactone) (PEG(43)-b-PNIPAM(82)-b-PCL(87)) triblock copolymer was synthesized by combination of atom transfer radical polymerization (ATRP), ring-opening polymerization (ROP), and click chemistry. The synthesis included the four steps, and all the structures of the polymers were determined. The thermoresponsive triblock copolymer can disperse in water at room temperature to form core-shell-corona micelles with the hydrophobic PCL block as core, the thermoresponsive PNIPAM block as shell, and the hydrophilic PEG block as corona. At temperatures above the lower critical solution temperature (LCST) of the PNIPAM block, the PNIPAM chains gradually collapse on the PCL core to shrink the size and change the structure of the resultant core-shell-corona micelles with temperature increasing.

SummaryThe objective of this review was to systematically evaluate common vetch seeds as a potential feedstuff for animals, by summarizing and discussing the available published literature covering their nutritional composition as well as their content of antinutritional factors and potential techniques for their reduction. In addition, animal feeding studies that have investigated the effect of inclusion of common vetch seeds on animal growth and performance were identified and evaluated to stimulate interest in their use as a good source of nutrients for inclusion in animal diets. The collective literature shows that common vetch seeds are a less costly (in comparison with alternatives) and rich source of protein and minerals for farmed animals, are of high digestibility and have a high energy content, and can be used to partially or totally replace soya bean meal and/or to replace a large proportion of cereals in the diet. Furthermore, the literature shows that common vetch seeds contain a range of antinutritional factors which, if they are to be utilized in non-ruminant diets and to increase their utilizing efficiency, need to be removed or inactivated. This can be achieved via certain pre-processing methods, the combination of which may deliver better results.

No abstract

No abstract

No abstract

Large-scale floating-point matrix multiplication is a fundamental kernel in many scientific and engineering applications. Most existing work only focus on accelerating matrix multiplication on FPGA by adopting a linear systolic array. This paper towards the extension of this architecture by proposing a scalable and highly configurable multi-array architecture. In addition, we propose a work-stealing scheme to ensure the equality in the workload partition among multiple linear arrays. Furthermore, an analytical model is developed to determine the optimal design parameters. Experiments on a real-life convolutional neural network (CNN) show that we can obtain the optimal extension of the linear array architecture.