Institution
Dublin City University
Education•Dublin, Ireland•
About: Dublin City University is a education organization based out in Dublin, Ireland. It is known for research contribution in the topics: Machine translation & Laser. The organization has 5904 authors who have published 17178 publications receiving 389376 citations. The organization is also known as: National Institute for Higher Education, Dublin & DCU.
Topics: Machine translation, Laser, Irish, Population, Context (language use)
Papers published on a yearly basis
Papers
More filters
••
TL;DR: The cell tests performed with primary human immune-competent cells confirmed the excellent biocompatibility of USIRONs and the potential of the DHAA-Fe(3)O(4) nanoparticles as negative contrast agents for MRI with optimal hydrodynamic size for extended blood circulation times was confirmed.
Abstract: Ultrasmall superparamagnetic Fe3O4 nanoparticles (USIRONs) were synthesized by a novel, easily scalable chemical reduction of colloidal iron hydroxide under hydrothermal conditions. The average crystallite size (5.1 ± 0.5 nm) and good crystallinity of the samples were supported by HR-TEM analysis and the saturation magnetization value (47 emu g–1). Vitamin C, used as a chemical reducing agent, also served as a capping agent in the oxidized form (dehydroascorbic acid, DHAA) to impart nanoparticles with exceptional solubility and stability in water, PBS buffer, and cell culture medium. Detailed physicochemical analysis of the USIRON suspensions provided insight into the magnetic ordering phenomena within the colloid, arising from the formation of uniform clusters displaying a hydrodynamic size of 41 nm. Phantom experiments on the contrast agent (clinical 3 T MRI scanner) revealed an enhanced r2/r1 ratio of 36.4 (r1= 5 s–1 mM–1 and r2= 182 s–1 mM–1) when compared to the clinically approved agents. The potent...
254 citations
••
TL;DR: The known mechanisms for Au NP biosynthesis in viable fungi and fungal protein extracts are described and the most suitable bioreactors for industrial AuNP biosynthesis are discussed.
Abstract: Gold nanoparticles (AuNPs) are a widespread research tool because of their oxidation resistance, biocompatibility and stability. Chemical methods for AuNP synthesis often produce toxic residues that raise environmental concern. On the other hand, the biological synthesis of AuNPs in viable microorganisms and their cell-free extracts is an environmentally friendly and low-cost process. In general, fungi tolerate higher metal concentrations than bacteria and secrete abundant extracellular redox proteins to reduce soluble metal ions to their insoluble form and eventually to nanocrystals. Fungi harbour untapped biological diversity and may provide novel metal reductases for metal detoxification and bioreduction. A thorough understanding of the biosynthetic mechanism of AuNPs in fungi is needed to reduce the time of biosynthesis and to scale up the AuNP production process. In this review, we describe the known mechanisms for AuNP biosynthesis in viable fungi and fungal protein extracts and discuss the most suitable bioreactors for industrial AuNP biosynthesis.
253 citations
•
TL;DR: This work shows that a naive pseudo-labeling overfits to incorrect pseudo-labels due to the so-called confirmation bias and demonstrates that mixup augmentation and setting a minimum number of labeled samples per mini-batch are effective regularization techniques for reducing it.
Abstract: Semi-supervised learning, i.e. jointly learning from labeled and unlabeled samples, is an active research topic due to its key role on relaxing human supervision. In the context of image classification, recent advances to learn from unlabeled samples are mainly focused on consistency regularization methods that encourage invariant predictions for different perturbations of unlabeled samples. We, conversely, propose to learn from unlabeled data by generating soft pseudo-labels using the network predictions. We show that a naive pseudo-labeling overfits to incorrect pseudo-labels due to the so-called confirmation bias and demonstrate that mixup augmentation and setting a minimum number of labeled samples per mini-batch are effective regularization techniques for reducing it. The proposed approach achieves state-of-the-art results in CIFAR-10/100, SVHN, and Mini-ImageNet despite being much simpler than other methods. These results demonstrate that pseudo-labeling alone can outperform consistency regularization methods, while the opposite was supposed in previous work. Source code is available at this https URL.
253 citations
••
TL;DR: In this paper, the early-time afterglow light curve carries information about 0, which determines the time of the peak of the gamma-ray burst (GRB) peak.
Abstract: Context. Gamma-ray burst (GRB) emission is believed to originate in highly relativistic fireballs. Aims. Currently, only lower limits were securely set to the initia l fireball Lorentz factor 0. We aim to provide a direct measure of 0. Methods. The early-time afterglow light curve carries information about 0, which determines the time of the afterglow peak. We have obtained early observations of the near-infrared afte rglows of GRB 060418 and GRB 060607A with the REM robotic telescope. Results. For both events, the afterglow peak could be clearly singled out, allowing a firm determination of the fireball Lorentz of 0∼ 400, fully confirming the highly relativistic nature of GRB fi reballs. The deceleration radius was inferred to be Rdec≈ 10 17 cm. This is much larger than the internal shocks radius (believed to power the prompt emission), thus providing further evidence for a different origin of the prompt and afterglow stages of the GRB.
253 citations
••
TL;DR: In this paper, the fuel cell cost, durability and performances challenges associated with using of fuel cell technology for transport applications are detailed and reviewed, and recent developments that deal with the proposed challenges are reported.
253 citations
Authors
Showing all 6059 results
Name | H-index | Papers | Citations |
---|---|---|---|
Joseph Wang | 158 | 1282 | 98799 |
David Cameron | 154 | 1586 | 126067 |
David Taylor | 131 | 2469 | 93220 |
Gordon G. Wallace | 114 | 1267 | 69095 |
David A. Morrow | 113 | 598 | 56776 |
G. Hughes | 103 | 957 | 46632 |
David Wilson | 102 | 757 | 49388 |
Muhammad Imran | 94 | 3053 | 51728 |
Haibo Zeng | 94 | 604 | 39226 |
David Lloyd | 90 | 1017 | 37691 |
Vikas Kumar | 89 | 859 | 39185 |
Luke P. Lee | 84 | 413 | 22803 |
James Chapman | 82 | 483 | 36468 |
Muhammad Iqbal | 77 | 961 | 23821 |
Michael C. Berndt | 76 | 228 | 16897 |