An equivalent distributed magnetic current based FDTD method for the calculation of e-fields induced by gradient coils in MRI

This paper evaluates a low-frequency FDTD method applied to the problem of induced E-fields/eddy currents in the human body resulting from the pulsed magnetic field gradients in MRI. In this algorithm, a distributed equivalent magnetic current (DEMC) is proposed as the electromagnetic source and is obtained by quasistatic calculation of the empty coil's vector potential or measurements therein. This technique circumvents the discretizing of complicated gradient coil geometries into a mesh of Yee cells, and thereby enables any type of gradient coil modeling or other complex low frequency sources. The proposed method has been verified against an example with an analytical solution. Results are presented showing the spatial distribution of gradient-induced electric fields in a multilayered spherical phantom model and a complete body model.

JORNAL SUPER NOTÍCIA: SENSACIONALISMO OU POPULARIZAÇÃO DA NOTÍCIA?

Esta dissertação apresenta um estudo do jornal impresso, com foco no jornal diário, delimitado no tabloide Super Notícia , de Belo Horizonte (MG). A escolha de tal impresso se deu por ser ele um fenômeno editorial de vendas e de publicidade, dentro dos estilos reconhecidos como popular e popularesco, com páginas diárias contendo desde notícia policial até receita de bolo. Ao analisar um objeto jornalístico é necessário verificar de que modo as matérias são veiculadas nas páginas desse impresso. Para tanto, nossa análise busca mostrar as relações entre os gêneros jornalísticos para então classificar o objeto de estudo. A metodologia privilegia a conceituação de jornalismo, notícia, reportagem e jornalismo sensacionalista e contempla também a pesquisa de campo, com visitas in loco na redação do periódico e entrevista com seus agentes. Quanto ao método de trabalho, a pesquisa qualitativa segue o modelo teórico-crítico. Foram feitas análises morfológica e de conteúdo do corpus, para identificar, caracterizar e classificar o objeto estudado. Do ponto de vista jornalístico, os aportes teóricos estão ancorados em Marques de Melo e outros autores que construíram teoria sobre Jornalismo. No que concerne ao objeto de estudo, no segmento tabloide os resultados apontam para uma nova visão do jornalismo atual.

An exploration of the effect of super large n-tuples on single layer RAMnets

N-tuple recognition systems (RAMnets) are normally modeled using a small number of input lines to each RAM, because the address space grows exponentially with the number of inputs. It is impossible to implement an arbitrarily-large address space as physical memory. But given modest amounts of training data, correspondingly modest numbers of bits will be set in that memory. Hash arrays can therefore be used instead of a direct implementation of the required address space. This paper describes some exploratory experiments using the hash array technique to investigate the performance of RAMnets with very large numbers of input lines. An argument is presented which concludes that performance should peak at a relatively small n-tuple size, but the experiments carried out so far contradict this. Further experiments are needed to confirm this unexpected result.

Training with noise is equivalent to Tikhonov regularization

It is well known that the addition of noise to the input data of a neural network during training can, in some circumstances, lead to significant improvements in generalization performance. Previous work has shown that such training with noise is equivalent to a form of regularization in which an extra term is added to the error function. However, the regularization term, which involves second derivatives of the error function, is not bounded below, and so can lead to difficulties if used directly in a learning algorithm based on error minimization. In this paper we show that, for the purposes of network training, the regularization term can be reduced to a positive definite form which involves only first derivatives of the network mapping. For a sum-of-squares error function, the regularization term belongs to the class of generalized Tikhonov regularizers. Direct minimization of the regularized error function provides a practical alternative to training with noise.

Super-hydrophobic surface of bulk carbon nanotubes compacted by spark plasma sintering followed by modification with polytetrofluorethylene

The surfaces of bulk carbon nanotubes compacted by plasma spark sintering have been modified with polytetrofluorethylene, thereby producing a super-hydrophobic surface with a contact angle above 160°. The surface roughness and air trapped in pores and between the polytetrofluorethylene particles are responsible for the super-hydrophobility. The material can be machined into desired shapes with fine and complex channels, allowing internal surfaces to also be super-hydrophobic.

Super-high temperature sensitivity of long-period gratings in B/Ge co-doped fiber

Long period fiber grating (LPFG) can be used as active gain controlling device in EDFA. However, LPFGs fabricated in the standard telecom fiber only have a typical temperature sensitivity of 3-10nm/100°C, which may not be sufficient for implementing tuneable filters capable of wide tuning range and high tuning efficiency. In this paper, we report a theoretical and experimental investigation of thermal properties of LPFGs fabricated in B/Ge co-doped optical fiber. We have found that the temperature sensitivity of the LPFGs in the B/Ge fiber is considerably increased compared with those produced in the standard fiber. The LPFGs written in the B/Ge fiber have achieved, on average, one order of magnitude higher sensitivity than that of the LPFGs produced in the standard telecom fiber. We have also identified that the thermal response of LPFG is strongly dependent on the order of the coupled resonant cladding mode. The maximum sensitivity of 1.75nm/°C achieved by the 10th cladding mode of the 240μm LPFG is nearly 24 times that of the minimum value (0.075nm/C) exhibited by the 30th mode of the 34μm LPFG. Such devices may lead to high-efficiency and low-cost thermal/electrical tunable loss filters or sensors with extremely high temperature resolution.