Bayesian inference for multidimensional NMR image reconstruction

Reconstruction of an image from a set of projections has been adapted to generate multidimensional nuclear magnetic resonance (NMR) spectra, which have discrete features that are relatively sparsely distributed in space. For this reason, a reliable reconstruction can be made from a small number of projections. This new concept is called Projection Reconstruction NMR (PR-NMR). In this paper, multidimensional NMR spectra are reconstructed by Reversible Jump Markov Chain Monte Carlo (RJMCMC). This statistical method generates samples under the assumption that each peak consists of a small number of parameters: position of peak centres, peak amplitude, and peak width. In order to find the number of peaks and shape, RJMCMC has several moves: birth, death, merge, split, and invariant updating. The reconstruction schemes are tested on a set of six projections derived from the three-dimensional 700 MHz HNCO spectrum of a protein HasA.

Modelling within-host spatiotemporal dynamics of invasive bacterial disease

Mechanistic determinants of bacterial growth, death, and spread within mammalian hosts cannot be fully resolved studying a single bacterial population. They are also currently poorly understood. Here, we report on the application of sophisticated experimental approaches to map spatiotemporal population dynamics of bacteria during an infection. We analyzed heterogeneous traits of simultaneous infections with tagged Salmonella enterica populations (wild-type isogenic tagged strains [WITS]) in wild-type and gene-targeted mice. WITS are phenotypically identical but can be distinguished and enumerated by quantitative PCR, making it possible, using probabilistic models, to estimate bacterial death rate based on the disappearance of strains through time. This multidisciplinary approach allowed us to establish the timing, relative occurrence, and immune control of key infection parameters in a true host-pathogen combination. Our analyses support a model in which shortly after infection, concomitant death and rapid bacterial replication lead to the establishment of independent bacterial subpopulations in different organs, a process controlled by host antimicrobial mechanisms. Later, decreased microbial mortality leads to an exponential increase in the number of bacteria that spread locally, with subsequent mixing of bacteria between organs via bacteraemia and further stochastic selection. This approach provides us with an unprecedented outlook on the pathogenesis of S. enterica infections, illustrating the complex spatial and stochastic effects that drive an infectious disease. The application of the novel method that we present in appropriate and diverse host-pathogen combinations, together with modelling of the data that result, will facilitate a comprehensive view of the spatial and stochastic nature of within-host dynamics. © 2008 Grant et al.

Damage development in an armor ceramic under quasi-static indentation

The objective of the present study is to assess the capabilities of a recently developed mechanism-based model for inelastic deformation and damage in structural ceramics. In addition to conventional lattice plasticity, the model accounts for microcrack growth and coalescence as well as granular flow following comminution. The assessment is made through a coupled experimental/computational study of the indentation response of a commercial armor ceramic. The experiments include examinations of subsurface damage zones along with measurements of residual surface profiles and residual near-surface stresses. Extensive finite element computations are conducted in parallel. Comparisons between experiment and simulation indicate that the most discriminating metric in the assessment is the spatial extent of subsurface damage following indentation. Residual stresses provide additional validation. In contrast, surface profiles of indents are dictated largely by lattice plasticity and thus provide minimal additional insight into the inelastic deformation resulting from microcracking or granular flow. A satisfactory level of correlation is obtained using property values that are either measured directly or estimated from physically based arguments, without undue reliance on adjustable (nonphysical) parameters. © 2011 The American Ceramic Society.

A systems approach to characterise the dissemination process of household water treatment systems in developing countries

Abstract: Starting in the 1980s, household-level water treatment and safe storage systems (HWTS) have been developed as simple, local, user-friendly, and low cost options to improve drinking water quality at the point of use. However, despite conclusive evidence of the health and economic benefits of HWTS, and promotion efforts in over 50 countries in the past 20 years, implementation outcomes have been slow, reaching only 5-10 million regular users. This study attempts to understand the barriers and drivers affecting HWTS implementation. Although existing literature related to HWTS and innovation diffusion theories proposed ample critical factors and recommendations, there is a lack of holistic and systemic approach to integrate these findings. It is proposed that system dynamics modelling can be a promising tool to map the inter-relationships of different critical factors and to understand the structure of HWTS dissemination process, which may lead to identifying high impact, leveraged mitigation strategies to scale-up HWTS adoption and sustained use.

Process development on the monolithic fabrication of an ultra-compact 4×4 optical switch matrix on InP/InGaAsP material

We have fabricated an ultra-compact 4×4 optical matrix on InP/InGaAsP material. 1×4 MMI couplers and TIR mirrors are employed to produce a compact 1×2 mm2 device. A CH4/H2/O2 RIE dry etch process has been used to realize two-level dry etching: deep-etch for both the MMI couplers and the mirrors and shallow-etch for the rest of the routing waveguides. It was found that a metal/dielectric bilayer mask is essential for multi-dry-etch processes and high profile verticality. We have found a Ti intermediate mask for the deep-etch process which is removable by SF6 dry-etch before the following shallow process. Dry-etch removal of the intermediate mask is necessary to protect the deep-etched mirror sidewall.

Viscous layer meshes from level sets on cartesian meshes

The long term goal of our work is to enable rapid prototyping design optimization to take place on geometries of arbitrary size in a spirit of a real time computer game. In recent papers we have reported the integration of a Level Set based geometry kernel with an octree-based cut-Cartesian mesh generator, RANS flow solver and post-processing all within a single piece of software - and all implemented in parallel with commodity PC clusters as the target. This work has shown that it is possible to eliminate all serial bottlenecks from the CED Process. This paper reports further progress towards our goal; in particular we report on the generation of viscous layer meshes to bridge the body to the flow across the cut-cells. The Level Set formulation, which underpins the geometry representation, is used as a natural mechanism to allow rapid construction of conformal layer meshes. The guiding principle is to construct the mesh which most closely approximates the body but remains solvable. This apparently novel approach is described and examples given.

Polymer-based board-level optical interconnects

This paper provides an overview of the rationale behind the significant interest in polymer-based on-board optical links together with a brief review of recently reported work addressing certain challenges in this field. Polymer-based optical links have garnered considerable research attention due to their important functional attributes and compelling cost-benefit advantages in on-board optoelectronic systems as they can be cost-effectively integrated on conventional printed circuit boards. To date, significant work on the polymer materials, their fabrication process and their integration on standard board substrates have enabled the demonstration of numerous high-speed on-board optical links. However, to be deployed in real-world systems, these optoelectronic printed circuit boards (OE PCBs) must also be cost-effective. Here, recent advances in the integration process focusing on simple direct end-fire coupling schemes and the use of low-cost FR4 PCB substrates are presented. Performance of two proof-of-principle 10 Gb/s systems based on this integration method are summarised while work in realising more complex yet compact planar optical components is outlined. © 2011 IEEE.

The Gaussian process density sampler

We present the Gaussian Process Density Sampler (GPDS), an exchangeable generative model for use in nonparametric Bayesian density estimation. Samples drawn from the GPDS are consistent with exact, independent samples from a fixed density function that is a transformation of a function drawn from a Gaussian process prior. Our formulation allows us to infer an unknown density from data using Markov chain Monte Carlo, which gives samples from the posterior distribution over density functions and from the predictive distribution on data space. We can also infer the hyperparameters of the Gaussian process. We compare this density modeling technique to several existing techniques on a toy problem and a skullreconstruction task.

New space-time trellis codes for two-antenna quasi-static channels

New space-time trellis codes with four- and eight-level phase-shift keying (PSK) and 16-phase quadrature amplitude modulation (QAM) for two transmit antennas in slow-fading channels are presented in this paper. Unlike most of the codes that are reported in the literature, the proposed codes are specifically designed to minimize the frame error probability from a union-bound perspective. The performance of the proposed codes with various memory orders and receive antennas is evaluated by simulation. It is shown that the proposed codes outperform previously known codes in all studied cases.

The Gaussian mixture MCMC particle algorithm for dynamic cluster tracking

We present a novel filtering algorithm for tracking multiple clusters of coordinated objects. Based on a Markov chain Monte Carlo (MCMC) mechanism, the new algorithm propagates a discrete approximation of the underlying filtering density. A dynamic Gaussian mixture model is utilized for representing the time-varying clustering structure. This involves point process formulations of typical behavioral moves such as birth and death of clusters as well as merging and splitting. For handling complex, possibly large scale scenarios, the sampling efficiency of the basic MCMC scheme is enhanced via the use of a Metropolis within Gibbs particle refinement step. As the proposed methodology essentially involves random set representations, a new type of estimator, termed the probability hypothesis density surface (PHDS), is derived for computing point estimates. It is further proved that this estimator is optimal in the sense of the mean relative entropy. Finally, the algorithm's performance is assessed and demonstrated in both synthetic and realistic tracking scenarios. © 2012 Elsevier Ltd. All rights reserved.

Industrial ecology at factory level: A prototype methodology

The concept of sustainable manufacturing is a form of pollution prevention that integrates environmental considerations in the production of goods while focusing on efficient resource use. Taking the industrial ecology perspective, this efficiency comes from improved resource flow management. The assessment of material, energy and waste resource flows, therefore, offers a route to viewing and analysing a manufacturing system as an ecosystem using industrial ecology biological analogy and can, in turn, support the identification of improvement opportunities in the material, energy and waste flows. This application of industrial ecology at factory level is absent from the literature. This article provides a prototype methodology to apply the concepts of industrial ecology using material, energy and waste process flows to address this gap in the literature. Various modelling techniques were reviewed and candidates selected to test the prototype methodology in an industrial case. The application of the prototype methodology showed the possibility of using the material, energy and waste resource flows through the factory to link manufacturing operations and supporting facilities, and to identify potential improvements in resource use. The outcomes of the work provide a basis to build the specifications for a modelling tool that can support those analysing their manufacturing system to improve their environmental performance and move towards sustainable manufacturing. © IMechE 2012.

Gaussian Process Vine Copulas for Multivariate Dependence

Copulas allow to learn marginal distributions separately from the multivariate dependence structure (copula) that links them together into a density function. Vine factorizations ease the learning of high-dimensional copulas by constructing a hierarchy of conditional bivariate copulas. However, to simplify inference, it is common to assume that each of these conditional bivariate copulas is independent from its conditioning variables. In this paper, we relax this assumption by discovering the latent functions that specify the shape of a conditional copula given its conditioning variables We learn these functions by following a Bayesian approach based on sparse Gaussian processes with expectation propagation for scalable, approximate inference. Experiments on real-world datasets show that, when modeling all conditional dependencies, we obtain better estimates of the underlying copula of the data.

State estimation schemes for independent component coupled hidden Markov models

Conventional Hidden Markov models generally consist of a Markov chain observed through a linear map corrupted by additive noise. This general class of model has enjoyed a huge and diverse range of applications, for example, speech processing, biomedical signal processing and more recently quantitative finance. However, a lesser known extension of this general class of model is the so-called Factorial Hidden Markov Model (FHMM). FHMMs also have diverse applications, notably in machine learning, artificial intelligence and speech recognition [13, 17]. FHMMs extend the usual class of HMMs, by supposing the partially observed state process is a finite collection of distinct Markov chains, either statistically independent or dependent. There is also considerable current activity in applying collections of partially observed Markov chains to complex action recognition problems, see, for example, [6]. In this article we consider the Maximum Likelihood (ML) parameter estimation problem for FHMMs. Much of the extant literature concerning this problem presents parameter estimation schemes based on full data log-likelihood EM algorithms. This approach can be slow to converge and often imposes heavy demands on computer memory. The latter point is particularly relevant for the class of FHMMs where state space dimensions are relatively large. The contribution in this article is to develop new recursive formulae for a filter-based EM algorithm that can be implemented online. Our new formulae are equivalent ML estimators, however, these formulae are purely recursive and so, significantly reduce numerical complexity and memory requirements. A computer simulation is included to demonstrate the performance of our results. © Taylor & Francis Group, LLC.