Non‐linear model reduction for uncertainty quantification in large‐scale inverse problems

We present a model reduction approach to the solution of large‐scale statistical inverse problems in a Bayesian inference setting. A key to the model reduction is an efficient representation of the non‐linear terms in the reduced model. To achieve this, we present a formulation that employs masked projection of the discrete equations; that is, we compute an approximation of the non‐linear term using a select subset of interpolation points. Further, through this formulation we show similarities among the existing techniques of gappy proper orthogonal decomposition, missing point estimation, and empirical interpolation via coefficient‐function approximation. The resulting model reduction methodology is applied to a highly non‐linear combustion problem governed by an advection–diffusion‐reaction partial differential equation (PDE). Our reduced model is used as a surrogate for a finite element discretization of the non‐linear PDE within the Markov chain Monte Carlo sampling employed by the Bayesian inference approach. In two spatial dimensions, we show that this approach yields accurate results while reducing the computational cost by several orders of magnitude. For the full three‐dimensional problem, a forward solve using a reduced model that has high fidelity over the input parameter space is more than two million times faster than the full‐order finite element model, making tractable the solution of the statistical inverse problem that would otherwise require many years of CPU time. Copyright © 2009 John Wiley & Sons, Ltd.     

Acute renal failure in Kuwait--a prospective study

This prospective study was conducted over a period of 18 months (February 1989 to July 1990) in the State of Kuwait. It covered a population of 1,024,211 and eight multidisciplinary hospitals with an in-patient admission of 118,079 per year. Two hundred and twenty-six adult patients with acute renal failure (ARF) were seen and followed up by nephrologists. This made the calculated annual incidence of ARF 14.7 per 100,000 population, nearly five times that reported by the EDTA registry (Biesenbach et al. 1991). Drugs, sepsis and volume depletion were the most frequent causes, with sepsis resulting in 36% cause specific mortality compared to zero mortality with the other two. The overall mortality rate was only 14% which clearly indicated a markedly improved prognosis in cases of ARF. The prognosis in ARF depended on two major factors, viz. the type of aetiological insult and the presence of predisposing associated medical illnesses. Multiple insults, though common, do not affect the mortality rate. Secondary sepsis or gastrointestinal bleeding as a cause of death in ARF was rarely seen in our study. Those who required dialytic support for renal failure had a 45% patient mortality rate in general. Over 40% of our patients were 60 years or older compared with only 3.5% in the local population. This indicated old age as a major risk factor in the development of ARF. The overall mortality in the elderly did not differ from that in the young, but sepsis in the elderly carried a mortality rate of 60% compared to only 14.8% in the younger age group.     

Left ventricular radial tagging acquisition using gradient-recalled-echo techniques: sequence optimization

Myocardial tagging with magnetic resonance (MR) imaging offers unique possibilities for noninvasive left ventricular (LV) strain analysis. True three-dimensional strain analysis can be achieved with tags implemented in cardiac short axis and long axis images. Spin-echo (SE) techniques have been used for these studies. However, this approach is time-consuming: images at different phases of the cardiac cycle have to be obtained in successive measurements and hence the total number of measurements equals the number of time frames. Moreover, the images are often degraded by flow and motion artifacts. The purpose of this study was to optimize a faster and more robust MR tagging sequence for use on a clinical whole-body 1 T MR system with optimal persistence of the tags during the entire cardiac cycle. The tagging pulses were implemented in gradient-recalled-echo (GRE) sequences and compared to SE-based acquisitions. The effects of the use of flow-compensating gradients, the excitation angles, and the angles of the saturation pulses have been studied with MR signal simulations and in comparative measurements in volunteers. GRE acquisitions with flow-compensating gradients are robust techniques for myocardial tagging acquisitions. Use of optimized flip angles and saturation pulses can significantly improve delineation of the tag and can be used up to at least 700 ms after the R-wave. Therefore, LV tagging with GRE acquisitions using optimized MR parameters is a robust and promising technique.     

Compact Patch Antenna Array for 60 GHz Millimeter-Wave Broadband Applications

Wireless Personal Communications - In this study, a wide-band compact patch antenna array is developed for 60 GHz band applications. The antenna array consists of eight identical elements....     

Plant-derived antimicrobials reduce Listeria monocytogenes virulence factors in vitro, and down-regulate expression of virulence genes

Listeria monocytogenes (LM) is a major foodborne pathogen causing septicemia, meningitis and death in humans. LM infection is preceded by its attachment to and invasion of human intestinal epithelium followed by systemic spread. The major virulence factors in LM include motility, hemolysin and lecithinase production. Reducing LM attachment to and invasion of host tissue and production of virulence factors could potentially control listeriosis in humans. This study investigated the efficacy of sub-inhibitory concentrations (SICs, concentrations not inhibiting bacterial growth) of three, generally regarded as safe (GRAS)-status, plant-derived antimicrobial compounds in reducing LM attachment to and invasion of human colon adenocarcinoma (Caco-2) and human brain microvascular endothelial cells (HBMEC). Additionally, the effect of these compounds on the aforementioned LM virulence factors was studied. The compounds and their respective SICs used relative to their MICs were trans-cinnamaldehyde (TC 0.50mM, 0.75mM with the MIC of 0.90mM), carvacrol (CR 0.50mM, 0.65mM with the MIC of 0.75mM), and thymol (TY 0.33mM, 0.50mM with the MIC of 0.60mM). All three-plant antimicrobials reduced LM adhesion to and invasion of Caco-2 and HBMEC (p<0.05). The compounds also decreased LM motility, hemolysin production and lecithinase activity (p<0.05). Real-time PCR data revealed that TC, CR, and TY down-regulated the expression of LM virulence genes by >3.0 folds compared to controls (p<0.05). Results suggest that TC, CR, and TY could potentially be used to control LM infection; however, in vivo studies are necessary to validate these results.     

Artificial Enzymes for Diels-Alder Reactions

The Diels-Alder (DA) reaction is a cycloaddition of a conjugated diene and an alkene (dienophile) leading to the formation of a cyclohexene derivative through a concerted mechanism. As DA reactions generally proceed with a high degree of regio- and stereoselectivity, they are widely used in synthetic organic chemistry. Considering eco-conscious public and governmental movements, efforts are now directed towards the development of synthetic processes that meet environmental concerns. Artificial enzymes, which can be developed to catalyze abiotic reactions, appear to be important synthetic tools in the synthetic biology field. This review describes the different strategies used to develop protein-based artificial enzymes for DA reactions, including for in cellulo approaches.     

Synthetic Phosphoethanolamine Cellobiose Promotes Escherichia coli Biofilm Formation and Congo Red Binding

Urinary tract infections (UTIs) are caused by bacteria growing in complex, multicellular enclosed aggregates known as biofilms. Recently, a zwitterionic cellulose derivative produced in Escherichia coli (E. coli) was determined to play an important role in the formation and assembly of biofilms. In order to produce a minimal, yet structurally defined tool compound to probe the biology of the naturally occurring polymer, we have synthesized a zwitterionic phosphoethanolamine cellobiose (pEtN cellobiose) and evaluated its biofilm activity in the Gram-negative bacterium E. coli, a pathogen implicated in the pathogenesis of UTIs. The impact of synthetic pEtN cellobiose on biofilm formation was examined via colorimetric assays which revealed an increase in cellular adhesion to an abiotic substrate compared to untreated samples. Additionally, Congo red binding assays indicate that culturing E. coli in the presence of pEtN cellobiose enhances Congo Red binding to bacterial cells. These results reveal new opportunities to study the impact glycopolymers have on cellular adhesion in Gram-negative pathogens.     

Influence of lithium substitution on the orange emission in manganese doped ZnO thin films

Lithium (Li) and Manganese (Mn) co-doped zinc oxide (ZnO) thin films have been deposited on glass substrates by sol–gel spin-coating technique. The influence of the defect states on the orange emission has been studied by altering the defect chemistry through appropriate Li substitution. The results of the defect related fluorescence in the blue and green regions have been combined to elucidate comprehensively the defect-mediated energy transfer mechanism of orange emission in Li and Mn co-doped ZnO thin films.