Parametric accelerated failure time models with random effects and an application to kidney transplant survival

Accelerated failure time models with a shared random component are described, and are used to evaluate the effect of explanatory factors and different transplant centres on survival times following kidney transplantation. Different combinations of the distribution of the random effects and baseline hazard function are considered and the fit of such models to the transplant data is critically assessed. A mixture model that combines short- and long-term components of a hazard function is then developed, which provides a more flexible model for the hazard function. The model can incorporate different explanatory variables and random effects in each component. The model is straightforward to fit using standard statistical software, and is shown to be a good fit to the transplant data. Copyright (C) 2004 John Wiley Sons, Ltd.

Purified galactooligosaccharide, derived from a mixture produced by the enzymic activity of Bifidobacterium bifidum, reduces Salmonella enterica serovar Typhimurium adhesion and invasion in vitro and in vivo

The prebiotic Bimuno (R) is a mixture containing galactooligosaccharides (GOSs), produced by the galactosyltransferase activity of Bifidobacterium bifidum NCIMB 411 71 using lactose as the substrate Previous in vivo and in vitro studies demonstrating the efficacy of Bimuno (R) in reducing Salmonella enterica serovar Typhimurium (S Typhimurium) colonization did not ascertain whether or not the protective effects could be attributed to the prebiotic component GOS Here we wished to test the hypothesis that GOS, derived from Bimuno (R) may confer the direct anti-invasive and protective effects of Bimuno (R) In this study the efficacy of Bimuno (R), a basal solution of Bimuno (R) without GOS [which contained glucose, galactose, lactose, maltodextrin and gum arabic in the same relative proportions (w/w) as they are found in Bimuno (R)] and purified GOS to reduce S Typhimurium adhesion and invasion was assessed using a series of in vitro and in vivo models The novel use of three dimensionally cultured HT-29-16E cells to study prebiotics in vitro demonstrated that the presence of similar to 5 mg Bimuno (R) ml(-1) or similar to 2 5 mg GOS ml(-1) significantly reduced the invasion of S Typhimurium (SL1344nal(r)) (P<0 0001) Furthermore, similar to 2 5 mg GOS ml(-1) significantly reduced the adherence of S Typhimurium (SU 344nal(r)) (P<0 0001) It was demonstrated that cells produced using this system formed multi-layered aggregates of cells that displayed excellent formation of brush borders and tight junctions In the murine ligated deal gut loops, the presence of Bimuno (R) or GOS prevented the adherence or invasion of S Typhimurium to enterocytes, and thus reduced its associated pathology This protection appeared to correlate with significant reductions in the neutral and acidic mucins detected in goblet cells, possibly as a consequence of stimulating the cells to secrete the mucin into the lumen In all assays, Bimuno (R) without GOS conferred no such protection, indicating that the basal solution confers no protective effects against S Typhimurium Collectively, the studies presented here clearly indicate that the protective effects conferred by Bimuno (R) can be attributed to GOS

Construction of neurofuzzy models for imbalanced data classification

We propose a new class of neurofuzzy construction algorithms with the aim of maximizing generalization capability specifically for imbalanced data classification problems based on leave-one-out (LOO) cross validation. The algorithms are in two stages, first an initial rule base is constructed based on estimating the Gaussian mixture model with analysis of variance decomposition from input data; the second stage carries out the joint weighted least squares parameter estimation and rule selection using orthogonal forward subspace selection (OFSS)procedure. We show how different LOO based rule selection criteria can be incorporated with OFSS, and advocate either maximizing the leave-one-out area under curve of the receiver operating characteristics, or maximizing the leave-one-out Fmeasure if the data sets exhibit imbalanced class distribution. Extensive comparative simulations illustrate the effectiveness of the proposed algorithms.