A Conditional Density Estimation Partition Model Using Logistic Gaussian Processes.

Conditional density estimation (density regression) estimates the distribution of a response variable y conditional on covariates x. Utilizing a partition model framework, a conditional density estimation method is proposed using logistic Gaussian processes. The partition is created using a Voronoi tessellation and is learned from the data using a reversible jump Markov chain Monte Carlo algorithm. The Markov chain Monte Carlo algorithm is made possible through a Laplace approximation on the latent variables of the logistic Gaussian process model. This approximation marginalizes the parameters in each partition element, allowing an efficient search of the posterior distribution of the tessellation. The method has desirable consistency properties. In simulation and applications, the model successfully estimates the partition structure and conditional distribution of y.

Systemic anti-inflammatory therapy aided by double-headed nanoparticles in a canine model of acute intraocular inflammation.

Novel approaches circumventing blood-ocular barriers in systemic drug delivery are lacking. We hypothesize receptor-mediated delivery of curcumin (CUR) across intestinal and ocular barriers leads to decreased inflammation in a model of lens-induced uveitis. CUR was encapsulated in double-headed polyester nanoparticles using gambogic acid (GA)-coupled polylactide-co-glycolide (PLGA). Orally administered PLGA-GA2-CUR led to notable aqueous humor CUR levels and was dosed (10 mg/kg twice daily) to adult male beagles (n = 8 eyes) with induced ocular inflammation. Eyes were evaluated using a semiquantitative preclinical ocular toxicology scoring (SPOTS) and compared to commercial anti-inflammatory treatment (oral carprofen 2.2 mg/kg twice daily) (n = 8) and untreated controls (n = 8). PLGA-GA2-CUR offered improved protection compared with untreated controls and similar protection compared with carprofen, with reduced aqueous flare, miosis, and chemosis in the acute phase (<4 hours). This study highlights the potential of PLGA-GA2 nanoparticles for systemic drug delivery across ocular barriers.

Postnatal changes of plasma amino acids in suckling pigs.

Amino acids, ammonia, urea, orotate, and nitrate plus nitrite (stable oxidation products of nitric oxide) were determined in plasma of 1- to 21-d-old suckling pigs. Jugular venous blood samples were obtained from pigs at 1, 3, 7, 14, and 21 d of age for analysis of plasma amino acids and metabolites by HPLC and enzymatic methods. Plasma concentrations of arginine and its immediate precursors (citrulline and ornithine) decreased (P < 0.01) progressively (20 to 41%) with increasing age from 3 to 14 d. Plasma concentrations of glutamine declined (P < 0.01) progressively (10 to 31%) during the 1st wk of life. Plasma concentrations of branched-chain amino acids, threonine, and alanine decreased (P < 0.01) (5 to 12%) in 14- and 21-d-old pigs, compared with 1- and 3-d-old pigs. There were no postnatal changes (P > 0.05) in plasma concentrations of other amino acids. Plasma concentrations of ammonia increased (P < 0.01) by 18 and 46%, whereas those of nitrate plus nitrite decreased (P < 0.01) by 16 and 29%, in 7- and 14-d-old pigs, respectively, compared with 1- to 3-d-old pigs. Because arginine plays a crucial role in ammonia detoxification via the hepatic urea cycle and is the physiological substrate for nitric oxide synthesis, our results of the decreased plasma concentrations of arginine and nitrate plus nitrite, as well as the increased plasma ammonia concentration, indicate a hitherto unrecognized deficiency of arginine in 7- to 21-d-old suckling pigs. Arginine is an essential amino acid for piglets and has a great potential to enhance neonatal growth; therefore, further studies are necessary to elucidate the mechanism responsible for arginine deficiency in sow-reared piglets and to identify hormonal and metabolic means for improving neonatal arginine nutrition and growth.

Bayesian wavelet networks for nonparametric regression.

Radial wavelet networks have recently been proposed as a method for nonparametric regression. In this paper we analyze their performance within a Bayesian framework. We derive probability distributions over both the dimension of the networks and the network coefficients by placing a prior on the degrees of freedom of the model. This process bypasses the need to test or select a finite number of networks during the modeling process. Predictions are formed by mixing over many models of varying dimension and parameterization.We show that the complexity of the models adapts to the complexity of the data and produces good results on a number of benchmark test series.

Bayesian survival analysis using a MARS model.

A Bayesian multivariate adaptive regression spline fitting approach is used to model univariate and multivariate survival data with censoring. The possible models contain the proportional hazards model as a subclass and automatically detect departures from this. A reversible jump Markov chain Monte Carlo algorithm is described to obtain the estimate of the hazard function as well as the survival curve.

A Bayesian semiparametric accelerated failure time model.

A Bayesian semiparametric approach is described for an accelerated failure time model. The error distribution is assigned a Pólya tree prior and the regression parameters a noninformative hierarchical prior. Two cases are considered: the first assumes error terms are exchangeable; the second assumes that error terms are partially exchangeable. A Markov chain Monte Carlo algorithm is described to obtain a predictive distribution for a future observation given both uncensored and censored data.

Bayesian analysis of proportional hazards models built from monotone functions.

We consider the usual proportional hazards model in the case where the baseline hazard, the covariate link, and the covariate coefficients are all unknown. Both the baseline hazard and the covariate link are monotone functions and thus are characterized using a dense class of such functions which arises, upon transformation, as a mixture of Beta distribution functions. We take a Bayesian approach for fitting such a model. Since interest focuses more upon the likelihood, we consider vague prior specifications including Jeffreys's prior. Computations are carried out using sampling-based methods. Model criticism is also discussed. Finally, a data set studying survival of a sample of lung cancer patients is analyzed.

Primary sclerosing cholangitis.

A case of primary sclerosing cholangitis is reported. There was no history of choledocholithiasis, operation or traumatic biliary injury.