Brown University

About: Brown University is a education organization based out in Providence, Rhode Island, United States. It is known for research contribution in the topics: Population & Poison control. The organization has 35778 authors who have published 90896 publications receiving 4471489 citations. The organization is also known as: brown.edu & Brown.

A Randomized, Controlled Trial of Interferon Alfa-2b Alone and after Prednisone Withdrawal for the Treatment of Chronic Hepatitis B

Abstract: Background and Methods. Chronic hepatitis B is a common and often progressive liver disorder for which there is no accepted therapy. To assess the efficacy of treatment with interferon, we randomly assigned patients with chronic hepatitis B to one of the following regimens: prednisone for 6 weeks followed by 5 million units of recombinant interferon alfa-2b daily for 16 weeks; placebo followed by 5 million units of interferon daily for 16 weeks; placebo followed by 1 million units of interferon daily for 16 weeks; or observation with no treatment. Results. Hepatitis B e antigen and hepatitis B viral DNA disappeared from serum significantly more often in the patients given prednisone plus interferon (16 of 44 patients, or 36 percent) or 5 million units of interferon alone (15 of 41; 37 percent) than in the untreated controls (3 of 43; 7 percent; P<0.001); the difference between those given 1 million units of interferon (7 of 41; 17 percent) and the controls was not significant. The strongest indep...

Universe reheating after inflation

Abstract: We study the problem of scalar particle production after inflation by an inflaton field which is oscillating rapidly relative to the expansion of the universe. We use the framework of the chaotic inflation scenario with quartic and quadratic inflaton potentials. Particles produced are described by a quantum scalar field \ensuremath{\chi}, which is coupled to the inflaton via linear and quadratic couplings. The particle production effect is studied using the standard technique of Bogolyubov transformations. Particular attention is paid to parametric resonance phenomena which take place in the presence of the quickly oscillating inflaton field. We have found that in the region of applicability of perturbation theory the effects of parametric resonance are crucial, and estimates based on first-order Born approximation often underestimate the particle production. In the case of the quartic inflaton potential V(cphi)=\ensuremath{\lambda}${\mathit{cphi}}^{4}$, the particle production process is very efficient for either type of coupling between the inflaton field and the scalar field \ensuremath{\chi} even for small values of coupling constants. The energy density of the universe after the decay of the inflaton oscillations is in this case a factor [\ensuremath{\lambda} ln(1/\ensuremath{\lambda})${]}^{\mathrm{\ensuremath{-}}1}$ times larger than the corresponding estimates based on first-order Born approximation. In the case of the quadratic inflaton potential the reheating process depends crucially on the type of coupling between the inflaton and the scalar field \ensuremath{\chi} and on the magnitudes of the coupling constants. If the inflaton coupling to fermions and its linear (in inflaton field) coupling to scalar fields are suppressed, then, as previously discussed by Kofman, Linde, and Starobinsky, the inflaton field will eventually decouple from the rest of the matter, and the residual inflaton oscillations may provide the (cold) dark matter of the universe. In the case of the quadratic inflaton potential we obtain the lowest and the highest possible bounds on the effective energy density of the inflaton field when it freezes out.

Infinite latent feature models and the Indian buffet process

Abstract: We define a probability distribution over equivalence classes of binary matrices with a finite number of rows and an unbounded number of columns. This distribution is suitable for use as a prior in probabilistic models that represent objects using a potentially infinite array of features. We identify a simple generative process that results in the same distribution over equivalence classes, which we call the Indian buffet process. We illustrate the use of this distribution as a prior in an infinite latent feature model, deriving a Markov chain Monte Carlo algorithm for inference in this model and applying the algorithm to an image dataset.