Augmented attribute representations

We propose a new learning method to infer a mid-level feature representation that combines the advantage of semantic attribute representations with the higher expressive power of non-semantic features. The idea lies in augmenting an existing attribute-based representation with additional dimensions for which an autoencoder model is coupled with a large-margin principle. This construction allows a smooth transition between the zero-shot regime with no training example, the unsupervised regime with training examples but without class labels, and the supervised regime with training examples and with class labels. The resulting optimization problem can be solved efficiently, because several of the necessity steps have closed-form solutions. Through extensive experiments we show that the augmented representation achieves better results in terms of object categorization accuracy than the semantic representation alone. © 2012 Springer-Verlag.

An augmented method for free boundary problems with moving contact lines

An augmented immersed interface method (IIM) is proposed for simulating one-phase moving contact line problems in which a liquid drop spreads or recoils on a solid substrate. While the present two-dimensional mathematical model is a free boundary problem, in our new numerical method, the fluid domain enclosed by the free boundary is embedded into a rectangular one so that the problem can be solved by a regular Cartesian grid method. We introduce an augmented variable along the free boundary so that the stress balancing boundary condition is satisfied. A hybrid time discretization is used in the projection method for better stability. The resultant Helmholtz/Poisson equations with interfaces then are solved by the IIM in an efficient way. Several numerical tests including an accuracy check, and the spreading and recoiling processes of a liquid drop are presented in detail. (C) 2010 Elsevier Ltd. All rights reserved.

Metacomputing on clusters augmented with reconfigurable hardware

Accepted Version

STAR 3 randomized controlled trial to compare sensor-augmented insulin pump therapy with multiple daily injections in the treatment of type 1 diabetes: research design, methods, and baseline characteristics of enrolled subjects.

BACKGROUND: Sensor-augmented pump therapy (SAPT) integrates real-time continuous glucose monitoring (RT-CGM) with continuous subcutaneous insulin infusion (CSII) and offers an alternative to multiple daily injections (MDI). Previous studies provide evidence that SAPT may improve clinical outcomes among people with type 1 diabetes. Sensor-Augmented Pump Therapy for A1c Reduction (STAR) 3 is a multicenter randomized controlled trial comparing the efficacy of SAPT to that of MDI in subjects with type 1 diabetes. METHODS: Subjects were randomized to either continue with MDI or transition to SAPT for 1 year. Subjects in the MDI cohort were allowed to transition to SAPT for 6 months after completion of the study. SAPT subjects who completed the study were also allowed to continue for 6 months. The primary end point was the difference between treatment groups in change in hemoglobin A1c (HbA1c) percentage from baseline to 1 year of treatment. Secondary end points included percentage of subjects with HbA1c < or =7% and without severe hypoglycemia, as well as area under the curve of time spent in normal glycemic ranges. Tertiary end points include percentage of subjects with HbA1c < or =7%, key safety end points, user satisfaction, and responses on standardized assessments. RESULTS: A total of 495 subjects were enrolled, and the baseline characteristics similar between the SAPT and MDI groups. Study completion is anticipated in June 2010. CONCLUSIONS: Results of this randomized controlled trial should help establish whether an integrated RT-CGM and CSII system benefits patients with type 1 diabetes more than MDI.

Accelerating self-consistent field convergence with the augmented Roothaan-Hall energy function.

Based on Pulay's direct inversion iterative subspace (DIIS) approach, we present a method to accelerate self-consistent field (SCF) convergence. In this method, the quadratic augmented Roothaan-Hall (ARH) energy function, proposed recently by Høst and co-workers [J. Chem. Phys. 129, 124106 (2008)], is used as the object of minimization for obtaining the linear coefficients of Fock matrices within DIIS. This differs from the traditional DIIS of Pulay, which uses an object function derived from the commutator of the density and Fock matrices. Our results show that the present algorithm, abbreviated ADIIS, is more robust and efficient than the energy-DIIS (EDIIS) approach. In particular, several examples demonstrate that the combination of ADIIS and DIIS ("ADIIS+DIIS") is highly reliable and efficient in accelerating SCF convergence.

Expression of the counter-regulatory peptide intermedin is augmented in the presence of oxidative stress in hypertrophied cardiomyocytes.

Background: Intermedin (IMD), a novel cardiac peptide related to adrenomedullin (AM), protects against myocardial ischemia-reperfusion injury and attenuates ventricular remodelling. IMD’s actions are mediated by a calcitonin receptor-like receptor in association with receptor activity modifying proteins (RAMPs 1-3). Aim/method: using the spontaneously hypertensive rat (SHR) and normotensive Wistar Kyoto (WKY) rat at 20 weeks of age, to examine (i) the presence of myocardial oxidative stress and concentric hypertrophy; (ii) expression of IMD, AM and receptor components. Results: In left and right ventricular cardiomyocytes from SHR vs. WKY cell width (26% left, 15% right) and mRNA expression of hypertrophic markers ANP (2.7 fold left, 2.7 fold right) and BNP (2.2 fold left, 2.0 fold right) were enhanced. In left ventricular cardiomyocytes only (i) oxidative stress was indicated by increased membrane protein carbonyl content (71%) and augmented production of O2- anion (64%); (ii) IMD (6.8 fold), RAMP1 (2.5 fold) and RAMP3 (2.0 fold) mRNA was increased while AM and RAMP2 mRNA was not altered; (iii) abundance of RAMP1 (by 48%), RAMP2 (by 41%) and RAMP3 (by 90%) monomers in cell membranes was decreased. Conclusion: robust augmentation of IMD expression in hypertrophied left ventricular cardiomyocytes indicates a prominent role for this counter-regulatory peptide in the adaptation of the SHR myocardium to the stresses imposed by chronic hypertension. The local concentration and action of IMD may be further enhanced by down-regulation of NEP within the left ventricle.