On a resampling approach for tests on the number of clusters with mixture model-based clustering of tissue samples

We consider the problem of assessing the number of clusters in a limited number of tissue samples containing gene expressions for possibly several thousands of genes. It is proposed to use a normal mixture model-based approach to the clustering of the tissue samples. One advantage of this approach is that the question on the number of clusters in the data can be formulated in terms of a test on the smallest number of components in the mixture model compatible with the data. This test can be carried out on the basis of the likelihood ratio test statistic, using resampling to assess its null distribution. The effectiveness of this approach is demonstrated on simulated data and on some microarray datasets, as considered previously in the bioinformatics literature. (C) 2004 Elsevier Inc. All rights reserved.

Speeding up the EM algorithm for mixture model-based segmentation of magnetic resonance images

Mixture models implemented via the expectation-maximization (EM) algorithm are being increasingly used in a wide range of problems in pattern recognition such as image segmentation. However, the EM algorithm requires considerable computational time in its application to huge data sets such as a three-dimensional magnetic resonance (MR) image of over 10 million voxels. Recently, it was shown that a sparse, incremental version of the EM algorithm could improve its rate of convergence. In this paper, we show how this modified EM algorithm can be speeded up further by adopting a multiresolution kd-tree structure in performing the E-step. The proposed algorithm outperforms some other variants of the EM algorithm for segmenting MR images of the human brain. (C) 2004 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

Angiotensin II facilitates autoregulation in the perfused mouse kidney: An optimized in vitro model for assessment of renal vascular and tubular function

Background and Aims: We have optimized the isolated perfused mouse kidney (IPMK) model for studying renal vascular and tubular function in vitro using 24-28 g C57BL6J mice; the wild type controls for many transgenic mice. Methods and Results: Buffer composition was optimized for bovine serum albumin concentration (BSA). The effect of adding erythrocytes on renal function and morphology was assessed. Autoregulation was investigated during stepped increases in perfusion pressure. Perfusion for 60 min at 90-110 mmHg with Krebs bicarbonate buffer containing 5.5% BSA, and amino acids produced functional parameters within the in vivo range. Erythrocytes increased renal vascular resistance (3.8 +/- 0.2 vs 2.4 +/- 0.1 mL/min.mmHg, P < 0.05), enhanced sodium reabsorption (FENa = 0.3 +/- 0.08 vs 1.5 +/- 0.7%, P < 0.05), produced equivalent glomerular filtration rates (GFR; 364 +/- 38 vs 400 +/- 9 muL/min per gkw) and reduced distal tubular cell injury in the inner stripe (5.8 +/- 1.7 vs 23.7 +/- 3.1%, P < 0.001) compared to cell free perfusion. The IPMK was responsive to vasoconstrictor (angiotensin II, EC50 100 pM) and vasodilator (methacholine, EC50 75 nM) mediators and showed partial autoregulation of perfusate flow under control conditions over 65-85 mmHg; autoregulatory index (ARI) of 0.66 +/- 0.11. Angiotensin II (100 pM) extended this range (to 65-120 mmHg) and enhanced efficiency (ARI 0.21 +/- 0.02, P < 0.05). Angiotensin II facilitation was antagonized by methacholine (ARI 0.76 +/- 0.08) and papaverine (ARI 0.91 +/- 0.13). Conclusion: The IPMK model is useful for studying renal physiology and pathophysiology without systemic neurohormonal influences.

An FDTD model for calculation of gradient-induced eddy currents in MRI system

In most magnetic resonance imaging (MRI) systems, pulsed magnetic gradient fields induce eddy currents in the conducting structures of the superconducting magnet. The eddy currents induced in structures within the cryostat are particularly problematic as they are characterized by long time constants by virtue of the low resistivity of the conductors. This paper presents a three-dimensional (3-D) finite-difference time-domain (FDTD) scheme in cylindrical coordinates for eddy-current calculation in conductors. This model is intended to be part of a complete FDTD model of an MRI system including all RF and low-frequency field generating units and electrical models of the patient. The singularity apparent in the governing equations is removed by using a series expansion method and the conductor-air boundary condition is handled using a variant of the surface impedance concept. The numerical difficulty due to the asymmetry of Maxwell equations for low-frequency eddy-current problems is circumvented by taking advantage of the known penetration behavior of the eddy-current fields. A perfectly matched layer absorbing boundary condition in 3-D cylindrical coordinates is also incorporated. The numerical method has been verified against analytical solutions for simple cases. Finally, the algorithm is illustrated by modeling a pulsed field gradient coil system within an MRI magnet system. The results demonstrate that the proposed FDTD scheme can be used to calculate large-scale eddy-current problems in materials with high conductivity at low frequencies.

The use of simultaneous H-1 & P-31 magic angle spinning nuclear magnetic resonance measurements to characterize energy metabolism during the conversion of muscle to meat

In order to investigate the potential of magic angle spinning nuclear magnetic resonance (MAS NMR) in the elucidation of post-mortem metabolism in muscle biopsies, simultaneous H-1 and (31)p MAS NMR measurements were made continuously on postmortem (20 min to 24 h) muscle longissimus samples from rabbits. The animals had either been or not been given adrenaline (0.5 mg kg(-1) 4 h pre-slaughter) to deplete stores of muscle glycogen. The intracellular pH was calculated from H-1 spectra, and the post-mortem rate of formation of lactate was followed and quantified. Comparison of measurements made on muscle samples from rabbits treated with adrenaline with measurements made on muscle samples from untreated' rabbits revealed significant effects of adrenaline treatment on both pH (pH24 h = 6.42 vs. pH24 It = 5.60) and formation of lactate (16 mmol g(-1) vs. 65 mmol g(-1)). The P-31 NMR spectra were used to follow the rate of degradation of ATP and phosphocreatine. The present study clearly shows that MAS NMR has potential for the study of post-mortem energy metabolism.

A Bethe ansatz solvable model for superpositions of Cooper pairs and condensed molecular bosons

We introduce a general Hamiltonian describing coherent superpositions of Cooper pairs and condensed molecular bosons. For particular choices of the coupling parameters, the model is integrable. One integrable manifold, as well as the Bethe ansatz solution, was found by Dukelsky et al. [J. Dukelsky, G.G. Dussel, C. Esebbag, S. Pittel, Phys. Rev. Lett. 93 (2004) 050403]. Here we show that there is a second integrable manifold, established using the boundary quantum inverse scattering method. In this manner we obtain the exact solution by means of the algebraic Bethe ansatz. In the case where the Cooper pair energies are degenerate we examine the relationship between the spectrum of these integrable Hamiltonians and the quasi-exactly solvable spectrum of particular Schrodinger operators. For the solution we derive here the potential of the Schrodinger operator is given in terms of hyperbolic functions. For the solution derived by Dukelsky et al., loc. cit. the potential is sextic and the wavefunctions obey PT-symmetric boundary conditions. This latter case provides a novel example of an integrable Hermitian Hamiltonian acting on a Fock space whose states map into a Hilbert space of PE-symmetric wavefunctions defined on a contour in the complex plane. (c) 2006 Elsevier B.V. All rights reserved.

Model driven distribution of collaborative business processes

In this paper, we present a top down approach for integrated process modelling and distributed process execution. The integrated process model can be utilized for global monitoring and visualization and distributed process models for local execution. Our main focus in this paper is the presentation of the approach to support automatic generation and linking of distributed process models from an integrated process definition.