A global algorithm to estimate the expectations of the components of an observed univariate mixture

This paper deals with the unsupervised classification of univariate observations. Given a set of observations originating from a K-component mixture, we focus on the estimation of the component expectations. We propose an algorithm based on the minimization of the “K-product” (KP) criterion we introduced in a previous work. We show that the global minimum of this criterion can be reached by first solving a linear system then calculating the roots of some polynomial of order K. The KP global minimum provides a first raw estimate of the component expectations, then a nearest-neighbour classification enables to refine this estimation. Our method’s relevance is finally illustrated through simulations of various mixtures. When the mixture components do not strongly overlap, the KP algorithm provides better estimates than the Expectation-Maximization algorithm.     

Generalizations of the Nikodym boundedness and Vitali–Hahn–Saks theorems

Generalizations of the Nikodym boundedness and Vitali–Hahn–Saks theorems for scalar-valued measures on rings of sets that are in general not σ-rings are presented. As a consequence, the rings of subsets of N with density zero and uniform density zero are shown to have the Nikodym property. In addition, vector measure generalizations of the Vitali–Hahn–Saks theorem are given.     

Criteria for right disfocality of linear difference equations

Right disfocality and criteria for right disfocality of linear difference equations are considered. The criteria require the positivity of appropriate minors of a Wronskian determinant and are analogues to criteria for the disconjugacy of linear ordinary differential equations and linear difference equations.     

Study on the Binding Characteristic of S-Naproxen Imprinted Polymer and the Interactions between Templates and Monomers

S-naproxen imprinted polymer was prepared with acrylamide as a functional monomer, 1,4-butanediyl diacrylate as a new kind of crossliner, and 2,4,6-trimethylbenzoylphenyl-phosphinic acid ethyl ester as photoinitiator. Their UV and IR spectra were applied to study the interactions between the template and the monomer. Binding experiments showed that the S-naproxen imprinted polymer had better recognition capability for the template than that of a blank polymer. The adsorption distribution coefficient KD of S-naproxen on molecularly-imprinted polymer was 153.2 g/mL, and the separation factor was 1.95 when the initial concentration of substrate was 2.0 mmol/L. Scatchard curves suggested that there were two classes of binding sites in the imprinted polymer and only one class in the blank polymer. Computer simulation using Hyperchem showed the existence of the interactions between the template and the functional monomer as well as the models of the complexes formed by the template and the monomer.     

Implementation of an integrated glucose POCT system

In response to a change of the Belgian National Directives whereby hospital laboratories became responsible for all point-of-care testing (POCT) performed within hospital walls a standardized and automated POC glucose-testing system was implemented in our hospital. The system consists of 50 AccuCheck Inform instruments (Roche Diagnostics, Vilvoorde, Belgium), 50 docking stations, a DataCare Server, and connections to the medical laboratory information system (MOLIS, Sysmex, Barchon, Belgium) and to the hospital information system. Implementation involved many parties and extensive preparation and communication. Key issues were bar-coded patient and user identification, training, and responsibilities. One year after the hospital wide implementation of this system the quality of POC glucose testing has significantly increased, thereby improving patient safety. This study describes a stepwise change over involving the medical laboratory and with a focus on hands-on quality.Presented at the ninth conference on Quality in the Spotlight, 18–19 March 2004, Antwerp, Belgium.     

Metallo‐supramolecular block copolymer micelles: Improved preparation and characterization

Micelles prepared from amphiphilic block copolymers in which a poly(styrene) segment is connected to a poly(ethylene oxide) block via a bis‐(2,2′:6′,2″‐terpyridine‐ruthenium) complex have been intensely studied. In most cases, the micelle populations were found to be strongly heterogeneous in size because of massive micelle/micelle aggregation. In the study reported in this article we tried to improve the homogeneity of the micelle population. The variant preparation procedure developed, which is described here, was used to prepare two “protomer”‐type micelles: PS₂₀‐[Ru]‐PEO₇₀ and PS₂₀‐[Ru]‐PEO₃₇₅. The dropwise addition of water to a solution of the compounds in dimethylformamide was replaced by the controlled addition of water by a syringe pump. The resulting micelles were characterized by sedimentation velocity and sedimentation equilibrium analyses in an analytical ultracentrifuge and by transmission electron microscopy of negatively stained samples. Sedimentation analysis showed virtually unimodal size distributions, in contrast to the findings on micelles prepared previously. PS₂₀‐[Ru]‐PEO₇₀ micelles were found to have an average molar mass of 318,000 g/mol (corresponding to 53 protomers per micelle, which is distinctly less than after micelle preparation by the standard method) and an average hydrodynamic diameter (d_h) of 18 nm. For PS₂₀‐[Ru]‐PEO₃₇₅ micelles, the corresponding values were M = 603,000 g/mol (31 protomers per micelle) and d_h = 34 nm. The latter particles were found to be identical to the “equilibrium” micelles prepared in pure water. Both micelle types had a very narrow molar mass distribution but a much broader distribution of s values and thus of hydrodynamic diameters. This indicates a conformational heterogeneity that is stable on the time scale of sedimentation velocity analysis. The findings from electron microscopy were in disagreement with those from the sedimentation analysis both in average micelle diameter and in the width of the distributions, apparently because of imperfections in the staining procedure. The preparation procedure described also may be useful in micelle formation from other types of protomers. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 4458–4465, 2004     

Effect of the particle size distribution on the low shear viscosity of high‐solid‐content latexes

The production of high‐solid‐content, low‐viscosity latexes is an active field in both industry and academia. The viscosity of polymer dispersions has a clear dependence on the particle size distribution (PSD). An example is the rule of thumb that a bimodal PSD enables the reduction of the viscosity with respect to monomodal systems. Despite important progress in theoretical work, not much has been done to quantitatively predict the low shear viscosity of aqueous polymer dispersions as a function of the complex PSD. In this work, the capability of a low‐shear‐viscosity equation to quantitatively account for the influence of both the PSD and the physicochemical characteristics of the dispersions is experimentally assessed. An analysis, consistent with theoretical concepts, of the data with semiempirical correlations is proposed. Next, with values of the parameters of the viscosity equation obtained experimentally, the effect of a latex with a 70% solid content on the low shear viscosity is examined. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 3936–3946, 2004     

Seizure warning algorithm based on optimization and nonlinear dynamics

There is growing evidence that temporal lobe seizures are preceded by a preictal transition, characterized by a gradual dynamical change from asymptomatic interictal state to seizure. We herein report the first prospective analysis of the online automated algorithm for detecting the preictal transition in ongoing EEG signals. Such, the algorithm constitutes a seizure warning system. The algorithm estimates STL_max, a measure of the order or disorder of the signal, of EEG signals recorded from individual electrode sites. The optimization techniques were employed to select critical brain electrode sites that exhibit the preictal transition for the warning of epileptic seizures. Specifically, a quadratically constrained quadratic 0-1 programming problem is formulated to identify critical electrode sites. The automated seizure warning algorithm was tested in continuous, long-term EEG recordings obtained from 5 patients with temporal lobe epilepsy. For individual patient, we use the first half of seizures to train the parameter settings, which is evaluated by ROC (Receiver Operating Characteristic) curve analysis. With the best parameter setting, the algorithm applied to all cases predicted an average of 91.7% of seizures with an average false prediction rate of 0.196 per hour. These results indicate that it may be possible to develop automated seizure warning devices for diagnostic and therapeutic purposes.Mathematics Subject Classification (1991):20E28, 20G40, 20C20     

Demand variability and incentives in inventory-based contracts

We present some results on the impact of demand variability on stocking policies and incentives to forecast, in the context of single period inventory-based contracts in a manufacturer-retailer channel.