Estimation of generalized entropies with sample spacing

In addition to the well-known Shannon entropy, generalized entropies, such as the Renyi and Tsallis entropies, are increasingly used in many applications. Entropies are computed by means of nonparametric kernel methods that are commonly used to estimate the density function of empirical data. Generalized entropy estimation techniques for one-dimensional data using sample spacings are proposed. By means of computational experiments, it is shown that these techniques are robust and accurate, compare favorably to the popular Parzen window method for estimating entropies, and, in many cases, require fewer computations than Parzen methods.     

A mathematical model for the transmission of Salmonella Typhimurium within a grower-finisher pig herd in Great Britain

In a study of pigs slaughtered at British abattoirs, approximately 23% carried Salmonella in their cecal (large intestine) contents. The most frequent serotype was Salmonella Typhimurium (STM), which was the second most common cause of human salmonellosis in Great Britain. A pig industry-monitoring program was developed to reduce Salmonella infection on British farms. The control of STM infection on the farm requires an understanding of STM transmission dynamics within the herd, and a mathematical model has been developed for an infected grower-finisher farm. The model estimates the probability of a random pig being infected with STM. There are three broad categories of STM infection in pigs: pigs that are infected but unable to transmit the infection (latent); pigs that are infectious, i.e., able to transmit the infection (shedders); and pigs that have stopped shedding but harbor STM in their internal organs (carriers). The model estimates that 21.0% (5th and 95th percentiles, 0.05 to 77.5%) of slaughter-age pigs on an infected farm are likely to be shedding STM. Although this range is wide, it is biologically plausible. Sensitivity analysis of the total number of infected pigs revealed that the most significant input parameters are the probability of effective contact between a specific infectious and susceptible pig and the duration of shedding. The model predicted that 11.5% of pigs would be shedding STM at slaughter age. This value is close to the estimate obtained from a British abattoir survey that 11. 1% of pigs carried STM in their ceca, indicating that the model has reasonable validity.     

A collaborative study to evaluate qualitatively powdered baby food validation samples artificially contaminated with Salmonella anatum

Nineteen laboratories participated in a collaborative study to evaluate the performance of validation samples consisting of powdered infant formula and cereal-based weaning food artificially contaminated with low levels of Salmonella anatum. The Standard method BS EN ISO 6579:2002 was to be followed for the trial. Salmonella counts in each baby food were around 10 CFU/25 g (low) or 10 CFU/g (high level). Trial participants received five samples of each sample type and five blank samples. All samples contained Pseudomonas aeruginosa and Lactobacillus spp. at concentrations between 10(4) and 10(5) CFU/g as background flora. Precision data for the powdered infant formula were similar for the solid selective plating-out medium xylose-lysine-deoxycholate (XLD) and a 2nd choice of agar suitable to isolate Salmonella spp. The sensitivity was 91.3%, accordance 82.5% and concordance 83.9% for the low and 100% for the high level in all cases. For the cereal-based weaning food, the precision data for the high level were similar for XLD and the 2nd choice medium (98.8%, 97.5% and 97.5%). A slight difference was observed for the low level where XLD obtained values of 82.5%, 73.8%, 70.6% for sensitivity, accordance and concordance and the second choice medium values of 81.5%, 72.5% and 69%. The specificity, accordance and concordance of blanks for the infant formula and cereal-based weaning food were 100%.     

Complexation of copper by zwitterionic aminosulfonic (good) buffers

Copper binding properties were investigated for several popular zwitterionic buffers. The two buffers 4-morpholinoethanesulfonic acid (MES) and 3-N-morpholinopropanesulfonic acid (MOPS) did not bind copper and would be good choices for metal speciation studies within their operational pH range. Conversely, 3-(N-morpholino)-2-hydroxypropanesulfonic acid (MOPSO) was observed to weakly bind copper directly (log Kc 2.02). Moreover, strong copper binding was observed for 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 4-(2-hydroxyethyl)piperazine-1-propanesulfonic acid (HEPPS), and N-(2-hydroxyethyl)piperazine-N'-(2-hydroxypropanesulfonic acid) (HEPPSO). Log Kc values range from 7.04 to 7.68 and are indicative of strong copper binding ligands. The latter buffer also exhibited weak binding characteristics with a log Kc of 2.05. The strong Cu binding ligands were present in HEPES, HEPPS, and HEPPSO at much lower concentrations than the total buffer concentration. MES, HEPES, MOPSO, and HEPPSO were analyzed by electrospray-ionization quadrapole time-of-flight mass spectroscopy. The most prominent feature of the spectra for each buffer analyzed was the presence of multiple oligomers, indicating a propensity of interaction between buffer molecules. In addition, the presence of several contaminants was identified in the mass spectrum of the HEPES matrix, including a prominent contaminant (at m/z 131) present in levels similar to those obtained from the modeling of the copper titration data. Other contaminants were found in the other matrixes but were not identified as possible copper binding agents.     

Off‐line monitoring of butyl acrylate and vinyl acetate homopolymerization and copolymerization in toluene

A study of butyl acrylate (BA) and vinyl acetate (VAc) solution homopolymerization and copolymerization in toluene was carried out. The conversion and copolymer composition were monitored using traditional techniques (gravimetry and ¹H‐NMR spectroscopy) and attenuated total reflectance‐Fourier transform IR (ATR‐FTIR) spectroscopy with a diamond‐composite probe and light conduit technology. The peak height of the characteristic absorbances of the monomer(s) during the course of the reaction was used to calculate the conversion and copolymer composition for the ATR‐FTIR monitoring. The data obtained using a ReactIR? 1000 reaction analysis system in the off‐line mode showed very good agreement with data obtained using traditional techniques. The solvent effects on BA and VAc solution homopolymerizations and copolymerizations in toluene were also investigated. Improvement to model predictions was obtained by allowing the lumped constant (k_p/k) to vary with the solvent concentration. Experimental data and model predictions of the number‐ and weight‐average molecular weights for the investigated systems are also presented. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 82: 2958–2977, 2001     

Unified methodology of neural analysis in decision support systems built for pharmaceutical technology

The objective of this study was to create universal methodology of artificial neural networks (ANNs) application in construction of decision support systems designed for various dosage forms. Two different dosage forms (solid dispersions and microemulsions) were modeled with use of the same methodology, software and hardware environments. Completely different models prepared confirmed their generalization ability both for solid dosage forms (solid dispersions) and liquid dosage forms (microemulsions). ME_expert and SD_expert systems basing on the neural expert committees were created. In the pilot study their application allowed for appropriate choice of qualitative and quantitative composition of particular pharmaceutical formulation. It was also proposed that ME_expert and SD_expert might provide in silico formulation procedures. Unified methodology of neural modeling in pharmaceutical technology was confirmed to be effective in providing valuable tools for pharmaceutical product development.     

Superparamagnetic relaxation evidences large surface contribution for the magnetic anisotropy of MnFe<Subscript>2</Subscript>0<Subscript>4</Subscript> nanoparticles of ferrofluids

Manganese ferrite nanoparticles, in the size range 3.3–9.0 nm, are prepared by a hydrothermal coprecipitation process and peptized in aqueous solution. The magnetization curves recorded at room temperature on diluted colloidal sols allow characterizing the distribution of magnetic moment by using a simple Langevin formalism. Mössbauer spectroscopy measurements performed on powder samples at 77 K exhibit a quadrupolar doublet which intensity grows at the expense of the hyperfine sextet pattern as the nanoparticles mean size decreases. The magnetic dynamics behavior is then investigated by measurements of magnetic hysteretic properties at 5 K and temperature dependence of the zero field cooling (ZFC) susceptibility. The values found for the effective anisotropy constant and the dependence of the irreversibility field, inversely proportional to the reference size, clearly indicate that the magnetic anisotropy of our nanoparticles finds its origin on the disordered surface layer.     

Clustering interval data through kernel-induced feature space

Recently, kernel-based clustering in feature space has shown to perform better than conventional clustering methods in unsupervised classification. In this paper, a partitioning clustering method in kernel-induce feature space for symbolic interval-valued data is introduced. The distance between an item and its prototype in feature space is expanded using a two-component mixture kernel to handle intervals. Moreover, tools for the partition and cluster interpretation of interval-valued data in feature space are also presented. To show the effectiveness of the proposed method, experiments with real and synthetic interval data sets were performed and a study comparing the proposed method with different clustering algorithms of the literature is also presented. The clustering quality furnished by the methods is measured by an external cluster validity index (corrected Rand index). These experiments showed the usefulness of the kernel K-means method for interval-valued data and the merit of the partition and cluster interpretation tools.     

Hydroxyalkylation of barbituric acid. III. Product analysis and reaction pathway

Polyetherols containing a thermally stable pyrimidine ring were obtained upon the reaction of hydroxymethyl derivatives of barbituric acid with an excess of ethylene or propylene oxide. The reaction was monitored by ¹H‐NMR and IR spectroscopy for the systems with variable starting molar ratios of reagents. We found that formaldehyde rearranged from N‐hydroxymethyl and oxymethylene bridges into the end of the polyetherol chain during the reaction. Simultaneously, the O‐hydroxymethyl groups underwent blocking by oxirane. The structures of the polyetherols was deduced on the basis of the course of the reaction and the analytical data. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2009