Maximum Likelihood Estimation of Mixture Densities for Binned and Truncated Multivariate Data

Binning and truncation of data are common in data analysis and machine learning. This paper addresses the problem of fitting mixture densities to multivariate binned and truncated data. The EM approach proposed by McLachlan and Jones (Biometrics, 44: 2, 571-578, 1988) for the univariate case is generalized to multivariate measurements. The multivariate solution requires the evaluation of multidimensional integrals over each bin at each iteration of the EM procedure. Naive implementation of the procedure can lead to computationally inefficient results. To reduce the computational cost a number of straightforward numerical techniques are proposed. Results on simulated data indicate that the proposed methods can achieve significant computational gains with no loss in the accuracy of the final parameter estimates. Furthermore, experimental results suggest that with a sufficient number of bins and data points it is possible to estimate the true underlying density almost as well as if the data were not binned. The paper concludes with a brief description of an application of this approach to diagnosis of iron deficiency anemia, in the context of binned and truncated bivariate measurements of volume and hemoglobin concentration from an individual's red blood cells.

A mixture model-based approach to the clustering of microarray expression data

Motivation: This paper introduces the software EMMIX-GENE that has been developed for the specific purpose of a model-based approach to the clustering of microarray expression data, in particular, of tissue samples on a very large number of genes. The latter is a nonstandard problem in parametric cluster analysis because the dimension of the feature space (the number of genes) is typically much greater than the number of tissues. A feasible approach is provided by first selecting a subset of the genes relevant for the clustering of the tissue samples by fitting mixtures of t distributions to rank the genes in order of increasing size of the likelihood ratio statistic for the test of one versus two components in the mixture model. The imposition of a threshold on the likelihood ratio statistic used in conjunction with a threshold on the size of a cluster allows the selection of a relevant set of genes. However, even this reduced set of genes will usually be too large for a normal mixture model to be fitted directly to the tissues, and so the use of mixtures of factor analyzers is exploited to reduce effectively the dimension of the feature space of genes. Results: The usefulness of the EMMIX-GENE approach for the clustering of tissue samples is demonstrated on two well-known data sets on colon and leukaemia tissues. For both data sets, relevant subsets of the genes are able to be selected that reveal interesting clusterings of the tissues that are either consistent with the external classification of the tissues or with background and biological knowledge of these sets.

Quantum and classical fidelities for Gaussian states

We examine the physical significance of fidelity as a measure of similarity for Gaussian states by drawing a comparison with its classical counterpart. We find that the relationship between these classical and quantum fidelities is not straightforward, and in general does not seem to provide insight into the physical significance of quantum fidelity. To avoid this ambiguity we propose that the efficacy of quantum information protocols be characterized by determining their transfer function and then calculating the fidelity achievable for a hypothetical pure reference input state. (c) 2007 Optical Society of America.

Long-run growth effects of taxation in a non-scale growth model with innovation

In previous studies, taxing income or consumption hinders long-run growth. Incorporating saving and leisure into the non-scale Schumpeterian model of [Journal of Political Economy 107 (1999) 715-730], we show that the usual growth effects of taxing consumption and labor income do not exist. (C) 2002 Elsevier Science B.V. All rights reserved.

Technical, allocative, cost and scale efficiencies in Bangladesh rice cultivation: A non-parametric approach

Applying programming techniques to detailed data for 406 rice farms in 21 villages, for 1997, produces inefficiency measures, which differ substantially from the results of simple yield and unit cost measures. For the Boro (dry) season, mean technical efficiency was efficiency was 56.2 per cent and 69.4 per cent, allocative efficiency was 81.3 per cent, cost efficiency was 56.2 per cent and scale efficiency 94.9 per cent. The Aman (wet) season results are similar, but a few points lower. Allocative inefficiency is due to overuse of labour, suggesting population pressure, and of fertiliser, where recommended rates may warrant revision. Second-stage regressions show that large families are more inefficient, whereas farmers with better access to input markets, and those who do less off-farm work, tend to be more efficient. The information on the sources of inter-farm performance differentials could be used by the extension agents to help inefficient farmers. There is little excuse for such sub-optimal use of survey data, which are often collected at substantial costs.

Translation of the leisure satisfaction scale into French: A validity study

Presents a study which described the process of translating an English standardized assessment into another language. Details of the study design; Translation of the Leisure Satisfaction Scale (LSS) into French using the translation/validation methodologies; Correlations between both language versions of LSS.

Glycogen-accumulating organisms in laboratory-scale and full scale wastewater treatment processes

Laboratory-scale sequencing batch reactors (SBRs) as models for wastewater treatment processes were used to identify glycogen-accumulating organisms (GAOs), which are thought to be responsible for the deterioration of enhanced biological phosphorus removal (EBPR). The SBRs (called Q and T), operated under alternating anaerobic-aerobic conditions typical for EBPR, generated mixed microbial communities (sludges) demonstrating the GAO phenotype. Intracellular glycogen and poly-beta-hydroxyalkanoate (PHA) transformations typical of efficient EBPR occurred but polyphosphate was not bioaccumulated and the sludges contained 1.8% P (sludge Q) and 1.5% P (sludge T). 16S rDNA clone libraries were prepared from DNA extracted from the Q and T sludges. Clone inserts were grouped into operational taxonomic units (OTUs) by restriction fragment length polymorphism banding profiles. OTU representatives were sequenced and phylogenetically analysed. The Q sludge library comprised four OTUs and all six determined sequences were 99.7% identical, forming a cluster in the gamma-Proteobacteria radiation. The T sludge library comprised eight OTUs and the majority of clones were Acidobacteria subphylum 4 (49% of the library) and candidate phylum OPU (39% of the library). One OTU (two clones, of which one was sequenced) was in the gamma-Proteobacteria radiation with 95% sequence identity to the Q sludge clones. Oligonucleotide probes (called GAOQ431 and GAOQ989) were designed from the gamma-Proteobacteria clone sequences for use in fluorescence in situ hybridization (FISH); 92 % of the Q sludge bacteria and 28 % of the T sludge bacteria bound these probes in FISH. FISH and post-FISH chemical staining for PHA were used to determine that bacteria from a novel gamma-Proteobacteria cluster were phenotypically GAOs in one laboratory-scale SBR and two fullscale wastewater treatment plants. It is suggested that the GAOs from the novel cluster in the gamma-Proteobacteria radiation be named 'Candidatus Competibacter phosphatis'.

Combined hydraulic and biological modelling and full-scale validation of SBR process

The biological reactions during the settling and decant periods of Sequencing Batch Reactors (SBRs) are generally ignored as they are not easily measured or described by modelling approaches. However, important processes are taking place, and in particular when the influent is fed into the bottom of the reactor at the same time (one of the main features of the UniFed process), the inclusion of these stages is crucial for accurate process predictions. Due to the vertical stratification of both liquid and solid components, a one-dimensional hydraulic model is combined with a modified ASM2d biological model to allow the prediction of settling velocity, sludge concentration, soluble components and biological processes during the non-mixed periods of the SBR. The model is calibrated on a full-scale UniFed SBR system with tracer breakthrough tests, depth profiles of particulate and soluble compounds and measurements of the key components during the mixed aerobic period. This model is then validated against results from an independent experimental period with considerably different operating parameters. In both cases, the model is able to accurately predict the stratification and most of the biological reactions occurring in the sludge blanket and the supernatant during the non-mixed periods. Together with a correct description of the mixed aerobic period, a good prediction of the overall SBR performance can be achieved.

Monitoring the composition and form of urban environments based on the vegetation-impervious surface-soil (VIS) model by sub-pixel analysis techniques

The majority of the world's population now resides in urban environments and information on the internal composition and dynamics of these environments is essential to enable preservation of certain standards of living. Remotely sensed data, especially the global coverage of moderate spatial resolution satellites such as Landsat, Indian Resource Satellite and Systeme Pour I'Observation de la Terre (SPOT), offer a highly useful data source for mapping the composition of these cities and examining their changes over time. The utility and range of applications for remotely sensed data in urban environments could be improved with a more appropriate conceptual model relating urban environments to the sampling resolutions of imaging sensors and processing routines. Hence, the aim of this work was to take the Vegetation-Impervious surface-Soil (VIS) model of urban composition and match it with the most appropriate image processing methodology to deliver information on VIS composition for urban environments. Several approaches were evaluated for mapping the urban composition of Brisbane city (south-cast Queensland, Australia) using Landsat 5 Thematic Mapper data and 1:5000 aerial photographs. The methods evaluated were: image classification; interpretation of aerial photographs; and constrained linear mixture analysis. Over 900 reference sample points on four transects were extracted from the aerial photographs and used as a basis to check output of the classification and mixture analysis. Distinctive zonations of VIS related to urban composition were found in the per-pixel classification and aggregated air-photo interpretation; however, significant spectral confusion also resulted between classes. In contrast, the VIS fraction images produced from the mixture analysis enabled distinctive densities of commercial, industrial and residential zones within the city to be clearly defined, based on their relative amount of vegetation cover. The soil fraction image served as an index for areas being (re)developed. The logical match of a low (L)-resolution, spectral mixture analysis approach with the moderate spatial resolution image data, ensured the processing model matched the spectrally heterogeneous nature of the urban environments at the scale of Landsat Thematic Mapper data.

Clinical evaluation of the WOMAC 3.0 OA Index in numeric rating scale format using a computerized touch screen version

Background: The Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index is a previously described self-administered questionnaire covering three domains: pain, stiffness and function. It has been validated in patients with osteoarthritis (OA) of the hip or knee in a paper-based format. Aim: To validate the WOMAC 3.0 using a numerical rating scale in a computerized touch screen format allowing immediate evaluation of the questionnaire. In the computed version cartoons, written and audio instruments were included in order facilitate application. Methods: Fifty patients, demographically balanced, with radiographically proven primary hip or knee OA completed the classical paper and the new computerized WOMAC version. Subjects were randomized either to paper format or computerized format first to balance possible order effects, Results: The intra-class correlation coefficients for pain, stiffness and function values were 0.915, 0.745 and 0.940, respectively. The Spearman correlation coefficients for pain, stiffness and function were 0.88, 0.77 and 0.87, respectively. Conclusion: These data indicate that the computerized WOMAC OA index 3.0 is comparable to the paper WOMAC in all three dimensions. The computerized version would allow physicians to get an immediate result and if present a direct comparison with a previous exam. (C) 2002 OsteoArthritis Research Society International. Published by Elsevier Science Ltd. All rights reserved.

Scale-up of mixer granulators for effective liquid distribution

There is considerable anecdotal evidence from industry that poor wetting and liquid distribution can lead to broad granule size distributions in mixer granulators. Current scale-up scenarios lead to poor liquid distribution and a wider product size distribution. There are two issues to consider when scaling up: the size and nature of the spray zone and the powder flow patterns as a function of granulator scale. Short, nucleation-only experiments in a 25L PMA Fielder mixer using lactose powder with water and HPC solutions demonstrated the existence of different nucleation regimes depending on the spray flux Psi(a)-from drop-controlled nucleation to caking. In the drop-controlled regime at low Psi(a) values. each drop forms a single nucleus and the nuclei distribution is controlled by the spray droplet size distribution. As Psi(a) increases, the distribution broadens rapidly as the droplets overlap and coalesce in the spray zone. The results are in excellent agreement with previous experiments and confirm that for drop-controlled nucleation. Psi(a) should be less than 0.1. Granulator flow studies showed that there are two powder flow regimes-bumping and roping. The powder flow goes through a transition from bumping to roping as impeller speed is increased. The roping regime gives good bed turn over and stable flow patterns. This regime is recommended for good liquid distribution and nucleation. Powder surface velocities as a function of impeller speed were measured using high-speed video equipment and MetaMorph image analysis software, Powder surface velocities were 0.2 to 1 ms(-1)-an order of magnitude lower than the impeller tip speed. Assuming geometrically similar granulators, impeller speed should be set to maintain constant Froude number during scale-up rather than constant tip speed to ensure operation in the roping regime. (C) 2002 Published by Elsevier Science B.V.