Hui and Walter's latent-class model extended to estimate diagnostic test properties from surveillance data: a latent model for latent data

Diagnostic test sensitivity and specificity are probabilistic estimates with far reaching implications for disease control, management and genetic studies. In the absence of 'gold standard' tests, traditional Bayesian latent class models may be used to assess diagnostic test accuracies through the comparison of two or more tests performed on the same groups of individuals. The aim of this study was to extend such models to estimate diagnostic test parameters and true cohort-specific prevalence, using disease surveillance data. The traditional Hui-Walter latent class methodology was extended to allow for features seen in such data, including (i) unrecorded data (i.e. data for a second test available only on a subset of the sampled population) and (ii) cohort-specific sensitivities and specificities. The model was applied with and without the modelling of conditional dependence between tests. The utility of the extended model was demonstrated through application to bovine tuberculosis surveillance data from Northern and the Republic of Ireland. Simulation coupled with re-sampling techniques, demonstrated that the extended model has good predictive power to estimate the diagnostic parameters and true herd-level prevalence from surveillance data. Our methodology can aid in the interpretation of disease surveillance data, and the results can potentially refine disease control strategies.

Contamination of Donana food-chains after the Aznalcollar mine disaster

On 25 April 1998 part of the tailings pond dike of the Aznalcollar Zn mine north of the Guadalquivir marshes (Donana) in southern Spain collapsed releasing an estimated 5 million m³ of acidic metal-rich waste. This event contaminated farmland and wetland up to >40 km downstream, including the 900-ha 'Entremuros', an important area for birds within the Donana world heritage site. In spite of the contamination, birds continued to feed in this area. Samples of two abundant macrophytes (Typha dominguensis and Scirpus maritimus) were taken from the Entremuros and nearby uncontaminated areas; these plants are important food items for several bird species. Analyses showed that in the Entremuros mean plant tissue concentrations of Cd were 3-40-fold (0.8-7.4 ppm) and Zn 20-100-fold (20-3384 ppm) greater than those from control areas. Comparable dietary concentrations of Zn have been reported to cause severe physiological damage to aquatic birds under experimental conditions. Elevated Cd concentrations are of concern as Cd bioconcentrates and is a cumulative poison. Metals released in this accident are moving into this food-chain and present a considerable risk to species feeding on Typha sp. and Scirpus sp. Many other food-webs exist in this area and require detailed examination to identify the species at risk, and to facilitate the management of these risks to minimise future impacts to the wildlife of Donana. Copyright (C) 1999 Elsevier Science Ltd.

Application of disturbance management principles in fresh food supply chains

The desire for more robust supply chains has led to a growth in theoretical and practical interest in the application of both preventive and impact reductive disturbance management principles. This application should ultimately lead to less vulnerable and more competitive supply chains. Based on the extant literature we identify fresh food supply chain's contextual factors (products, processes, supply chain networks and supply chain business environment) and their corresponding characteristics, as well as the main disturbance management principles used. To analyse their influence on the selection and application of disturbance management principles in fresh food supply chains we conducted three case studies. In each case we collected data on the relevant contextual factors, disturbance management principles applied and company background. As an underlying methodology, we first conduct within-case analysis and then expand the analyses to a cross-case context. Based on the findings from these case studies, propositions are built concerning the nature of contextual factors and their characteristics, and their influence on the selection and application of disturbance management principles in fresh food supply chains. Our main findings are related to the identification of contextual characteristics of fresh food supply chains that are either critical vulnerability sources, critical enablers or conditionals and as such require, facilitate or condition selection and application of disturbance management principles.

A Conventional Internal Combustion Engine Versus a Range Extended Electric Vehicle Under the Current Regulatory Regime in the United Kingdom

Transport accounts for 22% of greenhouse gas emissions in the United Kingdom and cars are expected tomore than double by 2050. Car manufacturers are continually aiming for a substantially reduced carbonfootprint through improved fuel efficiency and better powertrain performance due to the strict EuropeanUnion emissions standards. However, road tax, not just fuel efficiency, is a key consideration of consumerswhen purchasing a car. While measures have been taken to reduce emissions through stricter standards, infuture, alternative technologies will be used. Electric vehicles, hybrid vehicles and range extended electricvehicles have been identified as some of these future technologies. In this research a virtual test bed of aconventional internal combustion engine and a range extended electric vehicle family saloon car were builtin AVL’s vehicle and powertrain system level simulation tool, CRUISE, to simulate the New EuropeanDrive Cycle and the results were then soft-linked to a techno-economic model to compare the effectivenessof current support mechanisms over the full life cycle of both cars. The key finding indicates that althoughcarbon emissions are substantially reduced, switching is still not financially the best option for either theconsumer or the government in the long run.

Fish product mislabelling: failings of traceability in the production chain and implications for illegal, unreported and unregulated (IUU) fishing

Increasing consumer demand for seafood, combined with concern over the health of our oceans, has led to many initiatives aimed at tackling destructive fishing practices and promoting the sustainability of fisheries. An important global threat to sustainable fisheries is Illegal, Unreported and Unregulated (IUU) fishing, and there is now an increased emphasis on the use of trade measures to prevent IUU-sourced fish and fish products from entering the international market. Initiatives encompass new legislation in the European Union requiring the inclusion of species names on catch labels throughout the distribution chain. Such certification measures do not, however, guarantee accuracy of species designation. Using two DNA-based methods to compare species descriptions with molecular ID, we examined 386 samples of white fish, or products labelled as primarily containing white fish, from major UK supermarket chains. Species specific real-time PCR probes were used for cod (Gadus morhua) and haddock (Melanogrammus aeglefinus) to provide a highly sensitive and species-specific test for the major species of white fish sold in the UK. Additionally, fish-specific primers were used to sequence the forensically validated barcoding gene, mitochondrial cytochrome oxidase I (COI). Overall levels of congruence between product label and genetic species identification were high, with 94.34% of samples correctly labelled, though a significant proportion in terms of potential volume, were mislabelled. Substitution was usually for a cheaper alternative and, in one case, extended to a tropical species. To our knowledge, this is the first published study encompassing a large-scale assessment of UK retailers, and if representative, indicates a potentially significant incidence of incorrect product designation.

Averaged Extended Tree Augmented Naive Classifier

This work presents a new general purpose classifier named Averaged Extended Tree Augmented Naive Bayes (AETAN), which is based on combining the advantageous characteristics of Extended Tree Augmented Naive Bayes (ETAN) and Averaged One-Dependence Estimator (AODE) classifiers. We describe the main properties of the approach and algorithms for learning it, along with an analysis of its computational time complexity. Empirical results with numerous data sets indicate that the new approach is superior to ETAN and AODE in terms of both zero-one classification accuracy and log loss. It also compares favourably against weighted AODE and hidden Naive Bayes. The learning phase of the new approach is slower than that of its competitors, while the time complexity for the testing phase is similar. Such characteristics suggest that the new classifier is ideal in scenarios where online learning is not required.

Learning Extended Tree Augmented Naive Structures

This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds' algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). We enhance our procedure with a new score function that only takes into account arcs that are relevant to predict the class, as well as an optimization over the equivalent sample size during learning. These ideas may be useful for structure learning of Bayesian networks in general. A range of experiments shows that we obtain models with better prediction accuracy than naive Bayes and TAN, and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator (AODE). We release our implementation of ETAN so that it can be easily installed and run within Weka.

Production of extended-spectrum β-lactamases and the potential indirect pathogenic role of Prevotella isolates from the cystic fibrosis respiratory microbiota

Extended-spectrum β-lactamase (ESBL) production and the prevalence of the β-lactamase-encoding gene blaTEM were determined in Prevotella isolates (n=50) cultured from the respiratory tract of adults and young people with cystic fibrosis (CF). Time-kill studies were used to investigate the concept of passive antibiotic resistance and to ascertain whether a β-lactamase-positive Prevotella isolate can protect a recognised CF pathogen from the action of ceftazidime in vitro. The results indicated that approximately three-quarters (38/50; 76%) of Prevotella isolates produced ESBLs. Isolates positive for ESBL production had higher minimum inhibitory concentrations (MICs) of β-lactam antibiotics compared with isolates negative for production of ESBLs (P<0.001). The blaTEM gene was detected more frequently in CF Prevotella isolates from paediatric patients compared with isolates from adults (P=0.002), with sequence analysis demonstrating that 21/22 (95%) partial blaTEM genes detected were identical to blaTEM-116. Furthermore, a β-lactamase-positive Prevotella isolate protected Pseudomonas aeruginosa from the antimicrobial effects of ceftazidime (P=0.03). Prevotella isolated from the CF respiratory microbiota produce ESBLs and may influence the pathogenesis of chronic lung infection via indirect methods, including shielding recognised pathogens from the action of ceftazidime.