Discriminative classifiers for speaker recognition

Speaker Recognition, Speaker Verification, Sparse Kernel Logistic Regression, Support Vector Machine

First- Versus Second-Generation Drug-Eluting Stents in Acute Coronary Syndromes (Katowice-Zabrze Registry)

Abstract Background: There are sparse data on the performance of different types of drug-eluting stents (DES) in acute and real-life setting. Objective: The aim of the study was to compare the safety and efficacy of first- versus second-generation DES in patients with acute coronary syndromes (ACS). Methods: This all-comer registry enrolled consecutive patients diagnosed with ACS and treated with percutaneous coronary intervention with the implantation of first- or second-generation DES in one-year follow-up. The primary efficacy endpoint was defined as major adverse cardiac and cerebrovascular event (MACCE), a composite of all-cause death, nonfatal myocardial infarction, target-vessel revascularization and stroke. The primary safety outcome was definite stent thrombosis (ST) at one year. Results: From the total of 1916 patients enrolled into the registry, 1328 patients were diagnosed with ACS. Of them, 426 were treated with first- and 902 with second-generation DES. There was no significant difference in the incidence of MACCE between two types of DES at one year. The rate of acute and subacute ST was higher in first- vs. second-generation DES (1.6% vs. 0.1%, p < 0.001, and 1.2% vs. 0.2%, p = 0.025, respectively), but there was no difference regarding late ST (0.7% vs. 0.2%, respectively, p = 0.18) and gastrointestinal bleeding (2.1% vs. 1.1%, p = 0.21). In Cox regression, first-generation DES was an independent predictor for cumulative ST (HR 3.29 [1.30-8.31], p = 0.01). Conclusions: In an all-comer registry of ACS, the one-year rate of MACCE was comparable in groups treated with first- and second-generation DES. The use of first-generation DES was associated with higher rates of acute and subacute ST and was an independent predictor of cumulative ST.

Morphological characterization and sex-related differences of the mandible of the armadillos Chaetophractus vellerosus and Zaedyus pichiy (Xenarthra, Dasypodidae), with consideration of dietary aspects

The morphological characteristics of the mandible of adult Chaetophractus vellerosus (Gray, 1865) and Zaedyus pichiy (Desmarest, 1804) were studied to establish its generalized design and to identify inter- and intra- (sexual) specific differences. Morphological descriptions were complemented with the application of univariate and multivariate (analysis of correlation matrices, PCA, discriminant analysis) techniques. The mandible of both species is very similar, and is characterized by elevated condyle, well developed angular process, distinct coronoid process, tooth row which extends to the rear end of the angle between body and ramus, and unfused but firm symphysis. Although both armadillos are omnivorous, a more slender configuration of the jaw in Z. pichiy could be indicative of a better adaptation of its masticatory apparatus to insectivory. The PCA showed an almost total segregation of both species on PC1 (47.7% of the total variance), with C. vellerosus being associated to mandibles taller and with wider body and ramus. Zaedyus pichiy was characterized by heavy loadings of length parameters on PC2 (22.6% of the variance). A small degree of sexual dimorphism was found, with size-based differences in C. vellerosus (larger mandibles in females) and shape-based differences in Z. pichiy (taller mandibles in males, longer ones in females). Correlations between variables were higher in males of both species, indicating a more stable shape of the mandible than in females. The selected parameters to discriminate sexes were the body length of the mandible in C. vellerosus (correct classification: ca. 86% in males, 81% in females), and the height of the mandible at the level of the last tooth in Z. pichiy (near 85% of right assignment in both sexes). The inclusion of a new variable (body length) in the latter species improved the classification of the females to 100%. Teeth are typically 10 in C. vellerosus and 9 in Z. pichiy, but aberrancies in this basic number, such as unilateral or bilateral extra or fewer teeth, are common.

Sensitivity of simulation results to competing SAM updates

Recently there has been a renewed research interest in the properties of non survey updates of input-output tables and social accounting matrices (SAM). Along with the venerable and well known scaling RAS method, several alternative new procedures related to entropy minimization and other metrics have been suggested, tested and used in the literature. Whether these procedures will eventually substitute or merely complement the RAS approach is still an open question without a definite answer. The performance of many of the updating procedures has been tested using some kind of proximity or closeness measure to a reference input-output table or SAM. The first goal of this paper, in contrast, is the proposal of checking the operational performance of updating mechanisms by way of comparing the simulation results that ensue from adopting alternative databases for calibration of a reference applied general equilibrium model. The second goal is to introduce a new updatin! g procedure based on information retrieval principles. This new procedure is then compared as far as performance is concerned to two well-known updating approaches: RAS and cross-entropy. The rationale for the suggested cross validation is that the driving force for having more up to date databases is to be able to conduct more current, and hopefully more credible, policy analyses.

A Multisectorial model of prices: the SAM approach

Social Accounting Matrices (SAM) are normally used to analyse the income generation process. They are also useful, however, for analysing the cost transmission and price formation mechanisms. For price contributions, Roland-Holst and Sancho (1995) used the SAM structure to analyse the price and cost linkages through a representation of the interdependence between activities, households and factors. This paper is a further analysis of the cost transmission mechanisms, in which I add the capital account to the endogenous components of the Roland-Holst and Sancho approach. By doing this I reflect the responses of prices to the exogenous shocks in savings and investment. I also present an additive decomposition of the global price effects into categories of interdependence that isolates the impact on price levels of shocks in the capital account. I use a 1994 Social Accounting Matrix to make an empirical application of the Catalan economy. Keywords: social accounting matrix, cost linkages, price transmission, capital account. JEL Classification: C63, C69, D59.

Generalized factor models: a bayesian approach

There is recent interest in the generalization of classical factor models in which the idiosyncratic factors are assumed to be orthogonal and there are identification restrictions on cross-sectional and time dimensions. In this study, we describe and implement a Bayesian approach to generalized factor models. A flexible framework is developed to determine the variations attributed to common and idiosyncratic factors. We also propose a unique methodology to select the (generalized) factor model that best fits a given set of data. Applying the proposed methodology to the simulated data and the foreign exchange rate data, we provide a comparative analysis between the classical and generalized factor models. We find that when there is a shift from classical to generalized, there are significant changes in the estimates of the structures of the covariance and correlation matrices while there are less dramatic changes in the estimates of the factor loadings and the variation attributed to common factors.

Invariants de matrius Hadamard en MAGMA

L'objectiu d'aquest projecte ha estat generalitzar i integrar la funcionalitat de dos projectes anteriors que ampliaven el tractament que oferia el Magma respecte a les matrius de Hadamard. Hem implementat funcions genèriques que permeten construir noves matrius Hadamard de qualsevol mida per a cada rang i dimensió de nucli, i així ampliar la seva base de dades. També hem optimitzat la funció que calcula el nucli, i hem desenvolupat funcions que calculen la invariant Symmetric Hamming Distance Enumerator (SH-DE) proposada per Kai-Tai Fang i Gennian Gei que és més sensible per a la detecció de la no equivalència de les matrius Hadamard.

Effects of selection and drift on G matrix evolution in a heterogeneous environment: a multivariate Qst-Fst Test with the freshwater snail Galba truncatula.

Unraveling the effect of selection vs. drift on the evolution of quantitative traits is commonly achieved by one of two methods. Either one contrasts population differentiation estimates for genetic markers and quantitative traits (the Q(st)-F(st) contrast) or multivariate methods are used to study the covariance between sets of traits. In particular, many studies have focused on the genetic variance-covariance matrix (the G matrix). However, both drift and selection can cause changes in G. To understand their joint effects, we recently combined the two methods into a single test (accompanying article by Martin et al.), which we apply here to a network of 16 natural populations of the freshwater snail Galba truncatula. Using this new neutrality test, extended to hierarchical population structures, we studied the multivariate equivalent of the Q(st)-F(st) contrast for several life-history traits of G. truncatula. We found strong evidence of selection acting on multivariate phenotypes. Selection was homogeneous among populations within each habitat and heterogeneous between habitats. We found that the G matrices were relatively stable within each habitat, with proportionality between the among-populations (D) and the within-populations (G) covariance matrices. The effect of habitat heterogeneity is to break this proportionality because of selection for habitat-dependent optima. Individual-based simulations mimicking our empirical system confirmed that these patterns are expected under the selective regime inferred. We show that homogenizing selection can mimic some effect of drift on the G matrix (G and D almost proportional), but that incorporating information from molecular markers (multivariate Q(st)-F(st)) allows disentangling the two effects.

Coffee and tea intake and risk of head and neck cancer: pooled analysis in the International Head and Neck Cancer Epidemiology Consortium.

BACKGROUND: Only a few studies have explored the relation between coffee and tea intake and head and neck cancers, with inconsistent results. METHODS: We pooled individual-level data from nine case-control studies of head and neck cancers, including 5,139 cases and 9,028 controls. Logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (95% CI), adjusting for potential confounders. RESULTS: Caffeinated coffee intake was inversely related with the risk of cancer of the oral cavity and pharynx: the ORs were 0.96 (95% CI, 0.94-0.98) for an increment of 1 cup per day and 0.61 (95% CI, 0.47-0.80) in drinkers of >4 cups per day versus nondrinkers. This latter estimate was consistent for different anatomic sites (OR, 0.46; 95% CI, 0.30-0.71 for oral cavity; OR, 0.58; 95% CI, 0.41-0.82 for oropharynx/hypopharynx; and OR, 0.61; 95% CI, 0.37-1.01 for oral cavity/pharynx not otherwise specified) and across strata of selected covariates. No association of caffeinated coffee drinking was found with laryngeal cancer (OR, 0.96; 95% CI, 0.64-1.45 in drinkers of >4 cups per day versus nondrinkers). Data on decaffeinated coffee were too sparse for detailed analysis, but indicated no increased risk. Tea intake was not associated with head and neck cancer risk (OR, 0.99; 95% CI, 0.89-1.11 for drinkers versus nondrinkers). CONCLUSIONS: This pooled analysis of case-control studies supports the hypothesis of an inverse association between caffeinated coffee drinking and risk of cancer of the oral cavity and pharynx. IMPACT: Given widespread use of coffee and the relatively high incidence and low survival of head and neck cancers, the observed inverse association may have appreciable public health relevance.

Evaluation of cumulative PCB exposure estimated by a job exposure matrix versus PCB serum concentrations.

Although polychlorinated biphenyls (PCBs) have been banned in many countries for more than three decades, exposures to PCBs continue to be of concern due to their long half-lives and carcinogenic effects. In National Institute for Occupational Safety and Health studies, we are using semiquantitative plant-specific job exposure matrices (JEMs) to estimate historical PCB exposures for workers (n = 24,865) exposed to PCBs from 1938 to 1978 at three capacitor manufacturing plants. A subcohort of these workers (n = 410) employed in two of these plants had serum PCB concentrations measured at up to four times between 1976 and 1989. Our objectives were to evaluate the strength of association between an individual worker's measured serum PCB levels and the same worker's cumulative exposure estimated through 1977 with the (1) JEM and (2) duration of employment, and to calculate the explained variance the JEM provides for serum PCB levels using (3) simple linear regression. Consistent strong and statistically significant associations were observed between the cumulative exposures estimated with the JEM and serum PCB concentrations for all years. The strength of association between duration of employment and serum PCBs was good for highly chlorinated (Aroclor 1254/HPCB) but not less chlorinated (Aroclor 1242/LPCB) PCBs. In the simple regression models, cumulative occupational exposure estimated using the JEMs explained 14-24% of the variance of the Aroclor 1242/LPCB and 22-39% for Aroclor 1254/HPCB serum concentrations. We regard the cumulative exposure estimated with the JEM as a better estimate of PCB body burdens than serum concentrations quantified as Aroclor 1242/LPCB and Aroclor 1254/HPCB.

A comparison of results from MRIO and interregional computable general equilibrium (CGE) analyses of the impacts of a positive demand shock on the ‘CO2 trade balance’ between Scotland and the rest of the UK

In previous work we have applied the environmental multi-region input-output (MRIO) method proposed by Turner et al (2007) to examine the ‘CO2 trade balance’ between Scotland and the Rest of the UK. In McGregor et al (2008) we construct an interregional economy-environment input-output (IO) and social accounting matrix (SAM) framework that allows us to investigate methods of attributing responsibility for pollution generation in the UK at the regional level. This facilitates analysis of the nature and significance of environmental spillovers and the existence of an environmental ‘trade balance’ between regions. While the existence of significant data problems mean that the quantitative results of this study should be regarded as provisional, we argue that the use of such a framework allows us to begin to consider questions such as the extent to which a devolved authority like the Scottish Parliament can and should be responsible for contributing to national targets for reductions in emissions levels (e.g. the UK commitment to the Kyoto Protocol) when it is limited in the way it can control emissions, particularly with respect to changes in demand elsewhere in the UK. However, while such analysis is useful in terms of accounting for pollution flows in the single time period that the accounts relate to, it is limited when the focus is on modelling the impacts of any marginal change in activity. This is because a conventional demand-driven IO model assumes an entirely passive supply-side in the economy (i.e. all supply is infinitely elastic) and is further restricted by the assumption of universal Leontief (fixed proportions) technology implied by the use of the A and multiplier matrices. In this paper we argue that where analysis of marginal changes in activity is required, a more flexible interregional computable general equilibrium approach that models behavioural relationships in a more realistic and theory-consistent manner, is more appropriate and informative. To illustrate our analysis, we compare the results of introducing a positive demand stimulus in the UK economy using both IO and CGE interregional models of Scotland and the rest of the UK. In the case of the latter, we demonstrate how more theory consistent modelling of both demand and supply side behaviour at the regional and national levels affect model results, including the impact on the interregional CO2 ‘trade balance’.

The added value from adopting a CGE approach to analyse changes in environmental trade balances

The application of multi-region environmental input-output (IO) analysis to the problem of accounting for emissions generation (and/or resource use) under different accounting principles has become increasingly common in the ecological and environmental economics literature in particular, with applications at the international and interregional subnational level. However, while environmental IO analysis is invaluable in accounting for pollution flows in the single time period that the accounts relate to, it is limited when the focus is on modelling the impacts of any marginal change in activity. This is because a conventional demand-driven IO model assumes an entirely passive supply-side in the economy (i.e. all supply is infinitely elastic) and is further restricted by the assumption of universal Leontief (fixed proportions) technology implied by the use of the A and multiplier matrices. Where analysis of marginal changes in activity is required, extension from an IO accounting framework to a more flexible interregional computable general equilibrium (CGE) approach, where behavioural relationships can be modelled in a more realistic and theory-consistent manner, is appropriate. Our argument is illustrated by comparing the results of introducing a positive demand stimulus in the UK economy using IO and CGE interregional models of Scotland and the rest of the UK. In the case of the latter, we demonstrate how more theory consistent modelling of both demand and supply side behaviour at the regional and national levels effect model results, including the impact on the interregional CO2 ‘trade balance’.

Inflammatory markers and blood pressure: sex differences and the effect of fat mass in the CoLaus Study.

Several studies have reported high levels of inflammatory biomarkers in hypertension, but data coming from the general population are sparse, and sex differences have been little explored. The CoLaus Study is a cross-sectional examination survey in a random sample of 6067 Caucasians aged 35-75 years in Lausanne, Switzerland. Blood pressure (BP) was assessed using a validated oscillometric device. Anthropometric parameters were also measured, including body composition, using electrical bioimpedance. Crude serum levels of interleukin-6 (IL-6), tumor necrosis factor α (TNF-α) and ultrasensitive C-reactive protein (hsCRP) were positively and IL-1β (IL-1β) negatively (P<0.001 for all values), associated with BP. For IL-6, IL-1β and TNF-α, the association disappeared in multivariable analysis, largely explained by differences in age and body mass index, in particular fat mass. On the contrary, hsCRP remained independently and positively associated with systolic (β (95% confidence interval): 1.15 (0.64; 1.65); P<0.001) and diastolic (0.75 (0.42; 1.08); P<0.001) BP. Relationships of hsCRP, IL-6 and TNF-α with BP tended to be stronger in women than in men, partly related to the difference in fat mass, yet the interaction between sex and IL-6 persisted after correction for all tested confounders. In the general population, the associations between inflammatory biomarkers and rising levels of BP are mainly driven by age and fat mass. The stronger associations in women suggest that sex differences might exist in the complex interplay between BP and inflammation.

The ghost of past species occurrence: improving species distribution models for presence-only data

1. Model-based approaches have been used increasingly in conservation biology over recent years. Species presence data used for predictive species distribution modelling are abundant in natural history collections, whereas reliable absence data are sparse, most notably for vagrant species such as butterflies and snakes. As predictive methods such as generalized linear models (GLM) require absence data, various strategies have been proposed to select pseudo-absence data. However, only a few studies exist that compare different approaches to generating these pseudo-absence data. 2. Natural history collection data are usually available for long periods of time (decades or even centuries), thus allowing historical considerations. However, this historical dimension has rarely been assessed in studies of species distribution, although there is great potential for understanding current patterns, i.e. the past is the key to the present. 3. We used GLM to model the distributions of three 'target' butterfly species, Melitaea didyma, Coenonympha tullia and Maculinea teleius, in Switzerland. We developed and compared four strategies for defining pools of pseudo-absence data and applied them to natural history collection data from the last 10, 30 and 100 years. Pools included: (i) sites without target species records; (ii) sites where butterfly species other than the target species were present; (iii) sites without butterfly species but with habitat characteristics similar to those required by the target species; and (iv) a combination of the second and third strategies. Models were evaluated and compared by the total deviance explained, the maximized Kappa and the area under the curve (AUC). 4. Among the four strategies, model performance was best for strategy 3. Contrary to expectations, strategy 2 resulted in even lower model performance compared with models with pseudo-absence data simulated totally at random (strategy 1). 5. Independent of the strategy model, performance was enhanced when sites with historical species presence data were not considered as pseudo-absence data. Therefore, the combination of strategy 3 with species records from the last 100 years achieved the highest model performance. 6. Synthesis and applications. The protection of suitable habitat for species survival or reintroduction in rapidly changing landscapes is a high priority among conservationists. Model-based approaches offer planning authorities the possibility of delimiting priority areas for species detection or habitat protection. The performance of these models can be enhanced by fitting them with pseudo-absence data relying on large archives of natural history collection species presence data rather than using randomly sampled pseudo-absence data.

Spatial modelling of biodiversity at the community level

1. Statistical modelling is often used to relate sparse biological survey data to remotely derived environmental predictors, thereby providing a basis for predictively mapping biodiversity across an entire region of interest. The most popular strategy for such modelling has been to model distributions of individual species one at a time. Spatial modelling of biodiversity at the community level may, however, confer significant benefits for applications involving very large numbers of species, particularly if many of these species are recorded infrequently. 2. Community-level modelling combines data from multiple species and produces information on spatial pattern in the distribution of biodiversity at a collective community level instead of, or in addition to, the level of individual species. Spatial outputs from community-level modelling include predictive mapping of community types (groups of locations with similar species composition), species groups (groups of species with similar distributions), axes or gradients of compositional variation, levels of compositional dissimilarity between pairs of locations, and various macro-ecological properties (e.g. species richness). 3. Three broad modelling strategies can be used to generate these outputs: (i) 'assemble first, predict later', in which biological survey data are first classified, ordinated or aggregated to produce community-level entities or attributes that are then modelled in relation to environmental predictors; (ii) 'predict first, assemble later', in which individual species are modelled one at a time as a function of environmental variables, to produce a stack of species distribution maps that is then subjected to classification, ordination or aggregation; and (iii) 'assemble and predict together', in which all species are modelled simultaneously, within a single integrated modelling process. These strategies each have particular strengths and weaknesses, depending on the intended purpose of modelling and the type, quality and quantity of data involved. 4. Synthesis and applications. The potential benefits of modelling large multispecies data sets using community-level, as opposed to species-level, approaches include faster processing, increased power to detect shared patterns of environmental response across rarely recorded species, and enhanced capacity to synthesize complex data into a form more readily interpretable by scientists and decision-makers. Community-level modelling therefore deserves to be considered more often, and more widely, as a potential alternative or supplement to modelling individual species.