Maximum likelihood Linear Programming Data Fusion for Speaker Recognition

Biometric system performance can be improved by means of data fusion. Several kinds of information can be fused in order to obtain a more accurate classification (identification or verification) of an input sample. In this paper we present a method for computing the weights in a weighted sum fusion for score combinations, by means of a likelihood model. The maximum likelihood estimation is set as a linear programming problem. The scores are derived from a GMM classifier working on a different feature extractor. Our experimental results assesed the robustness of the system in front a changes on time (different sessions) and robustness in front a change of microphone. The improvements obtained were significantly better (error bars of two standard deviations) than a uniform weighted sum or a uniform weighted product or the best single classifier. The proposed method scales computationaly with the number of scores to be fussioned as the simplex method for linear programming.

A general theory of rank testing

This paper develops an approach to rank testing that nests all existing rank tests andsimplifies their asymptotics. The approach is based on the fact that implicit in every ranktest there are estimators of the null spaces of the matrix in question. The approach yieldsmany new insights about the behavior of rank testing statistics under the null as well as localand global alternatives in both the standard and the cointegration setting. The approach alsosuggests many new rank tests based on alternative estimates of the null spaces as well as thenew fixed-b theory. A brief Monte Carlo study illustrates the results.

Ultrathin Portland Cement Concrete Overlay Extended Evaluation, Interim Report, TR-432, 2002

The Iowa Department of Transportation (Iowa DOT) UTW Project (HR-559) initiated Ultra-Thin Whitetopping in Iowa. The project is located on Iowa Highway 21 between Iowa Highway 212 and U.S. Highway 6 in Iowa County, near Belle Plaine, Iowa. The above listed research project lasted for five years, and then was extended for another five year period. The new phase of the project (TR 432) was initiated by removing cracked panels existing in the 2-inch thick PCC sections and replacing them with three inches of PCC. The project extension provides an increased understanding of slab bonding conditions over a longer period, as well as knowledge regarding the behavior of the newly rehabilitated areas. This report documents the rehabilitation of the PCC patching of all fractured panels and several cracked panels, taking place in September of 2001.

Lack of extended venous thromboembolism prophylaxis in high-risk patients undergoing major orthopaedic or major cancer surgery. Electronic Assessment of VTE Prophylaxis in High-Risk Surgical Patients at Discharge from Swiss Hospitals (ESSENTIAL).

Extended pharmacological venous thromboembolism (VTE) prophylaxis beyond discharge is recommended for patients undergoing high-risk surgery. We prospectively investigated prophylaxis in 1,046 consecutive patients undergoing major orthopaedic (70%) or major cancer surgery (30%) in 14 Swiss hospitals. Appropriate in-hospital prophylaxis was used in 1,003 (96%) patients. At discharge, 638 (61%) patients received prescription for extended pharmacological prophylaxis: 564 (77%) after orthopaedic surgery, and 74 (23%) after cancer surgery (p < 0.001). Patients with knee replacement (94%), hip replacement (81%), major trauma (80%), and curative arthroscopy (73%) had the highest prescription rates for extended VTE prophylaxis; the lowest rates were found in patients undergoing major surgery for thoracic (7%), gastrointestinal (19%), and hepatobiliary (33%) cancer. The median duration of prescribed extended prophylaxis was longer in patients with orthopaedic surgery (32 days, interquartile range 14-40 days) than in patients with cancer surgery (23 days, interquartile range 11-30 days; p<0.001). Among the 278 patients with an extended prophylaxis order after hip replacement, knee replacement, or hip fracture surgery, 120 (43%) received a prescription for at least 35 days, and among the 74 patients with an extended prophylaxis order after major cancer surgery, 20 (27%) received a prescription for at least 28 days. In conclusion, approximately one quarter of the patients with major orthopaedic surgery and more than three quarters of the patients with major cancer surgery did not receive prescription for extended VTE prophylaxis. Future effort should focus on the improvement of extended VTE prophylaxis, particularly in patients undergoing major cancer surgery.

Troponin-based risk stratification of patients with acute nonmassive pulmonary embolism: systematic review and metaanalysis.

BACKGROUND: Controversy exists regarding the usefulness of troponin testing for the risk stratification of patients with acute pulmonary embolism (PE). We conducted an updated systematic review and a metaanalysis of troponin-based risk stratification of normotensive patients with acute symptomatic PE. The sources of our data were publications listed in Medline and Embase from 1980 through April 2008 and a review of cited references in those publications. METHODS: We included all studies that estimated the relation between troponin levels and the incidence of all-cause mortality in normotensive patients with acute symptomatic PE. Two reviewers independently abstracted data and assessed study quality. From the literature search, 596 publications were screened. Nine studies that consisted of 1,366 normotensive patients with acute symptomatic PE were deemed eligible. Pooled results showed that elevated troponin levels were associated with a 4.26-fold increased odds of overall mortality (95% CI, 2.13 to 8.50; heterogeneity chi(2) = 12.64; degrees of freedom = 8; p = 0.125). Summary receiver operating characteristic curve analysis showed a relationship between the sensitivity and specificity of troponin levels to predict overall mortality (Spearman rank correlation coefficient = 0.68; p = 0.046). Pooled likelihood ratios (LRs) were not extreme (negative LR, 0.59 [95% CI, 0.39 to 0.88]; positive LR, 2.26 [95% CI, 1.66 to 3.07]). The Begg rank correlation method did not detect evidence of publication bias. CONCLUSIONS: The results of this metaanalysis indicate that elevated troponin levels do not adequately discern normotensive patients with acute symptomatic PE who are at high risk for death from those who are at low risk for death.

Fusion of the extended modified liquid drop model for nucleation and dynamical nucleation theory

We present a new phenomenological approach to nucleation, based on the combination of the extended modified liquid drop model and dynamical nucleation theory. The new model proposes a new cluster definition, which properly includes the effect of fluctuations, and it is consistent both thermodynamically and kinetically. The model is able to predict successfully the free energy of formation of the critical nucleus, using only macroscopic thermodynamic properties. It also accounts for the spinodal and provides excellent agreement with the result of recent simulations.

Probabilistic evidential assessment of gunshot residue particle evidence (Part I): likelihood ratio calculation and case pre-assessment using Bayesian networks

Well developed experimental procedures currently exist for retrieving and analyzing particle evidence from hands of individuals suspected of being associated with the discharge of a firearm. Although analytical approaches (e.g. automated Scanning Electron Microscopy with Energy Dispersive X-ray (SEM-EDS) microanalysis) allow the determination of the presence of elements typically found in gunshot residue (GSR) particles, such analyses provide no information about a given particle's actual source. Possible origins for which scientists may need to account for are a primary exposure to the discharge of a firearm or a secondary transfer due to a contaminated environment. In order to approach such sources of uncertainty in the context of evidential assessment, this paper studies the construction and practical implementation of graphical probability models (i.e. Bayesian networks). These can assist forensic scientists in making the issue tractable within a probabilistic perspective. The proposed models focus on likelihood ratio calculations at various levels of detail as well as case pre-assessment.

Bayes linear spaces

Linear spaces consisting of σ-finite probability measures and infinite measures (improper priors and likelihood functions) are defined. The commutative group operation, called perturbation, is the updating given by Bayes theorem; the inverse operation is the Radon-Nikodym derivative. Bayes spaces of measures are sets of classes of proportional measures. In this framework, basic notions of mathematical statistics get a simple algebraic interpretation. For example, exponential families appear as affine subspaces with their sufficient statistics as a basis. Bayesian statistics, in particular some well-known properties of conjugated priors and likelihood functions, are revisited and slightly extended

On the inefficiency of the restricted maximum likelihood

The restricted maximum likelihood is preferred by many to the full maximumlikelihood for estimation with variance component and other randomcoefficientmodels, because the variance estimator is unbiased. It is shown that thisunbiasednessis accompanied in some balanced designs by an inflation of the meansquared error.An estimator of the cluster-level variance that is uniformly moreefficient than the fullmaximum likelihood is derived. Estimators of the variance ratio are alsostudied.

Extended access cocaine self-administration differentially activates dorsal raphe and amygdala corticotropin-releasing factor systems in rats.

Cocaine-induced neuroadaptation of stress-related circuitry and increased access to cocaine each putatively contribute to the transition from cocaine use to cocaine dependence. The present study tested the hypothesis that rats receiving extended versus brief daily access to cocaine would exhibit regional differences in levels of the stress-regulatory neuropeptide corticotropin-releasing factor (CRF). A secondary goal was to explore how CRF levels change in relation to the time since cocaine self-administration. Male Wistar rats acquired operant self-administration of cocaine and were assigned to receive daily long access (6 hours/day, LgA, n = 20) or short access (1 hour/day, ShA, n = 18) to intravenous cocaine self-administration (fixed ratio 1, ∼0.50 mg/kg/infusion). After at least 3 weeks, tissue CRF immunoreactivity was measured at one of three timepoints: pre-session, post-session or 3 hours post-session. LgA, but not ShA, rats showed increased total session and first-hour cocaine intake. CRF immunoreactivity increased within the dorsal raphe (DR) and basolateral, but not central, nucleus of the amygdala (BLA, CeA) of ShA rats from pre-session to 3 hours post-session. In LgA rats, CRF immunoreactivity increased from pre-session to 3 hours post-session within the CeA and DR but tended to decrease in the BLA. LgA rats showed higher CRF levels than ShA rats in the DR and, pre-session, in the BLA. Thus, voluntary cocaine intake engages stress-regulatory CRF systems of the DR and amygdala. Increased availability of cocaine promotes greater tissue CRF levels in these extrahypothalamic brain regions, changes associated here with a model of cocaine dependence.

BadiRate: Estimating Family Turnover Rates by Likelihood-Based Methods

Motivation: The comparative analysis of gene gain and loss rates is critical for understanding the role of natural selection and adaptation in shaping gene family sizes. Studying complete genome data from closely related species allows accurate estimation of gene family turnover rates. Current methods and software tools, however, are not well designed for dealing with certain kinds of functional elements, such as microRNAs or transcription factor binding sites. Results: Here, we describe BadiRate, a new software tool to estimate family turnover rates, as well as the number of elements in internal phylogenetic nodes, by likelihood-based methods and parsimony. It implements two stochastic population models, which provide the appropriate statistical framework for testing hypothesis, such as lineage-specific gene family expansions or contractions. We have assessed the accuracy of BadiRate by computer simulations, and have also illustrated its functionality by analyzing a representative empirical dataset.

Intraocular implants for extended drug delivery: therapeutic applications.

An overview of ocular implants with therapeutic application potentials is provided. Various types of implants can be used as slow release devices delivering locally the needed drug for an extended period of time. Thus, multiple periocular or intraocular injections of the drug can be circumvented and secondary complications minimized. The various compositions of polymers fulfilling specific delivery goals are described. Several of these implants are undergoing clinical trials while a few are already commercialized. Despite the paramount progress in design, safety and efficacy, the place of these implants in our clinical therapeutic arsenal remains limited. Miniaturization of the implants allowing for their direct injection without the need for a complicated surgery is a necessary development avenue. Particulate systems which can be engineered to target specifically certain cells or tissues are another promising alternative. For ocular diseases affecting the choroid and outer retina, transscleral or intrasscleral implants are gaining momentum.

Interrogating eleven fast-evolving genes for signatures of recent positive selection in worldwide human populations

Different signatures of natural selection persist over varying time scales in our genome, revealing possible episodes of adaptative evolution during human history. Here, we identify genes showing signatures of ancestral positive selection in the human lineage and investigate whether some of those genes have been evolving adaptatively in extant human populations. Specifically, we compared more than 11,000 human genes with their orthologs inchimpanzee, mouse, rat and dog and applied a branch-site likelihood method to test for positive selection on the human lineage. Among the significant cases, a robust set of 11 genes were then further explored for signatures of recent positive selection using SNP data. We genotyped 223 SNPs in 39 worldwide populations from the HGDP Diversity panel and supplemented this information with available genotypes for up to 4,814 SNPs distributed along 2 Mb centered on each gene. After exploring the allele frequency spectrum, population differentiation and the maintainance of long unbroken haplotypes, we found signals of recent adaptative phenomena in only one of the 11 candidate gene regions. However, the signal ofrecent selection in this region may come from a different, neighbouring gene (CD5) ratherthan from the candidate gene itself (VPS37C). For this set of positively-selected genes in thehuman lineage, we find no indication that these genes maintained their rapid evolutionarypace among human populations. Based on these data, it therefore appears that adaptation forhuman-specific and for population-specific traits may have involved different genes.