Cluster kernels for semisupervised classification of VHR urban images

In this paper, we present and apply a semisupervised support vector machine based on cluster kernels for the problem of very high resolution image classification. In the proposed setting, a base kernel working with labeled samples only is deformed by a likelihood kernel encoding similarities between unlabeled examples. The resulting kernel is used to train a standard support vector machine (SVM) classifier. Experiments carried out on very high resolution (VHR) multispectral and hyperspectral images using very few labeled examples show the relevancy of the method in the context of urban image classification. Its simplicity and the small number of parameters involved make it versatile and workable by unexperimented users.

Determinants and burden of chronic kidney disease in the population-based CoLaus study: a cross-sectional analysis.

BACKGROUND: Chronic kidney disease (CKD) represents an increasing health burden. We present the population-based prevalence of CKD and compare the CKD Epidemiology collaboration (CKD-EPI) and modification of diet in renal disease (MDRD) equations to estimate the glomerular filtration rate, using the revised CKD classification with three albuminuria classes. We also explore factors associated with CKD. METHODS: The Swiss population-based, cross-sectional CoLaus study conducted in Lausanne (2003-2006) included 2810 men and 3111 women aged 35-75. CKD prevalence was assessed using CKD-EPI and MDRD equations and albuminuria estimated by the albumin-to-creatinine ratio in spot morning urine. Multivariate logistic regression was used to analyse determinants of CKD. RESULTS: Prevalence [95% confidence interval (CI)] of all stages CKD was 10.0% (9.2-10.8%) with CKD-EPI and 13.8% (12.9-14.6%) with MDRD. Using the revised CKD classification, the prevalence of low-, medium-, high- and very high-risk groups was 90.0, 8.46, 1.18 and 0.35% with CKD-EPI, respectively. With MDRD, the corresponding values were 86.24, 11.86, 1.55 and 0.35%. Using the revised classification, CKD-EPI systematically reclassified people in a lower risk category than MDRD. Age and obesity were more strongly associated with CKD in men [odds ratio (95% CI): 2.23(1.95; 2.56) per 10 years and 3.05(2.08;4.47), respectively] than in women [1.46 (1.29; 1.65) and 1.78 (1.30;2.44), respectively]. Hypertension, type 2 diabetes, serum homocysteine and uric acid were positively independently associated with CKD in men and women. CONCLUSIONS: One in 10 adults suffers from CKD in the population of Lausanne. CKD-EPI systematically reclassifies people in a lower CKD risk category than MDRD. Serum homocysteine and uric acid levels are associated with CKD independently of classical risk factors such as age, hypertension and diabetes.

Automatic classification of breast density

A recent trend in digital mammography is computer-aided diagnosis systems, which are computerised tools designed to assist radiologists. Most of these systems are used for the automatic detection of abnormalities. However, recent studies have shown that their sensitivity is significantly decreased as the density of the breast increases. This dependence is method specific. In this paper we propose a new approach to the classification of mammographic images according to their breast parenchymal density. Our classification uses information extracted from segmentation results and is based on the underlying breast tissue texture. Classification performance was based on a large set of digitised mammograms. Evaluation involves different classifiers and uses a leave-one-out methodology. Results demonstrate the feasibility of estimating breast density using image processing and analysis techniques

Breast cancer prognostic classification in the molecular era: the role of histological grade.

Breast cancer is a heterogeneous disease with varied morphological appearances, molecular features, behavior, and response to therapy. Current routine clinical management of breast cancer relies on the availability of robust clinical and pathological prognostic and predictive factors to support clinical and patient decision making in which potentially suitable treatment options are increasingly available. One of the best-established prognostic factors in breast cancer is histological grade, which represents the morphological assessment of tumor biological characteristics and has been shown to be able to generate important information related to the clinical behavior of breast cancers. Genome-wide microarray-based expression profiling studies have unraveled several characteristics of breast cancer biology and have provided further evidence that the biological features captured by histological grade are important in determining tumor behavior. Also, expression profiling studies have generated clinically useful data that have significantly improved our understanding of the biology of breast cancer, and these studies are undergoing evaluation as improved prognostic and predictive tools in clinical practice. Clinical acceptance of these molecular assays will require them to be more than expensive surrogates of established traditional factors such as histological grade. It is essential that they provide additional prognostic or predictive information above and beyond that offered by current parameters. Here, we present an analysis of the validity of histological grade as a prognostic factor and a consensus view on the significance of histological grade and its role in breast cancer classification and staging systems in this era of emerging clinical use of molecular classifiers.

Risk of dengue occurrence based on the capture of gravid Aedes aegypti females using MosquiTRAP

We assessed the risk classification of dengue fever based on the capture of Aedes aegypti adults using MosquiTRAP, a type of sticky trap, in comparison with traditional larval infestation indices. A total of 27 MosquiTRAPs were installed, with one trap per block, and were inspected weekly between November 2008-February 2009. Infestation baseline data were obtained from a survey conducted prior to trap installation. The index generated by MosquiTRAP and house index (HI) classified the area "in alert situation". The set for risk of dengue occurrence proposed by the use of MosquiTRAP classify areas in the same way of the traditional HI.

Classification of immature mosquito species according to characteristics of the larval habitat in the subtropical province of Chaco, Argentina

To classify mosquito species based on common features of their habitats, samples were obtained fortnightly between June 2001-October 2003 in the subtropical province of Chaco, Argentina. Data on the type of larval habitat, nature of the habitat (artificial or natural), size, depth, location related to sunlight, distance to the neighbouring houses, type of substrate, organic material, vegetation and algae type and their presence were collected. Data on the permanence, temperature, pH, turbidity, colour, odour and movement of the larval habitat's water were also collected. From the cluster analysis, three groups of species associated by their degree of habitat similarity were obtained and are listed below. Group 1 consisted of Aedes aegypti. Group 2 consisted of Culex imitator, Culex davisi, Wyeomyia muehlensi and Toxorhynchites haemorrhoidalis separatus. Within group 3, two subgroups are distinguished: A (Psorophora ferox, Psorophora cyanescens, Psorophora varinervis, Psorophora confinnis, Psorophora cingulata, Ochlerotatus hastatus-oligopistus, Ochlerotatus serratus, Ochlerotatus scapularis, Culex intrincatus, Culex quinquefasciatus, Culex pilosus, Ochlerotatus albifasciatus, Culex bidens) and B (Culex maxi, Culex eduardoi, Culex chidesteri, Uranotaenia lowii, Uranotaenia pulcherrima, Anopheles neomaculipalpus, Anopheles triannulatus, Anopheles albitarsis, Uranotaenia apicalis, Mansonia humeralis and Aedeomyia squamipennis). Principal component analysis indicates that the size of the larval habitats and the presence of aquatic vegetation are the main characteristics that explain the variation among different species. In contrast, water permanence is second in importance. Water temperature, pH and the type of larval habitat are less important in explaining the clustering of species.

Conceptual framework and pilot study to benchmark phylogenomic databases based on reference gene trees.

Phylogenomic databases provide orthology predictions for species with fully sequenced genomes. Although the goal seems well-defined, the content of these databases differs greatly. Seven ortholog databases (Ensembl Compara, eggNOG, HOGENOM, InParanoid, OMA, OrthoDB, Panther) were compared on the basis of reference trees. For three well-conserved protein families, we observed a generally high specificity of orthology assignments for these databases. We show that differences in the completeness of predicted gene relationships and in the phylogenetic information are, for the great majority, not due to the methods used, but to differences in the underlying database concepts. According to our metrics, none of the databases provides a fully correct and comprehensive protein classification. Our results provide a framework for meaningful and systematic comparisons of phylogenomic databases. In the future, a sustainable set of 'Gold standard' phylogenetic trees could provide a robust method for phylogenomic databases to assess their current quality status, measure changes following new database releases and diagnose improvements subsequent to an upgrade of the analysis procedure.

Disease identification based on ambulatory drugs dispensation and in-hospital ICD-10 diagnoses: a comparison.

BACKGROUND: Pharmacy-based case mix measures are an alternative source of information to the relatively scarce outpatient diagnoses data. But most published tools use national drug nomenclatures and offer no head-to-head comparisons between drugs-related and diagnoses-based categories. The objective of the study was to test the accuracy of drugs-based morbidity groups derived from the World Health Organization Anatomical Therapeutic Chemical Classification of drugs by checking them against diagnoses-based groups. METHODS: We compared drugs-based categories with their diagnoses-based analogues using anonymous data on 108,915 individuals insured with one of four companies. They were followed throughout 2005 and 2006 and hospitalized at least once during this period. The agreement between the two approaches was measured by weighted kappa coefficients. The reproducibility of the drugs-based morbidity measure over the 2 years was assessed for all enrollees. RESULTS: Eighty percent used a drug associated with at least one of the 60 morbidity categories derived from drugs dispensation. After accounting for inpatient under-coding, fifteen conditions agreed sufficiently with their diagnoses-based counterparts to be considered alternative strategies to diagnoses. In addition, they exhibited good reproducibility and allowed prevalence estimates in accordance with national estimates. For 22 conditions, drugs-based information identified accurately a subset of the population defined by diagnoses. CONCLUSIONS: Most categories provide insurers with health status information that could be exploited for healthcare expenditure prediction or ambulatory cost control, especially when ambulatory diagnoses are not available. However, due to insufficient concordance with their diagnoses-based analogues, their use for morbidity indicators is limited.

Fair bounds based solidarity

How should scholarships be distributed among the (public) higher education students? We raise this situation as a redistribution problem. Following the approach developed in Fleurbaey (1994) and Bossert (1995), redistribution should be based on the notion of solidarity and it re-allocates resources taking into account only agents’ relevant characteristics. We also follow Luttens (2010a), who considers that compensation of relevant characteristics must be based on a lower bound on what every individual deserves. In doing so, we use the so-called fair bound (Moulin (2002)) to define an egalitarian redistribution mechanism and characterize it in terms of non-negativity, priority in lower bound and solidarity. Finally, we apply our approach to the scholarships redistribution problem. Keywords: Redistribution mechanism, Lower bounds, Scholarship, Solidarity. JEL classification: C71, D63, D71

Revised nomenclature and classification of inherited ichthyoses: results of the First Ichthyosis Consensus Conference in Sorèze 2009.

BACKGROUND: Inherited ichthyoses belong to a large, clinically and etiologically heterogeneous group of mendelian disorders of cornification, typically involving the entire integument. Over the recent years, much progress has been made defining their molecular causes. However, there is no internationally accepted classification and terminology. OBJECTIVE: We sought to establish a consensus for the nomenclature and classification of inherited ichthyoses. METHODS: The classification project started at the First World Conference on Ichthyosis in 2007. A large international network of expert clinicians, skin pathologists, and geneticists entertained an interactive dialogue over 2 years, eventually leading to the First Ichthyosis Consensus Conference held in Sorèze, France, on January 23 and 24, 2009, where subcommittees on different issues proposed terminology that was debated until consensus was reached. RESULTS: It was agreed that currently the nosology should remain clinically based. "Syndromic" versus "nonsyndromic" forms provide a useful major subdivision. Several clinical terms and controversial disease names have been redefined: eg, the group caused by keratin mutations is referred to by the umbrella term, "keratinopathic ichthyosis"-under which are included epidermolytic ichthyosis, superficial epidermolytic ichthyosis, and ichthyosis Curth-Macklin. "Autosomal recessive congenital ichthyosis" is proposed as an umbrella term for the harlequin ichthyosis, lamellar ichthyosis, and the congenital ichthyosiform erythroderma group. LIMITATIONS: As more becomes known about these diseases in the future, modifications will be needed. CONCLUSION: We have achieved an international consensus for the classification of inherited ichthyosis that should be useful for all clinicians and can serve as reference point for future research.

New classification of natural breeding habitats for Neotropical anophelines in the Yanomami Indian Reserve, Amazon Region, Brazil and a new larval sampling methodology

Here we present the first in a series of articles about the ecology of immature stages of anophelines in the Brazilian Yanomami area. We propose a new larval habitat classification and a new larval sampling methodology. We also report some preliminary results illustrating the applicability of the methodology based on data collected in the Brazilian Amazon rainforest in a longitudinal study of two remote Yanomami communities, Parafuri and Toototobi. In these areas, we mapped and classified 112 natural breeding habitats located in low-order river systems based on their association with river flood pulses, seasonality and exposure to sun. Our classification rendered seven types of larval habitats: lakes associated with the river, which are subdivided into oxbow lakes and nonoxbow lakes, flooded areas associated with the river, flooded areas not associated with the river, rainfall pools, small forest streams, medium forest streams and rivers. The methodology for larval sampling was based on the accurate quantification of the effective breeding area, taking into account the area of the perimeter and subtypes of microenvironments present per larval habitat type using a laser range finder and a small portable inflatable boat. The new classification and new sampling methodology proposed herein may be useful in vector control programs.

Classification and clinical behavior of blastic plasmacytoid dendritic cell neoplasms according to their maturation-associated immunophenotypic profile.

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare subtype of leukemia/lymphoma, whose diagnosis can be difficult to achieve due to its clinical and biological heterogeneity, as well as its overlapping features with other hematologic malignancies. In this study we investigated whether the association between the maturational stage of tumor cells and the clinico-biological and prognostic features of the disease, based on the analysis of 46 BPDCN cases classified into three maturation-associated subgroups on immunophenotypic grounds. Our results show that blasts from cases with an immature plasmacytoid dendritic cell (pDC) phenotype exhibit an uncommon CD56- phenotype, coexisting with CD34+ non-pDC tumor cells, typically in the absence of extramedullary (e.g. skin) disease at presentation. Conversely, patients with a more mature blast cell phenotype more frequently displayed skin/extramedullary involvement and spread into secondary lymphoid tissues. Despite the dismal outcome, acute lymphoblastic leukemia-type therapy (with central nervous system prophylaxis) and/or allogeneic stem cell transplantation appeared to be the only effective therapies. Overall, our findings indicate that the maturational profile of pDC blasts in BPDCN is highly heterogeneous and translates into a wide clinical spectrum -from acute leukemia to mature lymphoma-like behavior-, which may also lead to variable diagnosis and treatment.

Grain size analyses in tin-lead glazes based on 2d-sections

A problem in the archaeometric classification of Catalan Renaissance pottery is the fact, thatthe clay supply of the pottery workshops was centrally organized by guilds, and thereforeusually all potters of a single production centre produced chemically similar ceramics.However, analysing the glazes of the ware usually a large number of inclusions in the glaze isfound, which reveal technological differences between single workshops. These inclusionshave been used by the potters in order to opacify the transparent glaze and to achieve a whitebackground for further decoration.In order to distinguish different technological preparation procedures of the single workshops,at a Scanning Electron Microscope the chemical composition of those inclusions as well astheir size in the two-dimensional cut is recorded. Based on the latter, a frequency distributionof the apparent diameters is estimated for each sample and type of inclusion.Following an approach by S.D. Wicksell (1925), it is principally possible to transform thedistributions of the apparent 2D-diameters back to those of the true three-dimensional bodies.The applicability of this approach and its practical problems are examined using differentways of kernel density estimation and Monte-Carlo tests of the methodology. Finally, it istested in how far the obtained frequency distributions can be used to classify the pottery

A Novel Breast Tissue Density Classification Methodology

It has been shown that the accuracy of mammographic abnormality detection methods is strongly dependent on the breast tissue characteristics, where a dense breast drastically reduces detection sensitivity. In addition, breast tissue density is widely accepted to be an important risk indicator for the development of breast cancer. Here, we describe the development of an automatic breast tissue classification methodology, which can be summarized in a number of distinct steps: 1) the segmentation of the breast area into fatty versus dense mammographic tissue; 2) the extraction of morphological and texture features from the segmented breast areas; and 3) the use of a Bayesian combination of a number of classifiers. The evaluation, based on a large number of cases from two different mammographic data sets, shows a strong correlation ( and 0.67 for the two data sets) between automatic and expert-based Breast Imaging Reporting and Data System mammographic density assessment

Object and Scene Classification: what does a Supervised Approach Provide us?

Given a set of images of scenes containing different object categories (e.g. grass, roads) our objective is to discover these objects in each image, and to use this object occurrences to perform a scene classification (e.g. beach scene, mountain scene). We achieve this by using a supervised learning algorithm able to learn with few images to facilitate the user task. We use a probabilistic model to recognise the objects and further we classify the scene based on their object occurrences. Experimental results are shown and evaluated to prove the validity of our proposal. Object recognition performance is compared to the approaches of He et al. (2004) and Marti et al. (2001) using their own datasets. Furthermore an unsupervised method is implemented in order to evaluate the advantages and disadvantages of our supervised classification approach versus an unsupervised one