Cluster-based active learning for compact image classification

In this paper, we consider active sampling to label pixels grouped with hierarchical clustering. The objective of the method is to match the data relationships discovered by the clustering algorithm with the user's desired class semantics. The first is represented as a complete tree to be pruned and the second is iteratively provided by the user. The active learning algorithm proposed searches the pruning of the tree that best matches the labels of the sampled points. By choosing the part of the tree to sample from according to current pruning's uncertainty, sampling is focused on most uncertain clusters. This way, large clusters for which the class membership is already fixed are no longer queried and sampling is focused on division of clusters showing mixed labels. The model is tested on a VHR image in a multiclass classification setting. The method clearly outperforms random sampling in a transductive setting, but cannot generalize to unseen data, since it aims at optimizing the classification of a given cluster structure.

Current Issues in the Diagnosis and Classification of Peripheral T-Cell Lymphomas

Mature T-cell and T/NK-cell neoplasms are both uncommon and heterogeneous, among the broad category of non-Hodgkin's lymphomas. Due to the lack of specific genetic alterations in the vast majority of cases, most currently defined entities show overlapping morphologic and immunophenotypic features and therefore pose a challenge to the diagnostic pathologist. The goal of the symposium is to address current criteria for the recognition of specific subtypes of T-cell lymphoma, and to highlight new data regarding emerging immunophenotypic or molecular markers. This activity has been designed to meet the needs of practicing pathologists, and residents and fellows enrolled in training programs in anatomic and clinical pathology. It should be a particular benefit to those with an interest in hematopathology. Upon completion of this activity, participants should be better able to: -To be able to state the basis for the classification of mature T-cell malignancies involving nodal and extranodal sites. -To recognize and accurately diagnose the various subtypes of nodal and extranodal peripheral T-cell lymphomas. -To utilize immunohistochemical and molecular tests to characterize atypical T-cell proliferations. -To recognize and accurately diagnose T-cell lymphoproliferative lesions involving the skin and gastrointestinal tract, and be able to provide guidance regarding their clinical aggressiveness and management -To be able to utilize flow cytometric data to identify diverse functional T-cell subsets.

Image-Based Coral Reef Classification and Thematic Mapping

This paper presents a novel image classification scheme for benthic coral reef images that can be applied to both single image and composite mosaic datasets. The proposed method can be configured to the characteristics (e.g., the size of the dataset, number of classes, resolution of the samples, color information availability, class types, etc.) of individual datasets. The proposed method uses completed local binary pattern (CLBP), grey level co-occurrence matrix (GLCM), Gabor filter response, and opponent angle and hue channel color histograms as feature descriptors. For classification, either k-nearest neighbor (KNN), neural network (NN), support vector machine (SVM) or probability density weighted mean distance (PDWMD) is used. The combination of features and classifiers that attains the best results is presented together with the guidelines for selection. The accuracy and efficiency of our proposed method are compared with other state-of-the-art techniques using three benthic and three texture datasets. The proposed method achieves the highest overall classification accuracy of any of the tested methods and has moderate execution time. Finally, the proposed classification scheme is applied to a large-scale image mosaic of the Red Sea to create a completely classified thematic map of the reef benthos

Ability of different adiposity indicators to identify children with elevated blood pressure.

OBJECTIVE: Body composition measured by dual-energy X-ray absorptiometry (DXA) is believed to be superior to crude measures such as BMI or waist circumference (WC) to assess health risks associated with adiposity in adults. We compared the ability of BMI, WC, waist-to-height ratio (WHtR), percentage body fat from skinfold thickness, and measures of total and central fat assessed by DXA to identify children with elevated blood pressure (BP). STUDY DESIGN: The QUALITY Study follows 630 Caucasian families (father, mother, and child originally aged 8-10 years). BP, height, weight, WC, and skinfold thickness were measured according to standardized protocols. Elevated BP was defined as systolic or diastolic BP at least 90th age, sex, and height-specific percentile. Total and central fat were determined with DXA. The area under the receiver operating characteristic (ROC) curve (AUC) statistic was computed from logistic models that adjusted for age, sex, height, Tanner stage, and physical activity. RESULTS: All adiposity indicators were highly correlated. WC and WHtR did not show superior ability over BMI to identify children with elevated SBP (P = 0.421 and 0.473). Measures of total and central fat from DXA did not show an improved ability over BMI or WC to identify children with elevated SBP (P = 0.325-0.662). CONCLUSION: Results support the use of BMI in clinical and public health settings, at least in this age group. As all indicators had a limited ability to identify children with elevated BP, results also support measurement of BP in all children of this age independent of a weight status.

Dispersal ability modulates the strength of the latitudinal richness gradient in European beetles.

Aim A debate exists as to whether present-day diversity gradients are governed by current environmental conditions or by changes in environmental conditions through time. Recent studies have shown that latitudinal richness gradients might be partially caused by incomplete post-glacial recolonization of high-latitude regions; this leads to the prediction that less mobile taxa should have steeper gradients than more mobile taxa. The aim of this study is to test this prediction. Location Europe. Methods We first assessed whether spatial turnover in species composition is a good surrogate for dispersal ability by measuring the proportion of wingless species in 19 European beetle clades and relating this value to spatial turnover (beta sim) of the clade. We then linearly regressed beta sim values of 21 taxa against the slope of their respective diversity gradients. Results A strong relationship exists between the proportion of wingless species and beta sim, and beta sim was found to be a good predictor of latitudinal richness gradients. Main conclusions Results are consistent with the prediction that poor dispersers have steeper richness gradients than good dispersers, supporting the view that current beetle diversity gradients in Europe are affected by post-glacial dispersal lags.

Altruism, spite and choosy females

Natural selection favours the genes which are able to introduce replicates of themselves in the next generation with higher certainty than do rival genes (Hamilton 1963). The fitness of an individual, it?s ability to produce future parents, depends on it?s own behaviour as well as on the behaviour of other individuals in the population. For instance, the intensity of competition an individual experience depends on the exploitation of resources by neighbours. The fitness is thus frequency dependent on what neighbours do. Behaviours can be classified according to the costs and benefits they have on the fitness of the behaver and it?s neighbours (Hamilton 1964, Hamilton 1975). According to this classification there exist four distinct social behaviours. (1) A gene confering the ability to use a new ressource is called selfish because it has a positive e_ect on the bearer of the gene but a negative e_ect on neighbours by the concomitant increase in competition. (2) An altruistic behaviour is defined as an action where an individual increases the fitness of a neighbour at the expense of it?s own. The e_ect is deleterious for the actor but positive for the receptor. (3) More surprinsingly, an individual might sacrifice a fraction of it?s ressources to harm another at no direct benefits. This spitefull behaviour incurs a cost for the actor but is also deleterious for the receptor. (4) Finally a cooperative behaviour breeds benefits for both actors and neighbours. In this thesis I will continue on the path traced by numerous evolutionnary biologist which attempt to fine tune our understanding of the evolution of social behaviours since Hamilton?s foundation (1963, 1964). A critical development over the last 40 years has been the realisation that competition between kin can partly or completely cancel out the role of relatedness as an agent favouring altruism (Wilson et al., 1992; Taylor, 1992a,b). Of importance is thus to determine the scale at which competition and altruism occur. One mechanism avoiding the complete dilution of relatedness by competition is the conditionnal expression of the social behaviors. Focus will be given in this thesis at the role played by di_erent recognition mechanism in paving the way to altruism (Komdeur and Hatchwell, 1999) when the population has a spatial structure. Further, the evolution of spite will also be considered in these settings. The thesis is fractionated into two parts. First, di_erent models promoting altruism cooperation and spite will be compared under the same theoretical umbrella. This is a rather informal and more personnal part of my thesis. It also serve as a justification and basis to "Altruism among kin and non-kin individuals" which is an article attempting to clas- sify the mechanisms leading to altruism and cooperation. Second, in the annexe, there are three research papers about kin selection, altruism and dispersal: "Is sociality driven by the costs of dispersal or the benefits of philopatry?: A role for kin-discrimination mechanism", "Altruism, dispersal and phenotype kin recognition" and "Inbreeding avoidance through kin recognition: choosy female boost male dispersal" this last paper incorporates kin recognition as an agent favoring sex-biased dispersal.

Similarity measure based classification

Luokittelujärjestelmää suunniteltaessa tarkoituksena on rakentaa systeemi, joka pystyy ratkaisemaan mahdollisimman tarkasti tutkittavan ongelma-alueen. Hahmontunnistuksessa tunnistusjärjestelmän ydin on luokitin. Luokittelun sovellusaluekenttä on varsin laaja. Luokitinta tarvitaan mm. hahmontunnistusjärjestelmissä, joista kuvankäsittely toimii hyvänä esimerkkinä. Myös lääketieteen parissa tarkkaa luokittelua tarvitaan paljon. Esimerkiksi potilaan oireiden diagnosointiin tarvitaan luokitin, joka pystyy mittaustuloksista päättelemään mahdollisimman tarkasti, onko potilaalla kyseinen oire vai ei. Väitöskirjassa on tehty similaarisuusmittoihin perustuva luokitin ja sen toimintaa on tarkasteltu mm. lääketieteen paristatulevilla data-aineistoilla, joissa luokittelutehtävänä on tunnistaa potilaan oireen laatu. Väitöskirjassa esitetyn luokittimen etuna on sen yksinkertainen rakenne, josta johtuen se on helppo tehdä sekä ymmärtää. Toinen etu on luokittimentarkkuus. Luokitin saadaan luokittelemaan useita eri ongelmia hyvin tarkasti. Tämä on tärkeää varsinkin lääketieteen parissa, missä jo pieni tarkkuuden parannus luokittelutuloksessa on erittäin tärkeää. Väitöskirjassa ontutkittu useita eri mittoja, joilla voidaan mitata samankaltaisuutta. Mitoille löytyy myös useita parametreja, joille voidaan etsiä juuri kyseiseen luokitteluongelmaan sopivat arvot. Tämä parametrien optimointi ongelma-alueeseen sopivaksi voidaan suorittaa mm. evoluutionääri- algoritmeja käyttäen. Kyseisessä työssä tähän on käytetty geneettistä algoritmia ja differentiaali-evoluutioalgoritmia. Luokittimen etuna on sen joustavuus. Ongelma-alueelle on helppo vaihtaa similaarisuusmitta, jos kyseinen mitta ei ole sopiva tutkittavaan ongelma-alueeseen. Myös eri mittojen parametrien optimointi voi parantaa tuloksia huomattavasti. Kun käytetään eri esikäsittelymenetelmiä ennen luokittelua, tuloksia pystytään parantamaan.

Stratification of the severity of critically ill patients with classification trees

Background: Development of three classification trees (CT) based on the CART (Classification and Regression Trees), CHAID (Chi-Square Automatic Interaction Detection) and C4.5 methodologies for the calculation of probability of hospital mortality; the comparison of the results with the APACHE II, SAPS II and MPM II-24 scores, and with a model based on multiple logistic regression (LR). Methods: Retrospective study of 2864 patients. Random partition (70:30) into a Development Set (DS) n = 1808 and Validation Set (VS) n = 808. Their properties of discrimination are compared with the ROC curve (AUC CI 95%), Percent of correct classification (PCC CI 95%); and the calibration with the Calibration Curve and the Standardized Mortality Ratio (SMR CI 95%). Results: CTs are produced with a different selection of variables and decision rules: CART (5 variables and 8 decision rules), CHAID (7 variables and 15 rules) and C4.5 (6 variables and 10 rules). The common variables were: inotropic therapy, Glasgow, age, (A-a)O2 gradient and antecedent of chronic illness. In VS: all the models achieved acceptable discrimination with AUC above 0.7. CT: CART (0.75(0.71-0.81)), CHAID (0.76(0.72-0.79)) and C4.5 (0.76(0.73-0.80)). PCC: CART (72(69- 75)), CHAID (72(69-75)) and C4.5 (76(73-79)). Calibration (SMR) better in the CT: CART (1.04(0.95-1.31)), CHAID (1.06(0.97-1.15) and C4.5 (1.08(0.98-1.16)). Conclusion: With different methodologies of CTs, trees are generated with different selection of variables and decision rules. The CTs are easy to interpret, and they stratify the risk of hospital mortality. The CTs should be taken into account for the classification of the prognosis of critically ill patients.

Prediction of crude protein and classification of the growth stage of wheat plant samples from NIR spectra

Near-infrared spectroscopy (NIRS) was used to analyse the crude protein content of dried and milled samples of wheat and to discriminate samples according to their stage of growth. A calibration set of 72 samples from three growth stages of wheat (tillering, heading and harvest) and a validation set of 28 samples was collected for this purpose. Principal components analysis (PCA) of the calibration set discriminated groups of samples according to the growth stage of the wheat. Based on these differences, a classification procedure (SIMCA) showed a very accurate classification of the validation set samples : all of them were successfully classified in each group using this procedure when both the residual and the leverage were used in the classification criteria. Looking only at the residuals all the samples were also correctly classified except one of tillering stage that was assigned to both tillering and heading stages. Finally, the determination of the crude protein content of these samples was considered in two ways: building up a global model for all the growth stages, and building up local models for each stage, separately. The best prediction results for crude protein were obtained using a global model for samples in the two first growth stages (tillering and heading), and using a local model for the harvest stage samples.

Integrating defeasible argumentation with fuzzy ART neural networks for pattern classification

Many classification systems rely on clustering techniques in which a collection of training examples is provided as an input, and a number of clusters c1,...cm modelling some concept C results as an output, such that every cluster ci is labelled as positive or negative. Given a new, unlabelled instance enew, the above classification is used to determine to which particular cluster ci this new instance belongs. In such a setting clusters can overlap, and a new unlabelled instance can be assigned to more than one cluster with conflicting labels. In the literature, such a case is usually solved non-deterministically by making a random choice. This paper presents a novel, hybrid approach to solve this situation by combining a neural network for classification along with a defeasible argumentation framework which models preference criteria for performing clustering.

Evidence-based provisional clinical classification criteria for autoinflammatory periodic fevers.

The objective of this work was to develop and validate a set of clinical criteria for the classification of patients affected by periodic fevers. Patients with inherited periodic fevers (familial Mediterranean fever (FMF); mevalonate kinase deficiency (MKD); tumour necrosis factor receptor-associated periodic fever syndrome (TRAPS); cryopyrin-associated periodic syndromes (CAPS)) enrolled in the Eurofever Registry up until March 2013 were evaluated. Patients with periodic fever, aphthosis, pharyngitis and adenitis (PFAPA) syndrome were used as negative controls. For each genetic disease, patients were considered to be 'gold standard' on the basis of the presence of a confirmatory genetic analysis. Clinical criteria were formulated on the basis of univariate and multivariate analysis in an initial group of patients (training set) and validated in an independent set of patients (validation set). A total of 1215 consecutive patients with periodic fevers were identified, and 518 gold standard patients (291 FMF, 74 MKD, 86 TRAPS, 67 CAPS) and 199 patients with PFAPA as disease controls were evaluated. The univariate and multivariate analyses identified a number of clinical variables that correlated independently with each disease, and four provisional classification scores were created. Cut-off values of the classification scores were chosen using receiver operating characteristic curve analysis as those giving the highest sensitivity and specificity. The classification scores were then tested in an independent set of patients (validation set) with an area under the curve of 0.98 for FMF, 0.95 for TRAPS, 0.96 for MKD, and 0.99 for CAPS. In conclusion, evidence-based provisional clinical criteria with high sensitivity and specificity for the clinical classification of patients with inherited periodic fevers have been developed.