Subclass problem-dependent design for error-correcting output codes

A common way to model multiclass classification problems is by means of Error-Correcting Output Codes (ECOCs). Given a multiclass problem, the ECOC technique designs a code word for each class, where each position of the code identifies the membership of the class for a given binary problem. A classification decision is obtained by assigning the label of the class with the closest code. One of the main requirements of the ECOC design is that the base classifier is capable of splitting each subgroup of classes from each binary problem. However, we cannot guarantee that a linear classifier model convex regions. Furthermore, nonlinear classifiers also fail to manage some type of surfaces. In this paper, we present a novel strategy to model multiclass classification problems using subclass information in the ECOC framework. Complex problems are solved by splitting the original set of classes into subclasses and embedding the binary problems in a problem-dependent ECOC design. Experimental results show that the proposed splitting procedure yields a better performance when the class overlap or the distribution of the training objects conceal the decision boundaries for the base classifier. The results are even more significant when one has a sufficiently large training size.

A comparative long-memory Analysis between Spanish, Mexican and U.S. interest rates

Evidence exists that many natural facts are described better as a fractal. Although fractals are very useful for describing nature, it is also appropiate to review the concept of random fractal in finance. Due to the extraordinary importance of Brownian motion in physics, chemistry or biology, we will consider the generalization that supposes fractional Brownian motion introduced by Mandelbrot.The main goal of this work is to analyse the existence of long range dependence in instantaneous forward rates of different financial markets. Concretelly, we perform an empirical analysis on the Spanish, Mexican and U.S. interbanking interest rate. We work with three time series of daily data corresponding to 1 day operations from 28th March 1996 to 21st May 2002. From among all the existing tests on this matter we apply the methodology proposed in Taqqu, Teverovsky and Willinger (1995).

Term structure of interest rates. European financial integration

In this paper we estimate, analyze and compare the term structures of interest rates in six different countries over the period 1992-2004. We apply the Nelson-Siegel model to obtain the term structures of interest rates at weekly intervals. A total of 4,038 curves are estimated and analyzed. Four European Monetary Union countries¿Spain, France, Germany and Italy¿are included. The UK is also included as a European non-member of the Monetary Union. Finally the US completes the analysis. The goal is to determine the differences in the shapes of the term structure of interest rates among these countries. Likewise, we can determine the most usual term structure shapes that appear for each country.*****

Frequent accesses to totally implanted vascular ports in pediatric oncology patients are associated with higher infection rates.

PURPOSE: Totally implanted vascular (TIVA) ports are used in children for repeated blood samples or intravenous treatments. We have recently published a prospective evaluation of surgical incidents and early complications associated with these devices. This work is the final part of the same study, assessing late complications over a follow-up of 2 yrs. METHODS: From January 2006 to January 2008, children older than 1 yr of age with a diagnosis of solid or blood cell malignancy were included. Insertion technique and care of the device were standardized. Every manipulation was prospectively recorded by specialized nurses. Obstruction was documented clinically. When bacteremia was suspected, routine central and peripheral blood cultures were drawn. RESULTS: Forty-five consecutive patients were enrolled in the study. Mean age at the time of the procedure was 8.5 yrs. There was no catheter-related infection within the first 4 weeks post-surgery. No device had to be removed because of infection or obstruction during follow-up. Frequent accesses to the port (=3 per day over a 10-day period) were associated with an 8-fold risk of infection. CONCLUSION: Insertion and use of TIVA devices were frequently associated with complications. No device had to be removed because of infection or obstruction over the follow-up period, although no prophylactic antibiotic agent was used. Restrictive use of antibiotics may prevent opportunistic infection. Frequent access to the device was significantly associated with line infection (odds ratio=8.43). No risk factor was identified for obstruction which occurred at a rate of 5.3 per 10,000 accesses.

Rat splanchnic net oxygen consumption. Energetic implications

1. The blood flow, PO2, pH and PCO2 have been estimated in portal and suprahepatic veins as well as in hepatic artery of fed and overnight starved rats given an oral glucose load. From these data the net intestinal, hepatic and splanchnic balances for oxygen and bicarbonate were calculated. The oxygen consumption of the intact animal has also been measured under comparable conditions. 2. The direct utilization of oxygen balances as energy equivalents when establishing the contribution of energy metabolism of liver and intestine to the overall energy expenses of the rat, has been found to be incorrect, since it incorporates the intrinsic error of interorgan proton transfer through bicarbonate. Liver and intestine produced high net bicarbonate balances in all situations tested, implying the elimination (by means of oxidative pathways, i.e. consuming additional oxygen) of high amounts of H+ generated with bicarbonate. The equivalence in energy output of the oxygen balances was then corrected for bicarbonate production to 11-54% lower values. 3. Intestine and liver consume a high proportion of available oxygen, about one-half in basal (fed or starved) conditions and about one-third after gavage, the intestine consumption being about 15% in all situations tested and the liver decreasing its oxygen consumption with gavage.

Plasma leptin turnover rates in lean and obese Zucker rats

Conscious female adult lean and obese Zucker rats were injected through the jugular vein with radioactive iodine-labeled murine leptin; in the ensuing 8 min, four blood samples were sequentially extracted from the carotid artery. The samples were used in a modified RIA for leptin, in which paired tubes received the same amount of either labeled or unlabeled leptin, thus allowing us to estimate both leptin levels and specific radioactivity. The data were used to determine the decay curve parameters from which the half-life of leptin (5.46 ± 0.23 min for lean rats and 6.99 ± 0.75 min for obese rats) as well as the size of its circulating pool (32 pmol/kg for lean rats and 267 pmol/kg for obese rats) and the overall degradation rate (96 fkat/kg for lean rats and 645 fkat/kg for obese rats) were estimated. These values are consistent with the hormonal role of leptin and the need for speedy changes in its levels in response to metabolic challenge.

Implementation of raltegravir in routine clinical practice: selection criteria for choosing this drug, virologic response rates, and characteristics of failures.

BACKGROUND: Raltegravir (RAL) achieved remarkable virologic suppression rates in randomized-clinical trials, but today efficacy data and factors for treatment failures in a routine clinical care setting are limited. METHODS: First, factors associated with a switch to RAL were identified with a logistic regression including patients from the Swiss HIV Cohort Study with a history of 3 class failure (n = 423). Second, predictors for virologic outcome were identified in an intent-to-treat analysis including all patients who received RAL. Last observation carried forward imputation was used to determine week 24 response rate (HIV-1 RNA >or= 50 copies/mL). RESULTS: The predominant factor associated with a switch to RAL in patients with suppressed baseline RNA was a regimen containing enfuvirtide [odds ratio 41.9 (95% confidence interval: 11.6-151.6)]. Efficacy analysis showed an overall response rate of 80.9% (152/188), whereas 71.8% (84/117) and 95.8% (68/71) showed viral suppression when stratified for detectable and undetectable RNA at baseline, respectively. Overall CD4 cell counts increased significantly by 42 cells/microL (P < 0.001). Characteristics of failures were a genotypic sensitivity score of the background regimen <or=1, very low RAL plasma concentrations, poor adherence, and high viral load at baseline. CONCLUSIONS: Virologic suppression rates in our routine clinical care setting were promising and comparable with data from previously published randomized-controlled trials.

Estimation of soil nitrate in the spring as a basis for adjustment of nitrogen fertiliser rates

Selostus: Maassa olevan nitraattitypen arviointi simulointimallin avulla

Trace element supplementation modulates pulmonary infection rates after major burns: a double-blind, placebo-controlled trial.

Infections remain the leading cause of death after major burns. Trace elements are involved in immunity and burn patients suffer acute trace element depletion after injury. In a previous nonrandomized study, trace element supplementation was associated with increased leukocyte counts and shortened hospital stays. This randomized, placebo-controlled trial studied clinical and immune effects of trace element supplements. Twenty patients, aged 40 +/- 16 y (mean +/- SD), burned on 48 +/- 17% of their body surfaces, were studied for 30 d after injury. They consumed either standard trace element intakes plus supplements (40.4 micromol Cu, 2.9 micromol Se, and 406 micromol Zn; group TE) or standard trace element intakes plus placebo (20 micromol Cu, 0.4 micromol Se, and 100 micromol Zn; group C) for 8 d. Demographic data were similar for both groups. Mean plasma copper and zinc concentrations were below normal until days 20 and 15, respectively (NS). Plasma selenium remained normal for group TE but decreased for group C (P < 0.05 on days 1 and 5). Total leukocyte counts tended to be higher in group TE because of higher neutrophil counts. Proliferation to mitogens was depressed compared with healthy control subjects (NS). The number of infections per patient was significantly (P < 0.05) lower in group TE (1.9 +/- 0.9) than in group C (3.1 +/- 1.1) because of fewer pulmonary infections. Early trace element supplementation appears beneficial after major burns; it was associated with a significant decrease in the number of bronchopneumonia infections and with a shorter hospital stay when data were normalized for burn size.

Fracture rates of IPS Empress all-ceramic crowns: a systematic review.

Purpose: The aim of this study was to evaluate the clinical fracture rate of crowns fabricated with the pressable, leucite-reinforced ceramic IPS Empress, and relate the results to the type of tooth restored. Materials and Methods: The database SCOPUS was searched for clinical studies involving full-coverage crowns made of IPS Empress. To assess the fracture rate of the crowns in relation to the type of restored tooth and study, Poisson regression analysis was used. Results: Seven clinical studies were identified involving 1,487 adhesively luted crowns (mean observation time: 4.5 +/- 1.7 years) and 81 crowns cemented with zinc-phosphate cement (mean observation time: 1.6 +/- 0.8 years). Fifty-seven of the adhesively luted crowns fractured (3.8%). The majority of fractures (62%) occurred between the third and sixth year after placement. There was no significant influence regarding the test center on fracture rate, but the restored tooth type played a significant role. The hazard rate (per year) for crowns was estimated to be 5 in every 1,000 crowns for incisors, 7 in every 1,000 crowns for premolars, 12 in every 1,000 crowns for canines, and 16 in every 1,000 crowns for molars. One molar crown in the zinc-phosphate group fractured after 1.2 years. Conclusion: Adhesively luted IPS Empress crowns showed a low fracture rate for incisors and premolars and a somewhat higher rate for molars and canines. The sample size of the conventionally luted crowns was too small and the observation period too short to draw meaningful conclusions. Int J Prosthodont 2010;23:129-133.

Stratigraphie, morphodynamique, paléoenvironnements des terrains sédimentaires meubles à forte déclivité du domaine périglaciaire alpin

Dans le contexte d'un climat de plus en plus chaud, une étude « géosystémique » de la répartition du pergélisol dans l'ensemble d'un versant périglaciaire alpin, de la paroi rocheuse jusqu'au glacier rocheux, s'avère primordiale. S'insérant dans cette problématique, ce travail de thèse vise comme objectif général l'étude des versants d'éboulis situés à l'intérieur de la ceinture du pergélisol discontinu selon deux volets de recherche différents : une étude de la stratigraphie et de la répartition du pergélisol dans les éboulis de haute altitude et des processus qui lui sont associés ; une reconstitution de l'histoire paléoenvironnementale du domaine périglaciaire alpin pendant le Tardiglaciaire et l'Holocène. La stratigraphie et la répartition spatiale du pergélisol a été étudiée dans cinq éboulis des Alpes Valaisannes (Suisse), dont trois ont fait l'objet de forages profonds, grâce à la prospection géophysique de détail effectuée à l'aide de méthodes thermiques, de résistivité, sismiques et nucléaires. Les mesures effectuées ont permis de mettre en évidence que, dans les cinq éboulis étudiés, la répartition du pergélisol est discontinue et aucun des versants n'est intégralement occupé par du pergélisol. En particulier, il a été possible de prouver de manière directe que, dans un éboulis, le pergélisol est présent dans les parties inférieures du versant et absent dans les parties supérieures. Trois facteurs de contrôle principaux de la répartition du pergélisol déterminée au sein des éboulis étudiés ont été individualisés, pouvant agir seuls ou de manière combinée : la ventilation ascendante, l'augmentation de la granulométrie en direction de l'aval et la redistribution de la neige par le vent et les avalanches. Parmi ceux-ci, la relation ventilation-granulométrie semble être le facteur de contrôle principal permettant d'expliquer la présence de pergélisol dans les parties inférieures d'un éboulis et son absence dans les parties supérieures. Enfin, l'analyse de la structure des éboulis périglaciaires de haute altitude a permis de montrer que la stratigraphie du pergélisol peut être un élément important pour l'interprétation de la signification paléoclimatique de ce type de formes. Pour le deuxième volet de la recherche, grâce aux datations relatives effectuées à l'aide de l'utilisation conjointe de la méthode paléogéographique et du marteau de Schmidt, il a été possible de définir la chrono-stratigraphie du retrait glaciaire et du développement des glaciers rocheux et des versants d'éboulis des quatre régions des Alpes suisses étudiées (régions du Mont Gelé - Mont Fort, des Fontanesses et de Chamosentse, dans les Alpes Valaisannes, et Massif de la Cima di Gana Bianca, dans les Alpes Tessinoises). La compilation de toutes les datations effectuées a permis de montrer que la plupart des glaciers rocheux actifs étudiés se seraient développés soit juste avant et/ou pendant l'Optimum Climatique Holocène de 9.5-6.3 ka cal BP, soit au plus tard juste après cet évènement climatique majeur du dernier interglaciaire. Parmi les glaciers rocheux fossiles datés, la plupart aurait commencé à se former dans la deuxième moitié du Tardiglaciaire et se serait inactivé dans la première partie de l'Optimum Climatique Holocène. Pour les éboulis étudiés, les datations effectuées ont permis d'observer que leur surface date de la période entre le Boréal et l'Atlantique récent, indiquant que les taux d'éboulisation après la fin de l'Optimum Climatique Holocène ont dû être faibles, et que l'intervalle entre l'âge maximal et l'âge minimal est dans la plupart des cas relativement court (4-6 millénaires), indiquant que les taux d'éboulisation durant la période de formation des éboulis ont dû être importants. Grâce au calcul des taux d'érosion des parois rocheuses sur la base du volume de matériaux rocheux pour quatre des éboulis étudiés, il a été possible mettre en évidence l'existence d'une « éboulisation parapériglaciaire » liée à la dégradation du pergélisol dans les parois rocheuses, fonctionnant principalement durant les périodes de réchauffement climatique rapide comme cela a été le cas au début du Bølling, du Préboréal à la fin de l'Atlantique récent et, peut-être, à partir des années 1980. - In the context of a warmer climate, a « geosystemical » study of the permafrost distribution in a whole alpine periglacial hillslope, from the rockwall to the rockglacier, is of great importance. With respect to this problem, the general objective of this PhD thesis is the global study of talus slopes located within the alpine periglacial belt following two different research axes: the analysis of the internal structure and of the permafrost distribution of high altitude talus slopes and of the related processes; the reconstruction of the palaeoenvironmental history of the alpine periglacial belt during the Lateglacial and the Holocene. The stratigraphy and the permafrost distribution were studied in five talus slopes of the Valais Alps (Switzerland) with the analysis of borehole data (on three of the five talus slopes) and other methods of permafrost prospecting: Electrical Resistivity Tomography (ERT), Refraction Seismic Tomography (RST) and nuclear well logging. The collected data shows that, in all of the studied talus slopes, permafrost distribution is discontinuous and that neither of the hillslopes is integrally characterised by permafrost. In particular, this data proves by direct investigations that, in talus slopes, permafrost is present in the lower parts of the hillslope, whereas it is absent in the upper parts. Permafrost distribution in alpine talus slopes is depending of the combination of almost three controlling factors, whose respective importance is variable: the chimney effect, the increase of grain size downslope and the redistribution of snow by avalanches. Depending on the size of the talus and on topographical and geomorphological heterogeneities, various cases are possible: one dominant controlling factor or the combination of various factors. Nevertheless, it would be an error to consider each controlling factor independently, without considering their relationships. Between these controlling factors, the relationship chimney effect/grain size seems to be the most important factor controlling the presence of permafrost in the lowest part of periglacial talus slopes, and its absence in the upper parts. Finally, the analysis of the talus structure shows that the permafrost stratigraphy may be an important element of interpretation of the palaeoclimatic significance of an alpine talus slope. The second research axe focused on the establishment of a chronology of the Lateglacial glacier retreat and the dating of rockglaciers and talus slopes development in four studied regions of the Swiss Alps (Mont Gelé - Mont Fort, Fontanesses and Chamosentse regions, in the Valais Alps, and the Cima di Gana Bianca Massif, in the Ticino Alps). The compilation of the dates acquired through the combination of the palaeogeographical method and of the Schmidt hammer indicates that most of the investigated active rockglaciers started to evolve during the early phases of the Holocene or, at the latest, after the early-to-mid Holocene Climatic Optimum (ending around 6.3 ka cal BP). For the dated relict rockglaciers, most of them started to evolve in the second half of the Lateglacial, and probably became inactive at the beginning of the Holocene Climatic Optimum. For the investigated talus slopes, the relative dating carried out allowed to show that their surface date from the period included between the Boreal and the end of the Atlantic, pointing out that the rockwall retreat after the end of the Holocene Climatic Optimum was weak, and that the interval between maximal and minimal ages is in most cases relatively short (4-6 millennia). Therefore, the rockwall retreat during the development period of the talus slopes must has been considerable. Thanks to the calculation of rockwall erosion rates based on the volume of talus accumulations for four of the investigated hillslopes, it was possible to find evidences of the existence of "paraperiglacial rockfall phases" related to the permafrost degradation in rockwalls. These phases coincide with rapid climate warming periods, as at the beginning of the Bølling, during the Preboreal or, maybe, since 1980.

Les cegueses del coneixement: l'error i la il·lusió a partir de "El cónsul de sodoma"

Podeu consultar el document complet de la "XVI Setmana de Cinema Formatiu" a: http://hdl.handle.net/2445/22523

Subjetividad humana y "error" en la filosofía cartesiana

El presente trabajo, continuando la línea investigadora acerca de las nociones derazón, conciencia y subjetividad en Descartes, tal como se ha defendido en otros artículos ya publicados, aporta un nuevo argumento a una línea de trabajo previamente iniciada, poniendo de relieve que el problema gnoseológico del error viene condicionado por la misma noción cartesiana de racionalidad, y que ésta dista mucho de lo que tradicionalmente se ha entendido como una racionalidad abstracta y formal, libre de los imperativos humanos. Por otro lado, y a la inversa, también se intenta mostrar como el hecho del error contribuye, cartesianamente hablando, a definir un modelo de racionalidad profundamentehumanizada. El artículo, tras una introducción, se propone analizar las relaciones entre los conceptos básicos de racionalidad, dogma, y naturaleza, lo que permitirá a continuación dejar constancia de la copertenencia entre racionalidad y error, para acabar viendo como la libertad humana es la vez, y para ambos, su fundamento último.

Reproducibility of Fatmax and Fat Oxidation Rates during Exercise in Recreationally Trained Males.

Aerobic exercise training performed at the intensity eliciting maximal fat oxidation (Fatmax) has been shown to improve the metabolic profile of obese patients. However, limited information is available on the reproducibility of Fatmax and related physiological measures. The aim of this study was to assess the intra-individual variability of: a) Fatmax measurements determined using three different data analysis approaches and b) fat and carbohydrate oxidation rates at rest and at each stage of an individualized graded test. Fifteen healthy males [body mass index 23.1±0.6 kg/m2, maximal oxygen consumption ([Formula: see text]) 52.0±2.0 ml/kg/min] completed a maximal test and two identical submaximal incremental tests on ergocycle (30-min rest followed by 5-min stages with increments of 7.5% of the maximal power output). Fat and carbohydrate oxidation rates were determined using indirect calorimetry. Fatmax was determined with three approaches: the sine model (SIN), measured values (MV) and 3rd polynomial curve (P3). Intra-individual coefficients of variation (CVs) and limits of agreement were calculated. CV for Fatmax determined with SIN was 16.4% and tended to be lower than with P3 and MV (18.6% and 20.8%, respectively). Limits of agreement for Fatmax were -2±27% of [Formula: see text] with SIN, -4±32 with P3 and -4±28 with MV. CVs of oxygen uptake, carbon dioxide production and respiratory exchange rate were <10% at rest and <5% during exercise. Conversely, CVs of fat oxidation rates (20% at rest and 24-49% during exercise) and carbohydrate oxidation rates (33.5% at rest, 8.5-12.9% during exercise) were higher. The intra-individual variability of Fatmax and fat oxidation rates was high (CV>15%), regardless of the data analysis approach employed. Further research on the determinants of the variability of Fatmax and fat oxidation rates is required.