Bootstrapping pairs in Distance-Based Regression

La regressió basada en distàncies és un mètode de predicció que consisteix en dos passos: a partir de les distàncies entre observacions obtenim les variables latents, les quals passen a ser els regressors en un model lineal de mínims quadrats ordinaris. Les distàncies les calculem a partir dels predictors originals fent us d'una funció de dissimilaritats adequada. Donat que, en general, els regressors estan relacionats de manera no lineal amb la resposta, la seva selecció amb el test F usual no és possible. En aquest treball proposem una solució a aquest problema de selecció de predictors definint tests estadístics generalitzats i adaptant un mètode de bootstrap no paramètric per a l'estimació dels p-valors. Incluim un exemple numèric amb dades de l'assegurança d'automòbils.

Geomodelling of a fluvial system with semi-supervised support vector regression

Fluvial deposits are a challenge for modelling flow in sub-surface reservoirs. Connectivity and continuity of permeable bodies have a major impact on fluid flow in porous media. Contemporary object-based and multipoint statistics methods face a problem of robust representation of connected structures. An alternative approach to model petrophysical properties is based on machine learning algorithm ? Support Vector Regression (SVR). Semi-supervised SVR is able to establish spatial connectivity taking into account the prior knowledge on natural similarities. SVR as a learning algorithm is robust to noise and captures dependencies from all available data. Semi-supervised SVR applied to a synthetic fluvial reservoir demonstrated robust results, which are well matched to the flow performance

Predictors of poor response to methotrexate in polyarticular-course juvenile idiopathic arthritis: analysis of the PRINTO methotrexate trial.

OBJECTIVES: To determine whether baseline demographic, clinical, articular and laboratory variables predict methotrexate (MTX) poor response in polyarticular-course juvenile idiopathic arthritis. METHODS: Patients newly treated for 6 months with MTX enrolled in the Paediatric Rheumatology International Trials Organization (PRINTO) MTX trial. Bivariate and logistic regression analyses were used to identify baseline predictors of poor response according to the American College of Rheumatology pediatric (ACR-ped) 30 and 70 criteria. RESULTS: In all, 405/563 (71.9%) of patients were women; median age at onset and disease duration were 4.3 and 1.4 years, respectively, with anti-nuclear antibody (ANA) detected in 259/537 (48.2%) patients. With multivariate logistic regression analysis, the most important determinants of ACR-ped 70 non-responders were: disease duration > 1.3 years (OR 1.93), ANA negativity (OR 1.77), Childhood Health Assessment Questionnaire (CHAQ) disability index > 1.125 (OR 1.65) and the presence of right and left wrist activity (OR 1.55). Predictors of ACR-ped 30 non-responders were: ANA negativity (OR 1.92), CHAQ disability index > 1.14 (OR 2.18) and a parent's evaluation of child's overall well-being < or = 4.69 (OR 2.2). CONCLUSION: The subgroup of patients with longer disease duration, ANA negativity, higher disability and presence of wrist activity were significantly associated with a poorer response to a 6-month MTX course.

Irrigation with domestic wastewater: a multivariate analysis of main soil changes

Irrigation with treated domestic sewage wastewater (TSE) is an agricultural practice to reduce water requirements of agroecossystems and the nutrient load impact on freshwaters, but adverse effects on soil chemical (salinization, sodification, etc.) and soil physical properties (alteration in soil porosity and hydraulic conductivity, etc.) have been reported. This study aimed to define some relationships among these changes in an Oxisol using multivariate analysis. Corn (Zea mays L.) and sunflower (Helianthus annuus L.) were grown for two years, irrigated with TSE. The following soil properties were determined: Ca2+; Mg2+; Na+; K+ and H + Al contents, cationic exchangeable capacity (CEC), sum of bases (SB), base saturation (V), texture (sand, silt and clay), macro-, micro-, and cryptoporosity (V MA, V MI and V CRI), water content at soil saturation (θS) and at field capacity (θFC), residual water content (θR), soil bulk density (d s), water dispersed clay (WDC) and saturated hydraulic conductivity (K SAT). Factor analysis revealed the following six principal factors: Fine Porosity (composed of Na+; K+; WDC, θR, θRFC, and V CRI); Large Porosity (θS, d s, V MA, Vs); Soil CEC (Ca2+; Mg2+; CEC, SB, V); Soil Acidity (H + Al); and Soil Texture (factors 5 and 6). A dual pore structure appears clearly to the factors 1 and 2, with an apparent relationship between fine porosity and the monovalent cations Na+ and K+. The irrigation (with potable sodic tap water or sewage wastewater) only had a significant effect on Fine Porosity and Large Porosity factors, while factors 3 and 4 (Soil CEC and Soil Acidity) were correlated with soil depth. The main conclusion was a shift in pore distribution (large to fine pores) during irrigation with TSE, which induces an increase of water storage and reduces the capacity of drainage of salts.

Comment on "Reappraisal of experimental values of third-order elastic constants of some cubic semiconductors and metals"

In a recent paper A. S. Johal and D. J. Dunstan [Phys. Rev. B 73, 024106 (2006)] have applied multivariate linear regression analysis to the published data of the change in ultrasonic velocity with applied stress. The aim is to obtain the best estimates for the third-order elastic constants in cubic materials. From such an analysis they conclude that uniaxial stress data on metals turns out to be nearly useless by itself. The purpose of this comment is to point out that by a proper analysis of uniaxial stress data it is possible to obtain reliable values of third-order elastic constants in cubic metals and alloys. Cu-based shape memory alloys are used as an illustrative example.

Modelling non-industrial private forest landowners' strategic decision making by using logistic regression and neural networks: case of predicting the choice of forest taxation basis

Summary

Family history of cancer and the risk of laryngeal cancer: a case-control study from Italy and Switzerland.

Only limited data is available on the relationship between family history of laryngeal and other neoplasms and laryngeal cancer risk. We investigated the issue using data from a multicentre case-control study conducted in Italy and Switzerland between 1992 and 2009 including 852 cases with histologically confirmed laryngeal cancer and 1970 controls admitted to hospital for acute, non neoplastic conditions. Unconditional logistic regression models adjusted for age, sex, study center, education, tobacco smoking, alcohol drinking and number of siblings were used to estimate the odds ratios (ORs) of laryngeal cancer. The multivariate OR was 2.8 (95% confidence interval [CI], 1.5-5.3) in subjects reporting a first-degree relative with laryngeal cancer, as compared to subjects with no family history. The OR was higher when the relative was diagnosed before 60 years of age (OR = 3.5, 95% CI 1.4-8.8). As compared to subjects without family history, non-smokers, and moderate drinkers, the OR was 37.1 (95% CI 9.9-139.4) for current smokers, heavy drinkers, with family history of laryngeal cancer. Family history of colorectal (OR = 1.5, 95% CI 1.0-2.3) and kidney (OR = 3.8, 95% CI 1.2-12.1) cancer were also associated to an increased risk of laryngeal cancer, while no significant increase in risk was found for family history of cancer at all sites, excluding the larynx (OR = 1.1).

Multivariate analysis and geostatistics of the fertility of a humic rhodic hapludox under coffee cultivation

The spatial variability of soil and plant properties exerts great influence on the yeld of agricultural crops. This study analyzed the spatial variability of the fertility of a Humic Rhodic Hapludox with Arabic coffee, using principal component analysis, cluster analysis and geostatistics in combination. The experiment was carried out in an area under Coffea arabica L., variety Catucai 20/15 - 479. The soil was sampled at a depth 0.20 m, at 50 points of a sampling grid. The following chemical properties were determined: P, K+, Ca2+, Mg2+, Na+, S, Al3+, pH, H + Al, SB, t, T, V, m, OM, Na saturation index (SSI), remaining phosphorus (P-rem), and micronutrients (Zn, Fe, Mn, Cu and B). The data were analyzed with descriptive statistics, followed by principal component and cluster analyses. Geostatistics were used to check and quantify the degree of spatial dependence of properties, represented by principal components. The principal component analysis allowed a dimensional reduction of the problem, providing interpretable components, with little information loss. Despite the characteristic information loss of principal component analysis, the combination of this technique with geostatistical analysis was efficient for the quantification and determination of the structure of spatial dependence of soil fertility. In general, the availability of soil mineral nutrients was low and the levels of acidity and exchangeable Al were high.

Impact of targeted antifungal prophylaxis in heart transplant recipients at high risk for early invasive fungal infection.

BACKGROUND: Invasive fungal infection (IFI) is associated with high mortality after heart transplantation (HTx). After two undiagnosed fatal cases of early disseminated fungal infections in our heart transplant program, a retrospective analysis was conducted to identify risk factors for the development of IFI and implement a new antifungal prophylaxis policy. METHODS: Clinical characteristics of HTx recipients hospitalized in our center (2004-2010) were recorded (Period 1), and risk factors associated with IFI were investigated using Cox regression analysis. From October 2010 to October 2012 (Period 2), targeted caspofungin prophylaxis was administered to all recipients at high risk for IFI, based on the results of the Period 1 analysis. RESULTS: During Period 1, 10% (6/59) of the patients developed IFI at a median onset of 9 days after transplantation. By multivariate analysis, the use of posttransplant extracorporeal membrane oxygenation (ECMO) was the strongest predictor for fungal infection (OR, 29.93; 95% CI, 1.51-592.57, P=0.03), whereas renal replacement therapy (RRT) and Aspergillus colonization were significant predictors only by univariate analysis. During Period 2, only 4% (1/26) of the patients developed IFI. In patients at high risk for IFI, antifungal prophylaxis was administered to 17% (4/23) in Period 1 versus 100% (13/13) in Period 2 (P<0.01). By survival analysis, antifungal prophylaxis was associated with a reduction in 90-day IFI incidence (HR, 0.14; 95% CI, 0.03-0.84, P=0.03) and 30-day mortality (HR, 0.25; 95% CI, 0.09-0.8, P=0.02). CONCLUSION: Extracorporeal membrane oxygenation was identified an important risk factor for IFI after HTx, and its use may require targeted administration of antifungal prophylaxis in the immediate posttransplant period.

Combination of MRI and dynamic FET PET for initial glioma grading.

AIM: MRI and PET with 18F-fluoro-ethyl-tyrosine (FET) have been increasingly used to evaluate patients with gliomas. Our purpose was to assess the additive value of MR spectroscopy (MRS), diffusion imaging and dynamic FET-PET for glioma grading. PATIENTS, METHODS: 38 patients (42 ± 15 aged, F/M: 0.46) with untreated histologically proven brain gliomas were included. All underwent conventional MRI, MRS, diffusion sequences, and FET-PET within 3±4 weeks. Performances of tumour FET time-activity-curve, early-to-middle SUVmax ratio, choline / creatine ratio and ADC histogram distribution pattern for gliomas grading were assessed, as compared to histology. Combination of these parameters and respective odds were also evaluated. RESULTS: Tumour time-activity-curve reached the best accuracy (67%) when taken alone to distinguish between low and high-grade gliomas, followed by ADC histogram analysis (65%). Combination of time-activity-curve and ADC histogram analysis improved the sensitivity from 67% to 86% and the specificity from 63-67% to 100% (p < 0.008). On multivariate logistic regression analysis, negative slope of the tumour FET time-activity-curve however remains the best predictor of high-grade glioma (odds 7.6, SE 6.8, p = 0.022). CONCLUSION: Combination of dynamic FET-PET and diffusion MRI reached good performance for gliomas grading. The use of FET-PET/MR may be highly relevant in the initial assessment of primary brain tumours.

Retrospective feasibility study of simultaneous integrated boost in cervical cancer using Tomotherapy: the impact of organ motion and tumor regression.

BACKGROUND: Whole pelvis intensity modulated radiotherapy (IMRT) is increasingly being used to treat cervical cancer aiming to reduce side effects. Encouraged by this, some groups have proposed the use of simultaneous integrated boost (SIB) to target the tumor, either to get a higher tumoricidal effect or to replace brachytherapy. Nevertheless, physiological organ movement and rapid tumor regression throughout treatment might substantially reduce any benefit of this approach. PURPOSE: To evaluate the clinical target volume - simultaneous integrated boost (CTV-SIB) regression and motion during chemo-radiotherapy (CRT) for cervical cancer, and to monitor treatment progress dosimetrically and volumetrically to ensure treatment goals are met. METHODS AND MATERIALS: Ten patients treated with standard doses of CRT and brachytherapy were retrospectively re-planned using a helical Tomotherapy - SIB technique for the hypothetical scenario of this feasibility study. Target and organs at risk (OAR) were contoured on deformable fused planning-computed tomography and megavoltage computed tomography images. The CTV-SIB volume regression was determined. The center of mass (CM) was used to evaluate the degree of motion. The Dice's similarity coefficient (DSC) was used to assess the spatial overlap of CTV-SIBs between scans. A cumulative dose-volume histogram modeled estimated delivered doses. RESULTS: The CTV-SIB relative reduction was between 31 and 70%. The mean maximum CM change was 12.5, 9, and 3 mm in the superior-inferior, antero-posterior, and right-left dimensions, respectively. The CTV-SIB-DSC approached 1 in the first week of treatment, indicating almost perfect overlap. CTV-SIB-DSC regressed linearly during therapy, and by the end of treatment was 0.5, indicating 50% discordance. Two patients received less than 95% of the prescribed dose. Much higher doses to the OAR were observed. A multiple regression analysis showed a significant interaction between CTV-SIB reduction and OAR dose increase. CONCLUSIONS: The CTV-SIB had important regression and motion during CRT, receiving lower therapeutic doses than expected. The OAR had unpredictable shifts and received higher doses. The use of SIB without frequent adaptation of the treatment plan exposes cervical cancer patients to an unpredictable risk of under-dosing the target and/or overdosing adjacent critical structures. In that scenario, brachytherapy continues to be the gold standard approach.

Multivariate analysis applied to the study of the relationship between soil and plant properties in a peach orchard

In the State of Rio Grande do Sul, the municipality of Pelotas is responsible for 90 % of peach production due to its suitable climate and soil conditions. However, there is the need for new studies that aim at improved fruit quality and increased yield. The aim of this study was to evaluate the relationship that exists between soil physical properties and properties in the peach plant in the years 2010 and 2011 by the technique of multivariate canonical correlation. The experiment was conducted in a peach orchard located in the municipality of Morro Redondo, RS, Brazil, where an experimental grid of 101 plants was established. In a trench dug beside each one of the 101 plants, soil samples were collected to determine silt, clay, and sand contents, soil density, total porosity, macroporosity, microporosity, and volumetric water content in the 0.00-0.10 and 0.10-0.20 m layers, as well as the depth of the A horizon. In each plant and in each year, the following properties were assessed: trunk diameter, fruit size and number of fruits per plant, average weight of the fruit per plant, fruit pulp firmness, Brix content, and yield from the orchard. Exploratory analysis of the data was undertaken by descriptive statistics, and the relationships between the physical properties of the soil and of the plant were assessed by canonical correlation analysis. The results showed that the clay and microporosity variables were those that exhibited the highest coefficients of canonical cross-loading with the plant properties in the soil layers assessed, and that the variable of mean weight of the fruit per plant was that which had the highest coefficients of canonical loading within the plant group for the two years assessed.

Multitask learning of environmental spatial data

The present research deals with an application of artificial neural networks for multitask learning from spatial environmental data. The real case study (sediments contamination of Geneva Lake) consists of 8 pollutants. There are different relationships between these variables, from linear correlations to strong nonlinear dependencies. The main idea is to construct a subsets of pollutants which can be efficiently modeled together within the multitask framework. The proposed two-step approach is based on: 1) the criterion of nonlinear predictability of each variable ?k? by analyzing all possible models composed from the rest of the variables by using a General Regression Neural Network (GRNN) as a model; 2) a multitask learning of the best model using multilayer perceptron and spatial predictions. The results of the study are analyzed using both machine learning and geostatistical tools.

Regression analysis of spatial data.

Many of the most interesting questions ecologists ask lead to analyses of spatial data. Yet, perhaps confused by the large number of statistical models and fitting methods available, many ecologists seem to believe this is best left to specialists. Here, we describe the issues that need consideration when analysing spatial data and illustrate these using simulation studies. Our comparative analysis involves using methods including generalized least squares, spatial filters, wavelet revised models, conditional autoregressive models and generalized additive mixed models to estimate regression coefficients from synthetic but realistic data sets, including some which violate standard regression assumptions. We assess the performance of each method using two measures and using statistical error rates for model selection. Methods that performed well included generalized least squares family of models and a Bayesian implementation of the conditional auto-regressive model. Ordinary least squares also performed adequately in the absence of model selection, but had poorly controlled Type I error rates and so did not show the improvements in performance under model selection when using the above methods. Removing large-scale spatial trends in the response led to poor performance. These are empirical results; hence extrapolation of these findings to other situations should be performed cautiously. Nevertheless, our simulation-based approach provides much stronger evidence for comparative analysis than assessments based on single or small numbers of data sets, and should be considered a necessary foundation for statements of this type in future.