Poly-bagging predictors for classification modelling for credit scoring

Credit scoring modelling comprises one of the leading formal tools for supporting the granting of credit. Its core objective consists of the generation of a score by means of which potential clients can be listed in the order of the probability of default. A critical factor is whether a credit scoring model is accurate enough in order to provide correct classification of the client as a good or bad payer. In this context the concept of bootstraping aggregating (bagging) arises. The basic idea is to generate multiple classifiers by obtaining the predicted values from the fitted models to several replicated datasets and then combining them into a single predictive classification in order to improve the classification accuracy. In this paper we propose a new bagging-type variant procedure, which we call poly-bagging, consisting of combining predictors over a succession of resamplings. The study is derived by credit scoring modelling. The proposed poly-bagging procedure was applied to some different artificial datasets and to a real granting of credit dataset up to three successions of resamplings. We observed better classification accuracy for the two-bagged and the three-bagged models for all considered setups. These results lead to a strong indication that the poly-bagging approach may promote improvement on the modelling performance measures, while keeping a flexible and straightforward bagging-type structure easy to implement. (C) 2011 Elsevier Ltd. All rights reserved.

Risk stratification in neonates and infants submitted to cardiac surgery with cardiopulmonary bypass: A multimarker approach combining inflammatory mediators, N-terminal pro-B-type natriuretic peptide and troponin I

Low cardiac output syndrome (LCOS) is a common problem following cardiac surgery with cardiopulmonary bypass (CPB) in neonates and infants, and its early recognition remains a challenging task. We aimed to test whether a multimarker approach combining inflammatory and cardiac markers provides complementary information for prediction of LCOS and death in children submitted to cardiac surgery with CPB. Forty-six children younger than 18 months with congenital heart defects were prospectively enrolled. No intervention was made. Blood samples were collected pre-operatively, during CPB and post-operatively (PO) for measurement of interleukin (IL)-6, IL-8, IL-10, tumor necrosis factor (TNF)-alpha, cardiac troponin I (cTnI) and N-terminal pro-B-type natriuretic peptide (NT-proBNP). Clinical data and outcome variables were recorded. Logistic regression was used to identify predictors of LCOS and death. Multivariate logistic regression identified pre-operative NT-proBNP and IL-8 4 h PO as independent predictors of LCOS, while cTnI 4 h PO and CPB length were independent predictors of death. The use of inflammatory and cardiac markers in combination improved sensitivity, negative predictive value and accuracy of the models. In conclusion, the combined assessment of inflammatory and cardiac biochemical markers can be useful for identifying young children at increased risk for LCOS and death after heart surgery with CPB. (C) 2008 Elsevier Ltd. All rights reserved.

Time trends and predictors of mortality from ill-defined causes in old age: 9 year folllow-up of the Bambuí cohort study (Brazil)

The study objective was to examine differentials in time trends and predictors of deaths assigned to symptoms, signs and ill-defined conditions in comparison with other ill-defined conditions (ill-defined cardiovascular diseases, cancer and injury) in a population-based cohort study. Of 1,606 baseline participants aged 60 years and over, 524 died during 9-year follow-up and were included in this study. Deaths coded to "symptoms" declined by 77% in the period from 1997-1999 to 2003-2005. Deaths coded to other ill-defined conditions remained unchanged. The calendar period 2003-2005 (RR = 0.25; 95%CI: 0.09-0.70) and in-hospital deaths (RR = 0.16; 95%CI: 0.08-0.34) were independently associated with "symptoms", but not with other ill-defined conditions. Baseline socio-demographic characteristics and chronic diseases were not predictors of these outcomes. International and national agencies have focused on the reduction of deaths assigned to "symptoms" to improve the registration of vital statistics, while other ill-defined conditions have received little attention. Our data provide evidence supporting the need to redress this situation.

Influence of self-perceived oral health and socioeconomic predictors on the utilization of dental care services by schoolchildren

The influence of socioeconomic factors and self-rated oral health on children's dental health assistance was assessed. This study followed a cross-sectional design, with a multistage random sample of 792 12-year-old schoolchildren from Santa Maria, a city in southern Brazil. A dental examination provided information on the prevalence of dental caries (DMFT index). Data about the use of dental service, socioeconomic status, and self-perceived oral health were collected by means of structured interviews. These associations were assessed using Poisson regression models (prevalence ratio; 95% confidence interval). The prevalence of regular use of dental service was 47.8%. Children from low socioeconomic backgrounds and those who rated their oral health as "poor" used the service less frequently. The distribution of the kind of oral healthcare assistance used (public/private) varied across socioeconomic groups. The better-off children were less likely to have used the public service. Clinical, socioeconomic, and psychosocial factors were strong predictors for the utilization of dental care services by schoolchildren.

Combining Legendre's Polynomials and Genetic Algorithm in the Solution of Nonlinear Initial-Value Problems

This work develops a method for solving ordinary differential equations, that is, initial-value problems, with solutions approximated by using Legendre's polynomials. An iterative procedure for the adjustment of the polynomial coefficients is developed, based on the genetic algorithm. This procedure is applied to several examples providing comparisons between its results and the best polynomial fitting when numerical solutions by the traditional Runge-Kutta or Adams methods are available. The resulting algorithm provides reliable solutions even if the numerical solutions are not available, that is, when the mass matrix is singular or the equation produces unstable running processes.

The variability and predictors of quality of AIDS care services in Brazil

Background: Since establishing universal free access to antiretroviral therapy in 1996, the Brazilian Health System has increased the number of centers providing HIV/AIDS outpatient care from 33 to 540. There had been no formal monitoring of the quality of these services until a survey of 336 AIDS health centers across 7 Brazilian states was undertaken in 2002. Managers of the services were asked to assess their clinics according to parameters of service inputs and service delivery processes. This report analyzes the survey results and identifies predictors of the overall quality of service delivery. Methods: The survey involved completion of a multiple-choice questionnaire comprising 107 parameters of service inputs and processes of delivering care, with responses assessed according to their likely impact on service quality using a 3-point scale. K-means clustering was used to group these services according to their scored responses. Logistic regression analysis was performed to identify predictors of high service quality. Results: The questionnaire was completed by 95.8% (322) of the managers of the sites surveyed. Most sites scored about 50% of the benchmark expectation. K-means clustering analysis identified four quality levels within which services could be grouped: 76 services (24%) were classed as level 1 (best), 53 (16%) as level 2 (medium), 113 (35%) as level 3 (poor), and 80 (25%) as level 4 (very poor). Parameters of service delivery processes were more important than those relating to service inputs for determining the quality classification. Predictors of quality services included larger care sites, specialization for HIV/AIDS, and location within large municipalities. Conclusion: The survey demonstrated highly variable levels of HIV/AIDS service quality across the sites. Many sites were found to have deficiencies in the processes of service delivery processes that could benefit from quality improvement initiatives. These findings could have implications for how HIV/AIDS services are planned in Brazil to achieve quality standards, such as for where service sites should be located, their size and staffing requirements. A set of service delivery indicators has been identified that could be used for routine monitoring of HIV/AIDS service delivery for HIV/AIDS in Brazil (and potentially in other similar settings).

Predictors of HBeAg status and hepatitis B viraemia in HIV-infected patients with chronic hepatitis B in the HAART era in Brazil

Background: HBV-HIV co-infection is associated with an increased liver-related morbidity and mortality. However, little is known about the natural history of chronic hepatitis B in HIV-infected individuals under highly active antiretroviral therapy (HAART) receiving at least one of the two drugs that also affect HBV (TDF and LAM). Information about HBeAg status and HBV viremia in HIV/HBV co-infected patients is scarce. The objective of this study was to search for clinical and virological variables associated with HBeAg status and HBV viremia in patients of an HIV/HBV co-infected cohort. Methods: A retrospective cross-sectional study was performed, of HBsAg-positive HIV-infected patients in treatment between 1994 and 2007 in two AIDS outpatient clinics located in the Sao Paulo metropolitan area, Brazil. The baseline data were age, sex, CD4 T+ cell count, ALT level, HIV and HBV viral load, HBV genotype, and duration of antiretroviral use. The variables associated to HBeAg status and HBV viremia were assessed using logistic regression. Results: A total of 86 HBsAg patients were included in the study. Of these, 48 (56%) were using combination therapy that included lamivudine (LAM) and tenofovir (TDF), 31 (36%) were using LAM monotherapy, and 7 patients had no previous use of either one. Duration of use of TDF and LAM varied from 4 to 21 and 7 to 144 months, respectively. A total of 42 (48. 9%) patients were HBeAg positive and 44 (51. 1%) were HBeAg negative. The multivariate analysis revealed that the use of TDF for longer than 12 months was associated with undetectable HBV DNA viral load (serum HBV DNA level < 60 UI/ml) (p = 0. 047). HBeAg positivity was associated with HBV DNA > 60 UI/ml (p = 0. 001) and ALT levels above normality (p = 0. 038). Conclusion: Prolonged use of TDF containing HAART is associated with undetectable HBV DNA viral load. HBeAg positivity is associated with HBV viremia and increased ALT levels.

Incidence and predictors of severe liver fibrosis in HIV-infected patients with chronic hepatitis C in Brazil

The aim of this study was to examine the incidence and factors associated with the severity of liver fibrosis in 234 coinfected patients in Brazil. Patients were cared for in our clinic, from 1996 to 2004. Eligible patients were defined as patients with documented HIV and hepatitis C virus (HCV) infections and had previously undergone a liver biopsy. Patients with persistently normal alanine aminotransferase (ALT) were also included. The variables selected for study were age, gender, risk category, history of high alcohol consumption, CD4(+) T cell count, antiretroviral therapy usage, HCV genotype and duration of HCV infection. Stage of fibrosis was scored as follows: F0, no fibrosis; F1, portal fibrosis with no septa; F2, portal fibrosis with few septa; F3, bridging fibrosis with many septa; and F4, cirrhosis. The liver fibrosis stage was F3 in 39 (16.6%) and F4 in 20(8.5%) patients. Among patients with normal ALT, the liver fibrosis stage was F3-F4 in three patients (5.6%). Predictors of severe liver fibrosis (17344) by multivariate analysis were age (older patients) and genotype 3 (genotype I odds ratio [OR], 0.28; 95% confidence interval [0], 0.12 0.65). In summary, in the present study severe liver fibrosis was found in 25% of our patients and was associated with an age of more than 38 years at the time of liver biopsy as well as, HCV genotype 3. No differences were found with respect to CD4(+) T cell counts although patients with a CD4(+) T cell count greater than 50 were excluded.

Research Smad2 and Smad6 as predictors of overall survival in oral squamous cell carcinoma patients

Background: To test if the expression of Smad1-8 mRNAs were predictive of survival in patients with oral squamous cell carcinoma (SCC). Patients and Methods: We analyzed, prospectively, the expression of Smad1-8, by means of Ribonuclease Protection Assay in 48 primary, operable, oral SCC. In addition, 21 larynx, 10 oropharynx and 4 hypopharynx SCC and 65 matched adjacent mucosa, available for study, were also included. For survival analysis, patients were categorized as positive or negative for each Smad, according to median mRNA expression. We also performed real-time quantitative PCR (QRTPCR) to asses the pattern of TGF beta 1, TGF beta 2, TGF beta 3 in oral SCC. Results: Our results showed that Smad2 and Smad6 mRNA expression were both associated with survival in Oral SCC patients. Cox Multivariate analysis revealed that Smad6 positivity and Smad2 negativity were both predictive of good prognosis for oral SCC patients, independent of lymph nodal status (P = 0.003 and P = 0.029, respectively). In addition, simultaneously Smad2(-) and Smad6(+) oral SCC group of patients did not reach median overall survival (mOS) whereas the mOS of Smad2(+)/Smad6(-) subgroup was 11.6 months (P = 0.004, univariate analysis). Regarding to TGF beta isoforms, we found that Smad2 mRNA and TGF beta 1 mRNA were inversely correlated (p = 0.05, R = -0.33), and that seven of the eight TGF beta 1(+) patients were Smad2(-). In larynx SCC, Smad7(-) patients did not reach mOS whereas mOS of Smad7(+) patients were only 7.0 months (P = 0.04). No other correlations were found among Smad expression, clinico-pathological characteristics and survival in oral, larynx, hypopharynx, oropharynx or the entire head and neck SCC population. Conclusion: Smad6 together with Smad2 may be prognostic factors, independent of nodal status in oral SCC after curative resection. The underlying mechanism which involves aberrant TGF beta signaling should be better clarified in the future.

Combining Monte Carlo simulation and density-functional theory to describe the spectral changes of Na(2) in liquid helium

Spectral changes of Na(2) in liquid helium were studied using the sequential Monte Carlo-quantum mechanics method. Configurations composed by Na(2) surrounded by explicit helium atoms sampled from the Monte Carlo simulation were submitted to time-dependent density-functional theory calculations of the electronic absorption spectrum using different functionals. Attention is given to both line shift and line broadening. The Perdew, Burke, and Ernzerhof (PBE1PBE, also known as PBE0) functional, with the PBE1PBE/6-311++G(2d,2p) basis set, gives the spectral shift, compared to gas phase, of 500 cm(-1) for the allowed X (1)Sigma(+)(g) -> B (1)Pi(u) transition, in very good agreement with the experimental value (700 cm(-1)). For comparison, cluster calculations were also performed and the first X (1)Sigma(+)(g) -> A (1)Sigma(+)(u) transition was also considered.

Study of the optical and magnetic properties of pyrimidine in water combining PCM and QM/MM methodologies

The solvent effects on the low-lying absorption spectrum and on the (15)N chemical shielding of pyrimidine in water are calculated using the combined and sequential Monte Carlo simulation and quantum mechanical calculations. Special attention is devoted to the solute polarization. This is included by an iterative procedure previously developed where the solute is electrostatically equilibrated with the solvent. In addition, we verify the simple yet unexplored alternative of combining the polarizable continuum model (PCM) and the hybrid QM/MM method. We use PCM to obtain the average solute polarization and include this in the MM part of the sequential QM/MM methodology, PCM-MM/QM. These procedures are compared and further used in the discrete and the explicit solvent models. The use of the PCM polarization implemented in the MM part seems to generate a very good description of the average solute polarization leading to very good results for the n-pi* excitation energy and the (15)N nuclear chemical shield of pyrimidine in aqueous environment. The best results obtained here using the solute pyrimidine surrounded by 28 explicit water molecules embedded in the electrostatic field of the remaining 472 molecules give the statistically converged values for the low lying n-pi* absorption transition in water of 36 900 +/- 100 (PCM polarization) and 36 950 +/- 100 cm(-1) (iterative polarization), in excellent agreement among one another and with the experimental value observed with a band maximum at 36 900 cm(-1). For the nuclear shielding (15)N the corresponding gas-water chemical shift obtained using the solute pyrimidine surrounded by 9 explicit water molecules embedded in the electrostatic field of the remaining 491 molecules give the statistically converged values of 24.4 +/- 0.8 and 28.5 +/- 0.8 ppm, compared with the inferred experimental value of 19 +/- 2 ppm. Considering the simplicity of the PCM over the iterative polarization this is an important aspect and the computational savings point to the possibility of dealing with larger solute molecules. This PCM-MM/QM approach reconciles the simplicity of the PCM model with the reliability of the combined QM/MM approaches.

Combining Mager and Steinmetz: The Effect of Grain Size and Maximum Induction on Hysteresis Energy Loss

Twelve samples with different grain sizes were prepared by normal grain growth and by primary recrystallization, and the hysteresis dissipated energy was measured by a quasi-static method. Results showed a linear relation between hysteresis energy loss and the inverse of grain size, which is here called Mager`s law, for maximum inductions from 0.6 to 1.5 T, and a Steinmetz power law relation between hysteresis loss and maximum induction for all samples. The combined effect is better described by a Mager`s law where the coefficients follow Steinmetz law.

Chemical composition of whole body and carcass of Bos indicus and tropically adapted Bos taurus breeds

Relationships between the chemical composition of the 9th- to 11th-rib section and the chemical composition of the carcass and empty body were evaluated for Bos indicus (108 Nellore and 36 Guzerah; GuS) and tropically adapted Bos taurus (56 Caracu; CaS) bulls, averaging 20 to 24 mo of age at slaughter. Nellore cattle were represented by 56 animals from the selected herd (NeS) and 52 animals from the control herd (NeC). The CaS and GuS bulls were from selected herds. Selected herds were based on 20 yr of selection for postweaning BW. Carcass composition was obtained after grinding, homogenizing, sampling, and analyzing soft tissue and bones. Similarly, empty body composition was obtained after grinding, homogenizing, sampling, analyzing, and combining blood, hide, head + feet, viscera, and carcass. Bulls were separated into 2 groups. Group 1 was composed of 36 NeS, 36 NeC, 36 CaS, and 36 GuS bulls and had water, ether extract (EE), protein, and ash chemically determined in the 9th- to 11th-rib section and in the carcass. Group 2 was composed of 20 NeS, 16 NeC, and 20 CaS bulls and water, EE, protein, and ash were determined in the 9th-to 11th-rib section, carcass, and empty body. Linear regressions were developed between the carcass and the 9th-to 11th-rib section compositions for group 1 and between carcass and empty body compositions for group 2. The 9th-to 11th-rib section percentages of water (RWt) and EE (RF) predicted the percentages of carcass water (CWt) and carcass fat (CF) with high precision: CWt, % = 29.0806 + 0.4873 x RWt, % (r(2) = 0.813, SE = 1.06) and CF, % = 10.4037 + 0.5179 x RF, % (r(2) = 0.863, SE = 1.26), respectively. Linear regressions between percentage of CWt and CF and empty body water (EBWt) and empty body fat (EBF) were also predicted with high precision: EBWt, % = -9.6821 + 1.1626 x CWt, % (r(2) = 0.878, SE = 1.43) and EBF, % = 0.3739 + 1.0386 x CF, % (r(2) = 0.982, SE = 0.65), respectively. Chemical composition of the 9th-to 11th-rib section precisely estimated carcass percentages of water and EE. These regressions can accurately predict carcass and empty body compositions for Nellore, Guzerah, and Caracu breeds.

Preliminary study of human breast tissue using synchrotron radiation combining WAXS and SAXS techniques

Using synchrotron radiation, we combined simultaneously wide angle X-ray scattering (WAXS) and small angle X-ray scattering (SAXS) techniques to obtain the scattering profiles of normal and neoplastic breast tissu-es samples at the momentum transfer range 6.28 nm(-1) <= Q(=4 pi.sin(theta/2)lambda) <= 50.26 nm(-1) and 0.15 nm(-1) <= Q <= 1.90 nm(-1), respectively. The results obtained show considerable differences between the scattering profiles of these tissues. We verified that the combination of some parameters (ratio between glandular and adipose peak intensity and third-order axial peak intensity) extracted from scattering profiles can be used for identifying breast cancer. (c) 2009 Elsevier Ltd. All rights reserved.

Myocardial scintigraphy and clinical stratification as predictors of events in renal transplant candidates

Background: We tested the hypothesis that the universal application of myocardial scanning with single-photon emission computed tomography (SPECT) would result in better risk stratification in renal transplant candidates (RTC) compared with SPECT being restricted to patients who, in addition to renal disease, had other clinical risk factors. Methods: RTCs (n=363) underwent SPECT and clinical risk stratification according to the American Society of Transplantation (AST) algorithm and were followed up until a major adverse cardiovascular event (MACE) or death. Results: Of the 363 patients, 79 patients (22%) had an abnormal SPECT scan and 270 (74%) were classified as high risk. Both methods correctly identified patients with increased probability of MACE. However, clinical stratification performed better (sensitivity and negative predictive value 99% and 99% vs. 25% and 87%, respectively). High-risk patients with an abnormal SPECT scan had a modest increased risk of events (log-rank = 0.03; hazard ratio [HR] = 1.37; 95% confidence interval [95% CI], 1.02-1.82). Eighty-six patients underwent coronary angiography, and coronary artery disease (CAD) was found in 60%. High-risk patients with CAD had an increased incidence of events (log-rank = 0.008; HR=3.85; 95% CI, 1.46-13.22), but in those with an abnormal SPECT scan, the incidence of events was not influenced by CAD (log-rank = 0.23). Forty-six patients died. Clinical stratification, but not SPECT, correlated with the probability of death (log-rank = 0.02; HR=3.25; 95% CI, 1.31-10.82). Conclusion: SPECT should be restricted to high-risk patients. Moreover, in contrast to SPECT, the AST algorithm was also useful for predicting death by any cause in RTCs and for selecting patients for invasive coronary testing.