Robust accelerated failure time regression

Robust estimators for accelerated failure time models with asymmetric (or symmetric) error distribution and censored observations are proposed. It is assumed that the error model belongs to a log-location-scale family of distributions and that the mean response is the parameter of interest. Since scale is a main component of mean, scale is not treated as a nuisance parameter. A three steps procedure is proposed. In the first step, an initial high breakdown point S estimate is computed. In the second step, observations that are unlikely under the estimated model are rejected or down weighted. Finally, a weighted maximum likelihood estimate is computed. To define the estimates, functions of censored residuals are replaced by their estimated conditional expectation given that the response is larger than the observed censored value. The rejection rule in the second step is based on an adaptive cut-off that, asymptotically, does not reject any observation when the data are generat ed according to the model. Therefore, the final estimate attains full efficiency at the model, with respect to the maximum likelihood estimate, while maintaining the breakdown point of the initial estimator. Asymptotic results are provided. The new procedure is evaluated with the help of Monte Carlo simulations. Two examples with real data are discussed.

Blocking hypoxia-induced autophagy in tumors restores cytotoxic T-cell activity and promotes regression.

The relationship between hypoxic stress, autophagy, and specific cell-mediated cytotoxicity remains unknown. This study shows that hypoxia-induced resistance of lung tumor to cytolytic T lymphocyte (CTL)-mediated lysis is associated with autophagy induction in target cells. In turn, this correlates with STAT3 phosphorylation on tyrosine 705 residue (pSTAT3) and HIF-1α accumulation. Inhibition of autophagy by siRNA targeting of either beclin1 or Atg5 resulted in impairment of pSTAT3 and restoration of hypoxic tumor cell susceptibility to CTL-mediated lysis. Furthermore, inhibition of pSTAT3 in hypoxic Atg5 or beclin1-targeted tumor cells was found to be associated with the inhibition Src kinase (pSrc). Autophagy-induced pSTAT3 and pSrc regulation seemed to involve the ubiquitin proteasome system and p62/SQSTM1. In vivo experiments using B16-F10 melanoma tumor cells indicated that depletion of beclin1 resulted in an inhibition of B16-F10 tumor growth and increased tumor apoptosis. Moreover, in vivo inhibition of autophagy by hydroxychloroquine in B16-F10 tumor-bearing mice and mice vaccinated with tyrosinase-related protein-2 peptide dramatically increased tumor growth inhibition. Collectively, this study establishes a novel functional link between hypoxia-induced autophagy and the regulation of antigen-specific T-cell lysis and points to a major role of autophagy in the control of in vivo tumor growth.

Looking for Validity or Testing It? The Perils of Stepwise Regression, Extreme-Scores Analysis, Heteroscedasticity, and Measurement Error

When researchers introduce a new test they have to demonstrate that it is valid, using unbiased designs and suitable statistical procedures. In this article we use Monte Carlo analyses to highlight how incorrect statistical procedures (i.e., stepwise regression, extreme scores analyses) or ignoring regression assumptions (e.g., heteroscedasticity) contribute to wrong validity estimates. Beyond these demonstrations, and as an example, we re-examined the results reported by Warwick, Nettelbeck, and Ward (2010) concerning the validity of the Ability Emotional Intelligence Measure (AEIM). Warwick et al. used the wrong statistical procedures to conclude that the AEIM was incrementally valid beyond intelligence and personality traits in predicting various outcomes. In our re-analysis, we found that the reliability-corrected multiple correlation of their measures with personality and intelligence was up to .69. Using robust statistical procedures and appropriate controls, we also found that the AEIM did not predict incremental variance in GPA, stress, loneliness, or well-being, demonstrating the importance for testing validity instead of looking for it.

Ensemble methods for environmental data modelling with support vector regression

This paper investigates the use of ensemble of predictors in order to improve the performance of spatial prediction methods. Support vector regression (SVR), a popular method from the field of statistical machine learning, is used. Several instances of SVR are combined using different data sampling schemes (bagging and boosting). Bagging shows good performance, and proves to be more computationally efficient than training a single SVR model while reducing error. Boosting, however, does not improve results on this specific problem.

Ebstein's anomaly: one and a half ventricular repair.

Patients with Ebstein's anomaly can present after childhood or adolescence with cyanosis, arrhythmias, severe right ventricular dysfunction and frequently with left ventricular dysfunction secondary to the prolonged cyanosis and to the right ventricular interference. At this point conventional repair is accompanied by elevated mortality and morbidity and poor functional results. We report our experience with three patients (8, 16 and 35 years of age) with Ebstein's anomaly, very dilated right atrium, severe tricuspid valve regurgitation (4/4), bi-directional shunt through an atrial septal defect and reduced left ventricular function (mean ejection fraction = 58%, mean shortening fraction = 25%). All underwent one and a half ventricular repair consisting of closure of the atrial septal defect, tricuspid repair with reduction of the atrialised portion of the right ventricle and end-to-side anastomosis of the superior vena cava to the right pulmonary artery. All patients survived, with a mean follow-up of 33 months. In all there was complete regression of the cyanosis and of the signs of heart failure. Postoperative echocardiography showed reduced degree of tricuspid regurgitation (2/4) and improvement of the left ventricular function (mean ejection fraction = 77%, mean shortening fraction = 40%). In patients with Ebstein's anomaly referred late for surgery with severely compromised right ventricular function or even with reduced biventricular function, the presence of a relatively hypoplastic and/or malfunctioning right ventricular chamber inadequate to sustain the entire systemic venous return but capable of managing part of the systemic venous return, permits a one and a half ventricular repair with good functional results.

Evaluation of a multicomponent worksite health promotion program for cardiovascular risk factors-correcting for the regression towards the mean effect.

BACKGROUND: We assessed the impact of a multicomponent worksite health promotion program for0 reducing cardiovascular risk factors (CVRF) with short intervention, adjusting for regression towards the mean (RTM) affecting such nonexperimental study without control group. METHODS: A cohort of 4,198 workers (aged 42 +/- 10 years, range 16-76 years, 27% women) were analyzed at 3.7-year interval and stratified by each CVRF risk category (low/medium/high blood pressure [BP], total cholesterol [TC], body mass index [BMI], and smoking) with RTM and secular trend adjustments. Intervention consisted of 15 min CVRF screening and individualized counseling by health professionals to medium- and high-risk individuals, with eventual physician referral. RESULTS: High-risk groups participants improved diastolic BP (-3.4 mm Hg [95%CI: -5.1, -1.7]) in 190 hypertensive patients, TC (-0.58 mmol/l [-0.71, -0.44]) in 693 hypercholesterolemic patients, and smoking (-3.1 cig/day [-3.9, -2.3]) in 808 smokers, while systolic BP changes reflected RTM. Low-risk individuals without counseling deteriorated TC and BMI. Body weight increased uniformly in all risk groups (+0.35 kg/year). CONCLUSIONS: In real-world conditions, short intervention program participants in high-risk groups for diastolic BP, TC, and smoking improved their CVRF, whereas low-risk TC and BMI groups deteriorated. Future programs may include specific advises to low-risk groups to maintain a favorable CVRF profile.

Comparison of 3D segmented gradient-echo and steady-state free precession coronary MRI sequences in patients with coronary artery disease.

OBJECTIVE: Our objective was to compare two state-of-the-art coronary MRI (CMRI) sequences with regard to image quality and diagnostic accuracy for the detection of coronary artery disease (CAD). SUBJECTS AND METHODS: Twenty patients with known CAD were examined with a navigator-gated and corrected free-breathing 3D segmented gradient-echo (turbo field-echo) CMRI sequence and a steady-state free precession sequence (balanced turbo field-echo). CMRI was performed in a transverse plane for the left coronary artery and a double-oblique plane for the right coronary artery system. Subjective image quality (1- to 4-point scale, with 1 indicating excellent quality) and objective image quality parameters were independently determined for both sequences. Sensitivity, specificity, and accuracy for the detection of significant (> or = 50% diameter) coronary artery stenoses were determined as defined in invasive catheter X-ray coronary angiography. RESULTS: Subjective image quality was superior for the balanced turbo field-echo approach (1.8 +/- 0.9 vs 2.3 +/- 1.0 for turbo field-echo; p < 0.001). Vessel sharpness, signal-to-noise ratio, and contrast-to-noise ratio were all superior for the balanced turbo field-echo approach (p < 0.01 for signal-to-noise ratio and contrast-to-noise ratio). Of the 103 segments, 18% of turbo field-echo segments and 9% of balanced turbo field-echo segments had to be excluded from disease evaluation because of insufficient image quality. Sensitivity, specificity, and accuracy for the detection of significant coronary artery stenoses in the evaluated segments were 92%, 67%, 85%, respectively, for turbo field-echo and 82%, 82%, 81%, respectively, for balanced turbo field-echo. CONCLUSION: Balanced turbo field-echo offers improved image quality with significantly fewer nondiagnostic segments when compared with turbo field-echo. For the detection of CAD, both sequences showed comparable accuracy for the visualized segments.

Childhood epilepsy with neuropsychological regression and continuous spike waves during sleep: epilepsy surgery in a young adult.

We describe the case of a man with a history of complex partial seizures and severe language, cognitive and behavioural regression during early childhood (3.5 years), who underwent epilepsy surgery at the age of 25 years. His early epilepsy had clinical and electroencephalogram features of the syndromes of epilepsy with continuous spike waves during sleep and acquired epileptic aphasia (Landau-Kleffner syndrome), which we considered initially to be of idiopathic origin. Seizures recurred at 19 years and presurgical investigations at 25 years showed a lateral frontal epileptic focus with spread to Broca's area and the frontal orbital regions. Histopathology revealed a focal cortical dysplasia, not visible on magnetic resonance imaging. The prolonged but reversible early regression and the residual neuropsychological disorders during adulthood were probably the result of an active left frontal epilepsy, which interfered with language and behaviour during development. Our findings raise the question of the role of focal cortical dysplasia as an aetiology in the syndromes of epilepsy with continuous spike waves during sleep and acquired epileptic aphasia.

Robust Nonlinear Mapping of Soil Contamination Using Support Regression

Spatial data analysis mapping and visualization is of great importance in various fields: environment, pollution, natural hazards and risks, epidemiology, spatial econometrics, etc. A basic task of spatial mapping is to make predictions based on some empirical data (measurements). A number of state-of-the-art methods can be used for the task: deterministic interpolations, methods of geostatistics: the family of kriging estimators (Deutsch and Journel, 1997), machine learning algorithms such as artificial neural networks (ANN) of different architectures, hybrid ANN-geostatistics models (Kanevski and Maignan, 2004; Kanevski et al., 1996), etc. All the methods mentioned above can be used for solving the problem of spatial data mapping. Environmental empirical data are always contaminated/corrupted by noise, and often with noise of unknown nature. That's one of the reasons why deterministic models can be inconsistent, since they treat the measurements as values of some unknown function that should be interpolated. Kriging estimators treat the measurements as the realization of some spatial randomn process. To obtain the estimation with kriging one has to model the spatial structure of the data: spatial correlation function or (semi-)variogram. This task can be complicated if there is not sufficient number of measurements and variogram is sensitive to outliers and extremes. ANN is a powerful tool, but it also suffers from the number of reasons. of a special type ? multiplayer perceptrons ? are often used as a detrending tool in hybrid (ANN+geostatistics) models (Kanevski and Maignank, 2004). Therefore, development and adaptation of the method that would be nonlinear and robust to noise in measurements, would deal with the small empirical datasets and which has solid mathematical background is of great importance. The present paper deals with such model, based on Statistical Learning Theory (SLT) - Support Vector Regression. SLT is a general mathematical framework devoted to the problem of estimation of the dependencies from empirical data (Hastie et al, 2004; Vapnik, 1998). SLT models for classification - Support Vector Machines - have shown good results on different machine learning tasks. The results of SVM classification of spatial data are also promising (Kanevski et al, 2002). The properties of SVM for regression - Support Vector Regression (SVR) are less studied. First results of the application of SVR for spatial mapping of physical quantities were obtained by the authorsin for mapping of medium porosity (Kanevski et al, 1999), and for mapping of radioactively contaminated territories (Kanevski and Canu, 2000). The present paper is devoted to further understanding of the properties of SVR model for spatial data analysis and mapping. Detailed description of the SVR theory can be found in (Cristianini and Shawe-Taylor, 2000; Smola, 1996) and basic equations for the nonlinear modeling are given in section 2. Section 3 discusses the application of SVR for spatial data mapping on the real case study - soil pollution by Cs137 radionuclide. Section 4 discusses the properties of the modelapplied to noised data or data with outliers.

Mineral zoning and geochemistry of epithermal polymetallic Zn-Pb-Ag-Cu-Bi mineralization at Cerro de Pasco, Peru

The large Cerro de Pasco Cordilleran base metal deposit in central Peru is located on the eastern margin of a middle Miocene diatreme-dome complex and comprises two mineralization stages. The first stage consists of a large pyrite-quartz body replacing Lower Mesozoic Pucara carbonate rocks and, to a lesser extent, diatreme breccia. This body is composed of pyrite with pyrrhotite inclusions, quartz, and black and red chalcedony (containing hypogene hematite). At the contact with the pyrite-quartz body, the diatreme breccia is altered to pyrite-quartz-sericite-pyrite. This body was, in part, replaced by pipelike pyrrhotite bodies zoned outward to carbonate-replacement Zn-Pb ores hearing Fe-rich sphalerite (up to 24 mol % Fes). The second mineralization stage is partly superimposed on the first and consists of zoned east-west-trending Cu-Ag-(Au-Zn-Pb) enargite-pyrite veins hosted in the diatreme breccia in the western part of the deposit and well-zoned Zn-Pb-(Bi-Ag-Cu) carbonate-replacement orebodies; in both cases, sphalerite is Fe poor and the inner parts of the orebodies show typically advanced argillic alteration assemblages, including aluminum phosphate Sulfate (APS) minerals. The zoned enargite-pyrite veins display mineral zoning, from a core of enargite-pyrite +/- alunite with traces of Au, through an intermediate zone of tennantite, chalcopyrite, and Bi minerals to a poorly developed Outer zone hearing sphalerite-galena +/- kaolinite. The carbonate-hosted replacement ores are controlled along N 35 degrees E, N 90 degrees E, N 120 degrees E, and N 170 degrees E faults. They form well-zoned upward-flaring pipelike orebodies with a core of famatinite-pyrite and alunite, an intermediate zone with tetrahedrite-pyrite, chalcopyrite, matildite, cuprobismutite, emplectite, and other Bi minerals accompanied by APS minerals, kaolinite, and dickite, and an outer zone composed of Fe-poor sphalerite (in the range of 0.05-3.5 mol % Fes) and galena. The outermost zone consists of hematite, magnetite, and Fe-Mn-Zn-Ca-Mg carbonates. Most of the second-stage carbonate-replacement orebodies plunge between 25 degrees and 60 degrees to the west, suggesting that the hydrothermal fluids ascended from deeper levels and that no lateral feeding from the veins to the carbonate-replacement orebodies took place. In the Venencocha and Santa Rosa areas, located 2.5 km northwest of the Cerro de Pasco open pit and in the southern part of the deposit, respectively, advanced argillic altered dacitic domes and oxidized veins with advanced argillic alteration halos occur. The latter veins are possibly the oxidized equivalent of the second-stage enargite-pyrite veins located in the western part of the deposit. The alteration assemblage quartz-muscovite-pyrite associated with the pyrite-quartz body suggests that the first stage precipitated at slightly, acidic fin. The sulfide mineral assemblages define an evolutionary path close to the pyrite-pyrrhotite boundary and are characteristic of low-sulfidation states; they suggest that the oxidizing slightly acidic hydrothermal fluid was buffered by phyllite, shale, and carbonate host rock. However, the presence in the pyrite-quartz body of hematite within quartz suggests that, locally, the fluids were less buffered by the host rock. The mineral assemblages of the second mineralization stage are characteristic of high- to intermediate-sulfidation states. High-sulfidation states and oxidizing conditions were achieved and maintained in the cores of the second-stage orebodies, even in those replacing carbonate rocks. The observation that, in places, second-stage mineral assemblages are found in the inner and outer zones is explained in terms of the hydrothermal fluid advancing and waning. Microthermometric data from fluid inclusions in quartz indicate that the different ores of the first mineralization stage formed at similar temperatures and moderate salinities (200 degrees-275 degrees C and 0.2-6.8 wt % NaCl equiv in the pyrite-quartz body; 192 degrees-250 degrees C and 1.1-4.3 wt % NaCl equiv in the pyrrhotite bodies; and 183 degrees-212 degrees C and 3.2-4.0 wt % NaCl equiv in the Zn-Pb ores). These values are similar to those obtained for fluid inclusions in quartz and sphalerite from the second-stage ores (187 degrees-293 degrees C and 0.2-5.2 wt % NaCl equiv in the enargite-pyrite veins: 178 degrees-265 degrees C and 0.2-7.5 wt % NaCl equiv in quartz of carbonate-replacement orebodies; 168 degrees-999 degrees C and 3-11.8 wt % NaCl equiv in sphalerite of carbonate-replacement orebodies; and 245 degrees-261 degrees C and 3.2-7.7 wt % NaCl equiv in quartz from Venencocha). Oxygen and hydrogen isotope compositions oil kaolinite from carbonate-replacement orebodies (delta(18)O = 5.3-11.5%o, delta D = -82 to -114%o) and on alunite from the Venencocha and Santa Rosa areas (delta(18)O = 1.9-6.9%o, delta D = -56 to -73%o). Oxygen isotope compositions of quartz from the first and second stages have 6180 values from 9.1 to 1.7.8 per mil. Calculated fluids in equilibrium with kaolinite have delta(18)O values of 2.0 to 8.2 and delta D values of -69 to -97 per mil; values in equilibrium with alunite are -1.4 to -6.4 and -62 to -79 per mil. Sulfur isotope compositions of sulfides from both stages have a narrow range of delta(34)S values, between -3.7 and +4.2 per mil; values for sulfates from the second stage are between 4.2 and 31.2 per mil. These results define two mixing trends for the ore-forming fluids. The first trend reflects mixing between a moderately saline (similar to 10 wt % NaCl equiv) magmatic end member that had degassed (as indicated by the low delta D values) and meteoric water. The second mixing indicates condensation of magmatic vapor with HCl and SO(2) into meteoric water, which formed alunite. The hydrothermal system at Cerro de Pasco was emplaced at a shallow depth (similar to 500 m) in the epithermal and upper part of a porphyry environment. The similar temperatures and salinities obtained for the first stage and second stages, together with the stable isotope data, indicate that both stages are linked and represent successive stages of epithermal polymetallic mineralization in the upper part of a porphyry system.

Spatial prediction of monthly wind speeds in complex terrain with adaptive general regression neural networks

This paper presents the general regression neural networks (GRNN) as a nonlinear regression method for the interpolation of monthly wind speeds in complex Alpine orography. GRNN is trained using data coming from Swiss meteorological networks to learn the statistical relationship between topographic features and wind speed. The terrain convexity, slope and exposure are considered by extracting features from the digital elevation model at different spatial scales using specialised convolution filters. A database of gridded monthly wind speeds is then constructed by applying GRNN in prediction mode during the period 1968-2008. This study demonstrates that using topographic features as inputs in GRNN significantly reduces cross-validation errors with respect to low-dimensional models integrating only geographical coordinates and terrain height for the interpolation of wind speed. The spatial predictability of wind speed is found to be lower in summer than in winter due to more complex and weaker wind-topography relationships. The relevance of these relationships is studied using an adaptive version of the GRNN algorithm which allows to select the useful terrain features by eliminating the noisy ones. This research provides a framework for extending the low-dimensional interpolation models to high-dimensional spaces by integrating additional features accounting for the topographic conditions at multiple spatial scales. Copyright (c) 2012 Royal Meteorological Society.