Residuals for log-Burr XII regression models in survival analysis

In this paper, we compare three residuals to assess departures from the error assumptions as well as to detect outlying observations in log-Burr XII regression models with censored observations. These residuals can also be used for the log-logistic regression model, which is a special case of the log-Burr XII regression model. For different parameter settings, sample sizes and censoring percentages, various simulation studies are performed and the empirical distribution of each residual is displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extended to the modified martingale-type residual in log-Burr XII regression models with censored data.

Log-modified Weibull regression models with censored data: Sensitivity and residual analysis

This paper proposes a regression model considering the modified Weibull distribution. This distribution can be used to model bathtub-shaped failure rate functions. Assuming censored data, we consider maximum likelihood and Jackknife estimators for the parameters of the model. We derive the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and we also present some ways to perform global influence. Besides, for different parameter settings, sample sizes and censoring percentages, various simulations are performed and the empirical distribution of the modified deviance residual is displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extended for a martingale-type residual in log-modified Weibull regression models with censored data. Finally, we analyze a real data set under log-modified Weibull regression models. A diagnostic analysis and a model checking based on the modified deviance residual are performed to select appropriate models. (c) 2008 Elsevier B.V. All rights reserved.

A Log-Linear Regression Model for the Beta-Weibull Distribution

We introduce the log-beta Weibull regression model based on the beta Weibull distribution (Famoye et al., 2005; Lee et al., 2007). We derive expansions for the moment generating function which do not depend on complicated functions. The new regression model represents a parametric family of models that includes as sub-models several widely known regression models that can be applied to censored survival data. We employ a frequentist analysis, a jackknife estimator, and a parametric bootstrap for the parameters of the proposed model. We derive the appropriate matrices for assessing local influences on the parameter estimates under different perturbation schemes and present some ways to assess global influences. Further, for different parameter settings, sample sizes, and censoring percentages, several simulations are performed. In addition, the empirical distribution of some modified residuals are displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be extended to a modified deviance residual in the proposed regression model applied to censored data. We define martingale and deviance residuals to evaluate the model assumptions. The extended regression model is very useful for the analysis of real data and could give more realistic fits than other special regression models.

Sudden onset of log-periodicity and superdiffusion in non-Markovian random walks with amnestically induced persistence: exact results

Random walks can undergo transitions from normal diffusion to anomalous diffusion as some relevant parameter varies, for instance the L,vy index in L,vy flights. Here we derive the Fokker-Planck equation for a two-parameter family of non-Markovian random walks with amnestically induced persistence. We investigate two distinct transitions: one order parameter quantifies log-periodicity and discrete scale invariance in the first moment of the propagator, whereas the second order parameter, known as the Hurst exponent, describes the growth of the second moment. We report numerical and analytical results for six critical exponents, which together completely characterize the properties of the transitions. We find that the critical exponents related to the diffusion-superdiffusion transition are identical in the positive feedback and negative feedback branches of the critical line, even though the former leads to classical superdiffusion whereas the latter gives rise to log-periodic superdiffusion.

Spectral peak resolution and speech recognition in quiet: Normal hearing, hearing impaired and cochlear implant listeners

Spectral peak resolution was investigated in normal hearing (NH), hearing impaired (HI), and cochlear implant (CI) listeners. The task involved discriminating between two rippled noise stimuli in which the frequency positions of the log-spaced peaks and valleys were interchanged. The ripple spacing was varied adaptively from 0.13 to 11.31 ripples/octave, and the minimum ripple spacing at which a reversal in peak and trough positions could be detected was determined as the spectral peak resolution threshold for each listener. Spectral peak resolution was best, on average, in NH listeners, poorest in CI listeners, and intermediate for HI listeners. There was a significant relationship between spectral peak resolution and both vowel and consonant recognition in quiet across the three listener groups. The results indicate that the degree of spectral peak resolution required for accurate vowel and consonant recognition in quiet backgrounds is around 4 ripples/octave, and that spectral peak resolution poorer than around 1–2 ripples/octave may result in highly degraded speech recognition. These results suggest that efforts to improve spectral peak resolution for HI and CI users may lead to improved speech recognition

Expression of a truncated Brca1 protein delays lactational mammary development in transgenic mice

To address the hypothesis that certain disease-associated mutants of the breast-ovarian cancer susceptibility gene BRCA1 have biological activity in vivo, we have expressed a truncated Brca1 protein (trBrca1) in cell-lines and in the mammary gland of transgenic mice. Immunofluorescent analysis of transfected cell-lines indicates that trBRCA1 is a stable protein and that it is localized in the cell cytoplasm. Functional analysis of these cell-lines indicates that expression of trBRCA1 confers an increased radiosensitivity phenotype on mammary epithelial cells, consistent with abrogation of the BRCA1 pathway. MMTV-trBrca1 transgenic mice from two independent lines displayed a delay in lactational mammary gland development, as demonstrated by altered histological profiles of lobuloalveolar structures. Cellular and molecular analyses indicate that this phenotype results from a defect in differentiation, rather than altered rates of proliferation or apoptosis. The results presented in this paper are consistent with trBrca1 possessing dominant-negative activity and playing an important role in regulating normal mammary development. They may also have implications for germline carriers of BRCA1 mutations.

Alterações contráteis segmentares e sua associação com arritmias ventriculares complexas, em pacientes chagásicos com eletrocardiograma normal ou borderline

INTRODUÇÃO: Morte súbita representa a principal causa de óbito em chagásicos. Eventos fatais em pacientes saudáveis, com anormalidades contráteis, foram documentados. O objetivo deste trabalho é determinar a associação entre alteração contrátil e arritmias ventriculares complexas em pacientes chagásicos, classe funcional I, eletrocardiograma normal ou borderline e função ventricular preservada. MÉTODOS: Quarenta e nove pacientes com doença de Chagas e eletrocardiograma normal ou borderline realizaram ecocardiograma, teste ergométrico e Holter. Avaliou-se a contratilidade global e segmentar dos ventrículos e a presença de arritmias ventriculares complexas induzidas no esforço e espontâneas, respectivamente. A análise estatística foi feita pelo modelo Log-Linear geral. RESULTADOS: Idade média de 56 anos; 55% mulheres. Alterações contráteis segmentares em 24,5% dos pacientes; 12% dos Holter e 18% dos testes ergométricos positivos. Houve associação entre arritmia e alteração segmentar condicionada à presença da disfunção sistólica leve do ventrículo esquerdo. CONCLUSÕES: Alterações contráteis, na presença de disfunção global leve, indicam pacientes sob maior risco de arritmias complexas.

Detection of incipient left ventricular hypertrophy in mild to moderate arterial hypertension with normal electrocardiogram and echocardiogram: a new use for signal-averaged electrocardiography

OBJECTIVE: To assess signal-averaged electrocardiogram (SAECG) for diagnosing incipient left ventricular hypertrophy (LVH). METHODS: A study with 115 individuals was carried out. The individuals were divided as follows: GI - 38 healthy individuals; GII - 47 individuals with mild to moderate hypertension and normal findings on echocardiogram and ECG; and GIII - 30 individuals with hypertension and documented LVH. The magnitude vector of the SAECG was analyzed with the high-pass cutoff frequency of 40 Hz through the bidirectional four-pole Butterworth high-pass digital filter. The mean quadratic root of the total QRS voltage (RMST) and the two-dimensional integral of the QRS area of the spectro-temporal map were analyzed between 0 and 30 Hz for the frequency domain (Int FD), and between 40 and 250 Hz for the time domain (Int TD). The electrocardiographic criterion for LVH was based on the Cornell Product. Left ventricular mass was calculated with the Devereux formula. RESULTS: All parameters analyzed increased from GI to GIII, except for Int FD (GII vs GIII) and RMST log (GII vs GIII). Int TD showed greater accuracy for detecting LVH with an appropriate cutoff > 8 (sensitivity of 55%, specificity of 81%). Positive values (> 8) were found in 56.5% of the G II patients and in 18.4% of the GI patients (p< 0.0005). CONCLUSION: SAECG can be used in the early diagnosis of LVH in hypertensive patients with normal ECG and echocardiogram.

Long-term fluctuation of relative afferent pupillary defect in subjects with normal visual function.

PURPOSE: To determine whether the relative afferent pupillary defect (RAPD) remains constant over time in normal subjects. METHODS: Seventeen normal subjects were tested with infrared pupillography and automated perimetry in four sessions over 3 years. The changes in RAPD and visual field asymmetry between testing sessions were compared. RESULTS: The range of RAPD was 0.0 to 0.3 log unit, and the difference in the mean deviation between the eyes on automated static perimetry was 0 to 3 dB. Eight subjects repeatedly had an RAPD in the same eye. There was no correlation between the RAPD and the visual field asymmetry at the same visit. Changes in the magnitude of the RAPD between any two sessions were typically small (median, 0.08 log unit; 25th percentile, 0.04 log unit; 75th percentile, 0.15 log unit). CONCLUSIONS: Some normal subjects may show a persistent but small RAPD in the absence of detectable pathologic disease. Therefore, an isolated RAPD in the range of 0.3 log unit that is not associated with any other significant historical or clinical finding should probably be considered benign.

Oxidative and nonoxidative glucose metabolism following graded doses of oral glucose in man.

The oxidative and nonoxidative glucose metabolism represent the two major mechanisms of the utilization of a glucose load. Eight normal subjects were administered oral loads of 50, 100 and 150 g glucose and gas exchange measurements were performed for eight hours by means of computerized continuous indirect calorimetry. The glycemic peaks were almost identical with all three doses with a rise to between 141 and 147 mg/dl at 60 min. The fall back to basal level was reached later with the high than with the low glucose doses. The glucose oxidation rate rose to values between 223 and 253 mg/min after the three glucose doses, but while falling immediately after the peak at 120 min following the 50 g load, the glucose oxidation rate remained at its maximum rate until 210 min for the 100 g glucose load and plateaued up to 270 min for the 150 g glucose dose. The oxidation rates then fell gradually to reach basal levels at 270, 330 and 420 min according to the increasing size of the load. Altogether 55 +/- 3 g glucose were oxidized during the 8 hours following the 50 g glucose load, 75 +/- 3 g after the 100 g load and 80 +/- 5 g after the 150 g load. The nonoxidative glucose disposal, which corresponds essentially to glucose storage, varied according to the size of the glucose load, with uptakes of 20 +/- 1, 60 +/- 1 and 110 +/- 1 g glucose 180 min after the 50, 100 and 150 g glucose loads respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Hemodynamic and biochemical consequences of renin inhibition by infusion of CGP 38560A in normal volunteers.

Hemodynamic and biochemical effects of the new renin inhibitor CGP 38560A (molecular weight 826) were tested in 15 healthy volunteers after a single-blind, randomized, placebo-controlled protocol. At a 2-week interval, groups of five subjects received a 30-minute infusion of either 5% dextrose or CGP 38560A 50, 125, or 250 micrograms/kg. Blood pressure, heart rate, plasma renin activity, active and total renin, angiotensin-(1-8)octapeptide (angiotensin II), and aldosterone were sequentially measured up to 3 hours from the onset of the infusion. There was no consistent change in blood pressure or heart rate. Plasma renin activity and angiotensin II decreased dose dependently, and peak suppression was observed at the end of the infusion of CGP 38560A and after the 250-micrograms/kg dose. Plasma renin activity fell from 1.0 +/- 0.19 (mean +/- SEM) to less than 0.05 ng/ml/hr in all five subjects (p less than 0.001), and angiotensin II fell from 7.7 +/- 1.2 to 2.6 +/- 0.9 femtomole/ml (p less than 0.01). Active renin rose fourfold from 24 +/- 1.9 to 98 +/- 14 pg/ml (p less than 0.001) at the end of the infusion of the high dose. Plasma angiotensin II returned toward its initial values much faster than plasma renin activity and active renin. In conclusion, CGP 38560A was well tolerated. It induced a dose-dependent decrease in angiotensin II and plasma renin activity and a long-lasting and dose-dependent rise in active renin. The doses used did not reduce plasma angiotensin II maximally despite reduction of plasma renin activity to unmeasurable levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Glucagon-like peptide-I and the control of insulin secretion in the normal state and in NIDDM.

Potentiation of glucose-induced insulin secretion by intestinal factors has been described for many years. Today, two major peptides with potent insulinotropic action have been recognized: gastric inhibitory peptide and truncated forms of glucagon-like peptide I, GLP-I(7-37) or the related GLP-I(7-36)amide. These hormones have specific beta-cell receptors that are coupled to production of cAMP and activation of cAMP-dependent protein kinase. Elevation in intracellular cAMP levels is required to mediate the glucoincretin effect of these hormones: the potentiation of insulin secretion in the presence of stimulatory concentrations of glucose. In addition, circulating glucoincretins maintain basal levels of cAMP, which are necessary to keep beta-cells in a glucose-competent state. Interactions between glucoincretin signaling and glucose-induced insulin secretion may result from the phosphorylation of key elements of the glucose signaling pathway by cAMP-dependent protein kinase. These include the ATP-dependent K+ channel, the Ca++ channel, or elements of the secretory machinery itself. In NIDDM, the glucoincretin effect is reduced. However, basal or stimulated gastric inhibitory peptide and glucagon-like peptide I levels are normal or even elevated, suggesting that signals induced by these hormones on the beta-cells are probably altered. At pharmacological doses, infusion of glucagon-like peptide I but not gastric inhibitory peptide, can ameliorate postprandial insulin secretory response in NIDDM patients. Agonists of the glucagon-like peptide I receptor have been proposed as new therapeutic agents in NIDDM.

Comparison of thermogenic effect of fructose and glucose in normal humans.

After nutrient ingestion there is an increase in energy expenditure that has been referred to as dietary-induced thermogenesis. In the present study we have employed indirect calorimetry to compare the increment in energy expenditure after the ingestion of 75 g of glucose or fructose in 17 healthy volunteers. During the 4 h after glucose ingestion the plasma insulin concentration increased by 33 +/- 4 microU/ml and this was associated with a significant increase in carbohydrate oxidation and decrement in lipid oxidation. Energy expenditure increased by 0.08 +/- 0.01 kcal/min. When fructose was ingested, the plasma insulin concentration increased by only 8 +/- 2 microU/ml vs. glucose. Nonetheless, the increments in carbohydrate oxidation and decrement in lipid oxidation were significantly greater than with glucose. The increment in energy expenditure was also greater with fructose. When the mean increment in plasma insulin concentration after fructose was reproduced using the insulin clamp technique, the increase in carbohydrate oxidation and decrement in lipid oxidation were markedly reduced compared with the fructose-ingestion study; energy expenditure failed to increase above basal levels. To examine the role of the adrenergic nervous system in fructose-induced thermogenesis, fructose ingestion was also performed during beta-adrenergic blockade with propranolol. The increase in energy expenditure during fructose plus propranolol was lower than with fructose ingestion alone. These results indicate that the stimulation of thermogenesis after carbohydrate ingestion is related to an augmentation of cellular metabolism and is not dependent on an increase in the plasma insulin concentration per se.(ABSTRACT TRUNCATED AT 250 WORDS)

Robust Estimators of the Generalized Log-Gamma Distribution

We propose robust estimators of the generalized log-gamma distribution and, more generally, of location-shape-scale families of distributions. A (weighted) Q tau estimator minimizes a tau scale of the differences between empirical and theoretical quantiles. It is n(1/2) consistent; unfortunately, it is not asymptotically normal and, therefore, inconvenient for inference. However, it is a convenient starting point for a one-step weighted likelihood estimator, where the weights are based on a disparity measure between the model density and a kernel density estimate. The one-step weighted likelihood estimator is asymptotically normal and fully efficient under the model. It is also highly robust under outlier contamination. Supplementary materials are available online.

Cell lineage relationship in the stomach of normal and genetically manipulated mice

The oxyntic mucosa of the mouse stomach is lined with a heterogeneous population of cells that form numerous short pits continuous with long tubular glands. Tritiated thymidine radioautography has made it possible to pinpoint the origin of all cell types and to follow the differentiation/migration of different cell lineages along the pit-gland unit. The proliferating multipotent stem cells functionally anchored in the upper glandular region, the isthmus, give rise to three main lineage precursors: 1) pre-pit cells, which migrate upward to the pit while differentiating into mucus-producing pit cells; 2) pre-neck cells, which migrate downward to the glandular neck while differentiating into mucus-producing neck cells that, by approaching the glandular base, gradually change their phenotype into pepsinogen- and intrinsic factor-producing zymogenic cells; 3) pre-parietal cells, which differentiate into acid-producing parietal cells in the isthmus and then undergo bipolar migration towards the pit and the glandular base. Thus, parietal cells are the only cells that complete their differentiation in the isthmus and then migrate to be scattered throughout the pit-gland unit. To determine whether parietal cells play a role in controlling decisions about cell fate within the pit-gland unit, the gastric epithelium has been examined in transgenic mice expressing the H,K-ATPase ß-subunit-1035 to +24/simian virus 40 large T antigen fusion gene. The blockade in parietal cell differentiation in these mice produces an amplification of lineage precursors, a marked depletion of zymogenic cells and an increase in pit cell census. Ablation of parietal cells in another transgenic mouse model expressing the H,K-ATPase ß-subunit-1035 to +24/diphtheria toxin fragment A fusion gene also produces amplification of lineage precursors, and similar effects on zymogenic and pit cell census. These findings strongly suggest that parietal cells produce regulatory signals that control the cellular differentiation program of both pit and zymogenic cell lineages, and would hopefully improve our ability to identify the cellular pathways leading to malignant transformation