Silvia N. Elian

Streptococcus mutans-induced secondary caries adjacent to glass ionomer cement, composite resin and amalgam restorations in vitro

The aim of this study was to define, in vitro, the potential to inhibit secondary caries of restorative materials currently used in dental practice. Standard cavities were prepared on the buccal and lingual surfaces of fifty extracted human third molars. The teeth were randomly divided into five groups, each one restored with one of the following materials: glass ionomer cement (GIC); amalgam; light-cured composite resin; ion-releasing composite; and light-cured, fluoride-containing composite resin. The teeth were thermocycled, sterilized with gamma irradiation, exposed to a cariogenic challenge using a bacterial system using Streptococcus mutans, and then prepared for microscopic observation. The following parameters were measured in each lesion formed: extension, depth, and caries inhibition area. The outer lesions developed showed an intact surface layer and had a rectangular shape. Wall lesions were not observed inside the cavities. After Analysis of Variance and Component of Variance Models Analysis, it was observed that the GIC group had the smallest lesions and the greatest number of caries inhibition areas. The lesions developed around Amalgam and Ariston pHc restorations had an intermediate size and the largest lesions were observed around Z-100 and Heliomolar restorations. It may be concluded that the restorative materials GIC, amalgam and ion-releasing composites may reduce secondary caries formation.

The minimum sum of absolute errors regression: a robust alternative to the least squares regression

This paper concerns the minimum sum of absolute errors regression. It is a more robust alternative to the popular least squares regression whenever there are outliers in the values of the response variable, or the errors follow a long tailed distribution, or the loss function is proportional to the absolute errors rather than their squared values. We use data from a study of interstitial lung disease to illustrate the method, interpret the findings, and contrast with least squares regression. We point out some of the problems with the least squares analysis and show how to avoid these with the minimum sum of absolute errors analysis.

Influence measure for the L1 regression

Because outliers and leverage observations unduly affect the least squares regression, the identification of influential observations is considered an important and integrai part of the analysis. However, very few techniques have been developed for the residual analysis and diagnostics for the minimum sum of absolute errors, L1 regression. Although the L1 regression is more resistant to the outliers than the least squares regression, it appears that outliers (leverage) in the predictor variables may affect it. In this paper, our objective is to develop an influence measure for the L1 regression based on the likelihood displacement function. We illustrate the proposed influence measure with examples.

Simple Forms of the Best Linear Unbiased Predictor in the General Linear Regression Model

Abstract This article presents necessary and sufficient conditions to be satisfied by the best linear unbiased predictor of future observations in the general linear model in order to have a simple form. Under these conditions, the predictors have an expression similar to that in the uncorrelated case and some parameters related to the covariances between some observations need not to be known.

Coefficients of determinations for variable selection in the msae regression

Our objective is to modify a robust coefficient of determination for the minimum sum of absolute errors MSAE regression proposed by McKean and Sievers (1987) so that it satisfies all the desirable properties. We also propose an adjusted coefficient of determination that is appropriate for comparing several models with different number of variables. Further, it has the property that if it decreases with the addition of predictor variables to the model, then the contribution of these variables is statistically non-significant. We illustrate the results with an example.

Using electron microscopy and multivariate cluster analysis to determine diagnosis and prognosis in cases of neuroendocrine lung carcinoma

OBJECTIVE To establish reproducible electron microscopic criteria for identifying the four major types of neuroendocrine tumors of the lung: carcinoid; atypical carcinoid; large cell neuroendocrine carcinoma; and small cell carcinoma. METHODS Measurements were made on electron micrographs using a digital image analyzer. Sixteen morphometric variables related to tumor cell differentiation were assessed in 27 tumors. The examination under electron microscopy revealed that all of the tumors could be classified as belonging to one of the four categories listed above. Cluster analysis of the morphometry variables was used to group the tumors into three clusters, and Kaplan-Meier survival function curves were employed in order to draw correlations between each cluster and survival. RESULTS All three clusters of neuroendocrine carcinomas were found to be associated with survival curves, demonstrating the prognostic significance of electron microscopic features. The tumors fell into three well-defined clusters, which represent the spectrum of neuroendocrine differentiation: typical carcinoid (cluster 1); atypical carcinoid and large cell neuroendocrine carcinoma (cluster 2); and small cell carcinoma (cluster 3). Cluster 2 represents an intermediate step in neuroendocrine carcinogenesis, between typical carcinoid tumors and small cell carcinomas. CONCLUSIONS Our findings confirm that electron microscopy is useful in making the diagnosis and prognosis in cases of lung tumor.

Influence Measures in Quantile Regression Models

In this article, we use the asymmetric Laplace distribution to define a new method to determine the influence of a certain observation in the fit of quantile regression models. Our measure is based on the likelihood displacement function and we propose two types of measures in order to determine influential observations in a set of conditional quantiles conjointly or in each conditional quantile of interest. We verify the validity of our average measure in a simulated data set as well in an illustrative example with data about air pollution.