Erning Li | PubMed Health

PubMed2004

Conditional estimation for generalized linear models when covariates are subject-specific parameters in a mixed model for longitudinal measurements.

The relationship between a primary endpoint and features of longitudinal profiles of a continuous response is often of interest, and a relevant framework is that of a generalized linear model with covariates that are subject-specific random effects in a linear mixed model for the longitudinal measurements. Naive implementation by imputing subject-specific effects from individual regression fits yields biased inference, and several methods for reducing this bias have been proposed. These require a parametric (normality) assumption on the random effects, which may be unrealistic. Adapting a strategy of Stefanski and Carroll (1987, Biometrika74, 703-716), we propose estimators for the generalized linear model parameters that require no assumptions on the random effects and yield consistent inference regardless of the true distribution. The methods are illustrated via simulation and by application to a study of bone mineral density in women transitioning to menopause.

Biometry↗

PubMedDate not supplied

In vitro prediction of canine urolith mineral composition using computed tomographic mean beam attenuation measurements.

Determination of urolith mineral composition is critical for management of urolithiasis in dogs and cats. Using computed tomography, urolith physical density, and hence chemical composition, can be quantified using mean beam attenuation measurements (Hounsfield units; HU). This study was designed to establish in vitro reference ranges for three types of compositionally pure uroliths retrieved from dogs. Sixty-six canine uroliths (22 uric acid, 21 calcium oxalate, 14 struvite, nine mixed or compound) were placed in a phantom array. Uroliths were scanned at 120 kVp, 200 mA, and 80 kVp, 200 mA. The region of interest (ROI) for mean HU calculation was determined using two techniques, and reference ranges were calculated for each kVp using either ROI technique. HU for urolith types of pure composition were statistically different (Wilcoxon's two-sample test, P < 0.0083 [Bonferonni correction with six comparisons for total P < 0.05]) using both ROI techniques at either kVp. Struvite uroliths were not statistically different from mixed or compound uroliths. The accuracy for determination of composition of pure uroliths ranged from 86% to 93%; the prediction accuracy for each urolith mineral type and for all uroliths in general was highest when the ROI was hand-drawn just within the visible urolith border at 80 kVp. Technique of ROI determination and kVp that yielded the highest sensitivity, specificity, and positive and negative predictive values varied for each urolith type. Therefore, in this study, HU could be used to differentiate three types of uroliths of pure mineral composition in vitro. Further studies are needed to determine the predictive value of HU in vivo.

Animals↗