RANDOM EFFECTS MODELS IN A META-ANALYSIS OF THE ACCURACY OF DIAGNOSTIC TESTS WITHIN A GOLD STANDARD IN THE PRESENCE OF MISSING DATA

In evaluating the accuracy of diagnosis tests, it is common to apply two imperfect tests jointly or sequentially to a study population. In a recent meta-analysis of the accuracy of microsatellite instability testing (MSI) and traditional mutation analysis (MUT) in predicting germline mutations of the mismatch repair (MMR) genes, a Bayesian approach (Chen, Watson, and Parmigiani 2005) was proposed to handle missing data resulting from partial testing and the lack of a gold standard. In this paper, we demonstrate an improved estimation of the sensitivities and specificities of MSI and MUT by using a nonlinear mixed model and a Bayesian hierarchical model, both of which account for the heterogeneity across studies through study-specific random effects. The methods can be used to estimate the accuracy of two imperfect diagnostic tests in other meta-analyses when the prevalence of disease, the sensitivities and/or the specificities of diagnostic tests are heterogeneous among studies. Furthermore, simulation studies have demonstrated the importance of carefully selecting appropriate random effects on the estimation of diagnostic accuracy measurements in this scenario.

Pathomorphologic alterations predict presence or absence of hip osteoarthrosis

Abnormal morphology of the hip has been associated with primary osteoarthrosis. We evaluated the morphology of 464 consecutive hips contralateral to hips treated by THA. We excluded all hips with known diagnoses leading to secondary osteoarthritis and all hips with advanced arthrosis to eliminate the effect of arthritic remodeling on the morphologic measurements. Of the remaining 119 hips, 25 were in patients aged 60 years or older who had no or mild arthrosis (Tönnis Grade 0 or 1) and 94 hips had Tönnis Grade 2 osteoarthrosis. We quantified morphologic parameters on plain radiographs and CT images and simulated range of motion using virtual bone models from the CT data. The nonarthritic hips had fewer pathomorphologic findings. High alpha angles and high lateral center edge angles were strongly associated with the presence of arthritis; decreased internal and external rotation in 90 degrees flexion showed lesser correlation. The data confirm previous observations that abnormal hip morphology predates arthrosis and is not secondary to the osteoarthritic process. Hips at risk for developing arthrosis resulting from pathomorphologic changes may potentially be identified at the cessation of growth, long before the development of osteoarthrosis.

Expansion of mortar bars in the presence of potassium acetate deicer

Carboxylate-based deicing and anti-icing chemicals became widely used in the mid 1990s, replacing more environmentally burdensome chemicals. Within a few years of their adoption, distress of portland cement concrete runways was reported by a few airports using the new chemicals. Distress manifested characteristics identical to that of alkali silica reactivity (ASR), but onset occurred early in the pavement’s operating life and with pavements thought to contain innocuous aggregate. The carboxylate-based deicing chemicals were suspected of exacerbating ASR-like expansion. Innocuous, moderately, and highly reactive aggregates were tested using modified ASTM C1260 and ASTM C1567 procedures with soak solutions containing deicer solutions and sodium hydroxide or potassium hydroxide. ASR-like expansion is exacerbated in the presence of potassium acetate. The expansion rate produced by a given aggregate is also a function of the alkali hydroxide used. Petrographic analyses were performed on thin sections prepared from mortar bars used in the experiments. Expansion occurred via two mechanisms; rupture of aggregate grains and expansion of paste.

DESIGN AND FABRICATION OF RESONANT GAS SENSOR FOR HIGH SENSITIVITY IN THE PRESENCE OF AIR DAMPING

Gas sensors have been used widely in different important area including industrial control, environmental monitoring, counter-terrorism and chemical production. Micro-fabrication offers a promising way to achieve sensitive and inexpensive gas sensors. Over the years, various MEMS gas sensors have been investigated and fabricated. One significant type of MEMS gas sensors is based on mass change detection and the integration with specific polymer. This dissertation aims to make contributions to the design and fabrication of MEMS resonant mass sensors with capacitance actuation and sensing that lead to improved sensitivity. To accomplish this goal, the research has several objectives: (1) Define an effective measure for evaluating the sensitivity of resonant mass devices; (2) Model the effects of air damping on microcantilevers and validate models using laser measurement system (3) Develop design guidelines for improving sensitivity in the presence of air damping; (4) Characterize the degree of uncertainty in performance arising from fabrication variation for one or more process sequences, and establish design guidelines for improved robustness. Work has been completed toward these objectives. An evaluation measure has been developed and compared to an RMS based measure. Analytic models of air damping for parallel plate that include holes are compared with a COMSOL model. The models have been used to identify cantilever design parameters that maximize sensitivity. Additional designs have been modeled with COMSOL and the development of an analytical model for Fixed-free cantilever geometries with holes has been developed. Two process flows have been implemented and compared. A number of cantilever designs have been fabricated and the uncertainty in process has been investigated. Variability from processing have been evaluated and characterized.