March 1, 2017
Researchers from UAB’s Nutrition Obesity Research Center (NORC) and Office of Energetics distinguished themselves in Session 5 at the recent 14th Annual Postdoctoral Research Day competition. Presentations were scored on clarity, the significance of the research, and the scientific approach to the problem.
Marissa Gowey, PhD (right), postdoctoral scholar in the NORC, was awarded second place for “Executive Function in Individuals with Intentional Clinically Significant Weight Loss,” which primarily aims to characterize executive function (EF) in individuals who lost a clinically significant amount of weight through behavioral intervention. Study results indicate that the majority of participants, who had previously lost at least 5% of their body weight in a behavioral intervention, performed in the average to above average range across six domains of validated EF tests, while only a small portion of these participants performed in the below average range on tests of EF. Future research is recommended to administer pre- and post-treatment assessments in determining whether individual variability in EF can predict behavioral treatment success. UAB co-investigators are Samantha Henry, MA; William Neumeier, PhD; Janice Phillips; and mentor Gareth R. Dutton, PhD.
Anarina Murillo, PhD (left), postdoctoral scholar in the NORC and Office of Energetics, was awarded third place for “Estimating Physical Activity from 2D Body Composition Data Using Regression Calibration Methods For Measurement Error Correction,” which seeks to apply regression calibration (RC) methods to predict physical activity levels using model parameter estimates of body fat percentage that have been corrected for measurement error. While dual-energy X-ray absorptiometry (DXA) scans are often treated as the “gold standard” for measuring body fat percentage, they are costly and time-consuming. In this work, physical activity levels were estimated from 2D photographic data, a new promising method for measuring body fat for which model coefficients were corrected for possible inherent error using DXA data. This work illustrates the utility of RC methods to adjust for measurement error and, therefore, have the potential to lead to less biased estimates of model parameters in statistical analyses. UAB co-investigators are Olivia Affuso, PhD; Courtney M. Peterson, PhD; and David B. Allison, PhD.
Open the original version of this page.