Statistics Seminar: STA 290
Thursday, November 15th, 2012 at 4.10pm, MSB 1147 (Colloquium Room)
Refreshments 3:30pm, prior to seminar in MSB 4110 (Statistics Lounge)
Speaker: Abhyuday Mandal, University of Georgia
Title: Multi-objective optimal experimental designs for event-related fMRIstudies
Abstract: Functional magnetic resonance imaging (fMRI) is an advanced technology for studying brain functions. Due to the complexity and high cost of fMRI experiments, high quality multi-objective fMRI designs are in great demand. We propose an efficient approach to find optimal experimental designs for event-related functional magnetic resonance imaging (ER-fMRI).
We consider multiple objectives, including estimating the hemodynamic response function (HRF), detecting activation, circumventing psychological confounds and fulfilling customized requirements. Taking into account these goals, we formulate a family of multi-objective design criteria and develop a genetic-algorithm-based technique to search for optimal designs. Our proposed technique incorporates existing knowledge about the performance of fMRI designs, and its usefulness is shown through simulations. We also consider a nonlinear model to accommodate a wide spectrum of feasible HRF shapes, and propose an approach for obtaining maximin efficient designs.
Our approach involves a reduction in the parameter space and an efficient search algorithm. The designs that we obtain are much more robust against misspecified HRF shapes than designs widely used by researchers. This research is joint with Ming-Hung (Jason) Kao, Dibyen Majumdar and John Stufken.