Abstract #301966

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #301966
Activity Number: 106
Type: Invited
Date/Time: Monday, August 4, 2003 : 10:30 AM to 12:20 PM
Sponsor: IMS
Abstract - #301966
Title: Optimization of fMRI Experimental Designs with the Genetic Algorithm
Author(s): Thomas E. Nichols*+
Companies: University of Michigan
Address: 1420 Washington Height SPH II, Ann Arbor, MI, 48109,
Keywords: experimental design ; neuroscience ; functional MRI ; genetic algorithm
Abstract:

The statistical modeling of fMRI is challenging due to complicated nature of the signal and noise. The signal is temporally delayed and blurred relative to the experimental stimuli, and the noise is autocorrelated, only approximately modeled by ARMA(1,1), and is heterogeneous across the brain. Experimental designs in fMRI are also nontrivial, consisting of two or more types of conditions, presented rapidly as one per second. Additionally, the order in which conditions are presented may need to counterbalanced, or arranged so that the subject cannot predict what the next condition will be. We present a method for optimizing fMRI experiments to account for both statistical and psychological properties of the design. Using a genetic algorithm, we optimize a fitness measure that is a linear combination of estimation efficiency, a measure of counterbalancing, and a measure of how close the condition frequencies are to the desired proportions. We will show how the genetic algorithm successfully optimizes this arbitrary objective function, finding compromises between statistical efficiency psychological properties of the design.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003