|
Activity Number:
|
27
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 8, 2004 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #302073 |
|
Title:
|
Resampling for Brain Volumetric Analysis
|
|
Author(s):
|
Wei Zhu*+ and Yeming Ma and Nora D. Volkow and Manlong Rao and Gene-Jack Wang
|
|
Companies:
|
SUNY, Stony Brook and National Institutes of Health and National Institutes of Health and SUNY, Stony Brook and Brookhaven National Laboratory
|
|
Address:
|
Dept. of Applied Math & Statistics, Stony Brook, NY, 11794-3600,
|
|
Keywords:
|
bootstrap ; jackknife ; resampling ; PET ; SPM ; imaging
|
|
Abstract:
|
Brain activation/deactivation volume is as important an indicator of brain functional activities as (de)activation intensity. The widely used Statistical Parametric Mapping software (SPM) reports the (de)activation volume from one condition to the other. However, no statistical test is available in SPM to compare the (de)activation volume between two or more stimuli. There are two approaches to accomplish the statistical comparison of (de)activation volumes. The first approach is highly intuitive. For each subject, one would claim a certain brain region or voxel being activated from baseline to stimulus if it exceeds a certain threshold, and deactivated if it is less than another threshold. The difficulty with this approach lies in the determination of the thresholds, which is usually heuristic and has no statistical basis. We propose to adopt the statistical Resampling method including both the bootstrap method and the jackknife method as a natural solution to the brain volumetric analysis. The major difficulty with resampling is the computational intensity. We have overcome this obstacle by developing the SPM Batch-Mode software in-house.
|
- 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 2004 program |
