|
Activity Number:
|
494
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #310020 |
|
Title:
|
The Use of Multiscale Methods To Characterize Resting fMRI Data
|
|
Author(s):
|
Nicole Lazar*+
|
|
Companies:
|
University of Georgia
|
|
Address:
|
, Athens, GA, 30602,
|
|
Keywords:
|
|
|
Abstract:
|
Current approaches to the analysis of fMRI data assume simple models for the measurement error or for the structure of resting data. These include temporal independence and models for (short or long term) temporal dependence. In the absence of a long time series of data, which will permit the systematic examination of the various models, they remain assumptional in nature. However, statistically valid statements regarding performance of tests require a better characterization of the dependence structure at the voxel level. We investigate the temporal dependence in a long resting fMRI data set; such a time horizon is sufficient to provide insight into possible long-range structure, something that is not feasible with the typical length of scan. Using SiZer and wavelet spectra, we show that voxels in different parts of the brain should be modeled with different dependence structure.
|
- 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 2007 program |
