|
Activity Number:
|
417
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #309368 |
|
Title:
|
Weighted Fourier Series Analysis of Anatomical Brain Structures
|
|
Author(s):
|
Shubing Wang*+
|
|
Companies:
|
University of Wisconsin-Madison
|
|
Address:
|
Department of Statistics, 1245 Medical Sciences Center, Madison, WI, 53705,
|
|
Keywords:
|
Weighted ; AIR ; registration
|
|
Abstract:
|
As the generalized forms of Fourier Series and Spherical Harmonics, WFS are superior since they not only effectively reduce Gibb's phenomenon, but provide a convenient way to compute smoothness when applying random field theories. An Adaptive Iterative Regression (AIR) method is proposed for WFS representations of large-sized data, which preserves the computational efficiency of previous iterative fitting methods and improve the computational efficiency of Least-Squared Estimations. Using a novel method of calculating curvatures, we propose a curvature-based global shift registration method. Decision-tree-based classification techniques and random field theories to study the link between autism and its underlying neuroanatomy.
|
- 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 |
