Title page for ETD etd-12182009-121450

Type of Document

Dissertation

Author

Rajagopalan, Vidya

URN

etd-12182009-121450

Title

Increasing DBM Reliability using Distribution Independent Tests and Information Fusion Techniques

Degree

PhD

Department

Electrical and Computer Engineering

Advisory Committee

Advisor Name	Title
Wyatt, Christopher L.	Committee Chair
Kim, Inyoung	Committee Member
Mili, Lamine M.	Committee Member
Wang, Ge	Committee Member
Wang, Yue J.	Committee Member

Keywords

information fusion
deformation based morphometry
MRI
image registration
distribution independent tests

Date of Defense

2009-10-06

Availability

unrestricted

Abstract

In deformation based morphometry (DBM) group-wise differences in brain structure are measured
using deformable registration and some form of statistical test. However, it is known that DBM
results are sensitive to both the registration method and statistical test used. Given the lack of an
objective model of group variation it has been difficult to determine the extent of the influence
of registration implementation or contraints on DBM analysis. In this thesis, we use registration
methods with varying levels of theoretic similarity to study the influence of registration mechanics
on DBM results. We show that because of the extent of the influence of registration mechanics
on DBM results, analysis of changes should always be made with a thorough understanding
of the registration method used. We also show that minor variations in registration methods can
lead to large changes in DBM results. When using DBM, it would be imprudent to use only one
registration method to draw any conclusions about the variations being studied. In order to provide
a more complete representation of inter-group changes, we propose a method for combining
multiple registration methods using Dempster-Shafer evidence theory to produce belief maps of
categorical changes between groups. We show that the Dempster-Shafer combination produces a
unique and easy to interpret belief map of regional changes between and within groups without the
complications associated with hypothesis testing.
Another, often confounding, element of DBM is the parametric hypothesis test used to specify
voxels undergoing significant change between the two groups. The accuracy and reliability of
these tests are contingent on a number of fundamental assumptions made about the distribution of
the data used in the tests. Many DBM studies often overlook these assumptions and fail to verify
their validity for the data being tested. This raises many doubts about the credibility of the results
from such tests. In this thesis, we propose to perform statistical analysis on DBM data using nonparametric,
distribution independent hypothesis tests. With no data distributional assumptions,
these tests provide both increased flexibility and reliability of DBM statistical analysis

Files

Filename		Size	Approximate Download Time (Hours:Minutes:Seconds)
Filename		Size	28.8 Modem	56K Modem	ISDN (64 Kb)	ISDN (128 Kb)	Higher-speed Access
	Rajagopalan_PhdDissertation_Oct2009.pdf	26.86 Mb	02:04:20	01:03:57	00:55:57	00:27:58	00:02:23