PhD Thesis Proposal Defence


"Robust Multimodal Medical Image Registration and Statistical
Cerebrovascular Segmentation"

by

Miss Rui Gan


Abstract:

In this proposal, we focus on two important medical image processing
and analysis problems: multimodal image registration and
cerebrovascular segmentation.

First, to increase image registration robustness, we design a new
spatial feature, namely maximum distance-gradient-magnitude (MDGM), to
encode spatial information at a global level, including both edges and
distances. By associating MDGM with intensity into a two-element
attribute vector, we adopt multi-dimensional mutual information as
similarity measure on the vector space. Alternatively, if statistical
joint intensity mappings obtained from the pre-aligned training image
pairs are available, this a priori knowledge can be used as reference
to guide the registration by minimizing the Kullback-Leibler distance
(KLD) between the observed and reference joint intensity
distributions. Second, this proposal presents a statistical
segmentation technique for extraction of vasculature in 3D rotational
angiography (3D-RA). This method uses maximum intensity projections
(MIP) to improve the accuracy and robustness of threshold estimation
using the expectation maximization (EM) algorithm, and it is fully
automatic and computationally efficient. Finally, encouraging
preliminary results of the proposed approaches are presented in the
proposal.


Date:     		Thursday, 4 May 2006

Time:                   11:30a.m.-1:30p.m.

Venue:                  Room 3464
			lifts 25-26

Committee Members:      Dr. Albert Chung (Supervisor)
                        Dr. Chiew-Lan Tai (Chairperson)
			Dr. Huamin Qu
			Dr. Chi-Keung Tang


**** ALL are Welcome ****