MPhil Thesis Defence


"Three dimensional vascular segmentation based on maximum intensity
projections and orientation tensors"

By

Mr. Chun Kit Wong


Abstract

Angiographic images are routinely acquired for the medical diagnosis of
arterial diseases. Digital images acquired are in the form of
three-dimensional (3D) volume, which consists of a stack of slices.
Because of lack of 3D information in those angiographic slices,
radiologists generate maximum intensity projection (MIP) images at
different view angles to facilitate the diagnosis.  This two-dimensional
(2D) representation of the arterial morphology creates ambiguities in the
assessment of position and size of arterial abnormalities.  The motivation
of this work stems from the clinical need for a 3D representation of
vascular structure obtained from those angiograms, in order to overcome
the problem.

In this work, two new segmentation techniques are presented for extracting
3D vessels from angiographic images.  The two segmentation techniques can
be used complementarily.  The first technique presented extracts 3D
vessels based on MIP images.  It is a unified platform to segment
different kinds of angiograms; the second technique presented refines an
initial 3D vascular segmentation based on orientation tensor.  Orientation
tensor is used to analyze local structure of the arterial morphology.
Initial segmentation is then refined on a probabilistic framework based on
local structure descriptions.  We take the maximum a posteriori
(MAP)-Markov random field (MRF) approach to solve this segmentation
problem.  The benefits of our segmentation technique are twofold.  It is a
unified algorithm to segment different kinds of angiograms and it uses
only intrinsic information (local structure) from arterial morphology to
extract 3D vessels.

Experimental results on time-of-flight (TOF) magnetic resonance
angiography (MRA), phase contrast (PC) MRA and three-dimensional
rotational angiography (3D-RA) show that our algorithms outperform other
methods in extracting vascular structures with low signal-to-noise ratio
(SNR), e.g., small vessels and aneurysms.


Date:			Monday, 25 August 2003

Time:			2:00p.m.-4:00p.m.

Venue:			Room 3315
			Lifts 17-18

Committee Members:		Dr. Albert Chung (Supervisor)
			Dr. Long Quan (Chairman)
			Dr. Michael Brown


**** ALL are Welcome ****