Geodesic Image Normalization in the Space of Diffeomorphisms

Speaker:	Dr. James C. Gee
		Radiologic Science and Computer and Information Science
		University of Pennsylvania

Title:		"Geodesic Image Normalization in the Space of
		 Diffeomorphisms"

Date:		Monday, 21 August 2006

Time:		10:00am -11:00am

Venue:		Room 3598 (via lift nos. 27/28)
		HKUST

Abstract.

Medical image analysis based on diffeomorphisms (differentiable one to one
and onto maps with differentiable inverse) has placed computational
analysis of anatomy and physiology on firm theoretical ground.We detail
our approach to diffeomorphic computational anatomy while highlighting
both theoretical and practical benefits. We first introduce the metric
used to locate geodesics in the diffeomorphic space. Second, we give a
variational energy that parameterizes the image normalization problem in
terms of a geodesic diffeomorphism, enabling a fundamentally symmetric
solution. This approach to normalization is extended for optimal template
population studies using general imaging data. Finally, we show how the
temporal parameterization and large deformation capabilities of
diffeomorphisms make them appropriate for longitudinal analysis,
particularly of neurodegenerative data.



*************************
Biography:

James C. Gee, Ph.D., Associate Professor of Radiologic Science and
Computer and Information Science, is Director of the Penn Image Computing
and Science Laboratory (PICSL) and Program Co-Director of the Howard
Hughes Medical Institute Integrated Graduate Training Program in Clinical
Imaging and Informational Sciences.  Internationally recognized for
seminal contributions to computational anatomy, Dr. Gee's current work
spans numerous collaborations across a variety of disciplines and includes
applications of image analysis to study the biomechanics of moving organs;
the normal development and pathological correlates of brain structure; and
the correlation between brain structural changes and cognitive deficits in
central nervous system disorders.

PICSL is a part of the graduate groups of the Departments of Computer and
Information Science, and Bioengineering.  It is affiliated with the
Centers for Functional Neuroimaging, for Bioinformatics and for Cognitive
Neuroscience, the General Robotics, Automation, Sensing and Perception
Laboratory, the Working Group on Applied Mathematics and Computational
Science, and the Leonard Davis Institute of Health Economics, and is a
founding member of the Center for Health Informatics at Penn and the
National Library of Medicine Insight Consortium.