The introduction of Diffusion MRI, which measures the diffusion of
water molecules in live tissue,
has lead to exciting results in medical imaging. In the presence of
diffusion tends to be restricted in the direction perpendicular to
bundles of axons.
As a consequence, the direction of maximal local diffusion typically
coincides with that of fiber tracts.
This allows for image processing techniques to recover putative
connections between distinct anatomical
regions. In this talk I will review our work in Diffusion MRI
analysis, focusing on the role of both
geometric models, and stochastic processes to interpret such data. A
will be the need for the pattern analysis of curve like structures
using both deterministic
and probabilistic completion fields.
Kaleem Siddiqi received the BS degree from Lafayette College in 1988
and the MS and PhD
degrees from Brown University in 1990 and 1995, all in the field of
electrical engineering. He
is currently a Full Professor and William Dawson Scholar in the
School of Computer Science at
McGill University. He is also a member of McGill’s Center for
Intelligent Machines. Before
moving to McGill in 1998, he was a postdoctoral associate in the
Department of Computer
Science at Yale University (96-98) and held a visiting position in
the Department of Electrical
Engineering at McGill University (95-96). His research interests are
in the areas of computer vision,
image analysis and medical imaging. He is a member of Phi Beta Kappa,
Tau Beta Pi, and Eta Kappa Nu.
He is a senior member of IEEE.
Consider the problem of visually finding an object in a mostly unknown
space with a mobile robot. It is clear that all possible views and images cannot be
examined in a practical system. Visual attention is a complex phenomenon; we view
it as a mechanism that optimizes the search processes inherent in vision.
Here, we describe a particular example of a practical robotic vision system that employs some
of these attentive processes. We cast this as an optimization problem,
i.e., optimizing the probability of finding the target given a fixed cost limit in terms
of total number of robotic actions required to find the visual target. Due to the
inherent intractability of this problem, we present an approximate solution and
investigate its performance and properties. We conclude that our approach, named the SYT
algorithm, is sufficient to solve this problem and has additional desirable empirical
John K. Tsotsos (Ph.D. Computer Science, University of Toronto, 1980).
Currently, I am Distinguished Research Professor of Vision Science at
York University and also hold the Canada Research Chair in Computational
Vision. Previously, I was on faculty at the University of Toronto from
1980 -99. I moved to York University in January 2000 to become Director
of the Centre for Vision Research, a position I held until late 2006.
Editorial board memberships include Image & Vision Computing, Computer
Vision and Image Understanding, Computational Intelligence and
Artificial Intelligence and Medicine. Current research foci are human
and machine vision, attention, autonomous wheelchairs, and visual
monitoring for companion assistive systems. More detail may be found at
There is considerable research into the development of useful computer-assisted diagnostic tools for
medical imaging (medical CAD). Drivers include an aging population, increasing health-care costs,
availability of new treatments and interventions, a growing desire for early detection and treatment, and
improving technology. However, the rate of adoption of medical CAD has not met the expectations of
its developers. Why is this, and what can we do about it?
The talk will present some experiences in developing medical CAD systems for:
creating ground truth data from cineMRI as part of the development of a new cardiac monitoring
device (an on-going project),
detecting precursors to cerebral palsy in premature neonates using cranial ultrasound, and
identifying part of a coordinate system for the heart in echocardiograms.
In all three cases, the presentation will focus on experiences from the trenches with regard to task
identification, performance requirements, willingness to adopt the research results, and eventual outcome
of the work. Some ideas on increasing research impact will be included.
Peter Gregson received his Ph.D. in Electrical Engineering from the Technical University of Nova Scotia in 1988. He is the past-Director of the Innovation in Design Lab (iDLab), past-holder of the NSERC Chair in Design Innovation and a Professor in both the Department of Electrical and Computer Engineering and the School of Biomedical Engineering in the Faculty of Engineering at Dalhousie University. He has been heavily involved in product development in computer vision, medical imaging, electronics for vision, and intelligent control for over 35 years. Dr. Gregson's interest in design theory, methodology and teaching stems from his engineering, business and product development experience. He was the recipient of the 1996 Wighton Fellowship for innovative laboratory presentation, awarded to one faculty member in engineering and applied science in Canada annually.
ICIAR - International Conference on Image Analysis and Recognition
Send your comments to email@example.com