Announcements I: February 21, 2008

Posted 10:55 AM, February 21, 2008, by jaruiz

"Representing Uncertainty via the Logarithm of Odds": Monday, February 25th, at 1:00 PM in the Clark Center S360. Kilian Pohl, PhD, from Harvard Medical School will be delivering a lecture on "Representing Uncertainty via the Logarithm of Odds." In this talk, Dr. Pohl will describe a new representation for capturing the uncertainty of objects in images based on the logarithm of odds. This representation addresses several problems in vision as it provides an intrinsic, probabilistic representation for combining and deforming objects. He will show how this representation preserves the statistical characteristics of interpolated shapes, which is an important aspect for many longitudinal neuroscience studies. He will also use the technology in order to solve the mean-field approximation in the level set framework. Conventional likelihood models are combined with a curve length prior on boundaries, and an approximate posterior distribution on labels is sought via the mean field approach. Optimizing the resulting estimator by gradient descent leads to a level set style algorithm where the level set functions are the logarithm of odds encoding of the posterior label probabilities. Applications with more than two labels are easily accommodated. The label assignment is accomplished by the maximum a posteriori rule, so there are no problems of "overlap" or "vacuum."

BIOGRAPHY:
Kilian Pohl received his doctorate in computer science from the Medical Vision Lab at MIT and is currently an instructor for Harvard Medical School. His main research area is computational image analysis with an emphasis on studying statistical models from a Bayesian perspective. Kilian has been the recipient of several awards such as last year's Medical Image Analysis--Medical Image Computing and Computer-Assisted Intervention (MICCAI) '06 Best Paper Prize for his publication, "Using the Logarithm of Odds to Define a Vector Space on Probabilistic Atlases." For more details about his research, please visit his website at http://people.csail.mit.edu/pohl.