Image Segmentation with Optimization Techniques Used for Medical Imaging
<-- Return to the list
Date: 10-02-2007
Start Time:
1:00pm
End Time: 2:00pm
Speaker: Dorit Hochbaum, University of California , Berkeley
Location: Mudd 303
ABSTRACT
Image segmentation is to determine a partition to the "main" areas
of the image and identify them as associated with different types of
objects. This is of particular importance in medical imaging
where blur conceals information of critical importance. The
problem is modeled as minimization of deviation penalty, from the
captured colors of the pixels, and separation penalty, which is
associated with two adjacent images having different colors.
We describe a very efficient and best possible polynomial time
algorithm for the problem. This algorithm is more efficient than
most procedures based on spectral techniques, partitioning approaches
or heuristic clustering. We then demonstrate how to apply the
procedure for the purpose of de-blurring medical images.
BIO
Dorit S. Hochbaum is a full professor at UC Berkeley. She is a professor of Business Administration and of Industrial Engineering and Operations Research (IEOR). Professor Hochbaum holds a Ph.D from the Wharton school of Business at the University of Pennsylvania. Prior to joining UC Berkeley in 1981, Profeesor Hochbaum held a faculty position at Carnegie Mellon university's GSIA. Her research interests are in areas of approximation algorithms, supply chain management, efficient utilization of resources, design and analysis of computer algorithms and discrete and continuous optimization. Her recent applications work is on problems related to customer segmentation, prediction, ranking, group decision making and data mining. Recent theoretical work focuses on efficient techniques for network flow related problems and inverseon efficient techniques for network flow related problems and inverse problems, with applications varying from medical prognosis, error correction, financial risk assessment and prediction.
Professor Hochbaum served as the chair of the Manufacturing and Information Technology group at the Haas School of Business. She is the founder and director of the UC Berkeley Supply Chain Initiative. She is the founder and co-director of the RIOT project.
Professor Hochbaum is the author of over 130 papers that appeared in the Operations Research, Management Science and Theoretical Computer Science literature. She serves as department editor for Management Science department of Optimization and Modelling and on the editorial board of Networks and on the advisory board of Algorithms and Operations Research.
Professor Hochbaum was named in 2004 an honorary doctorate of Sciences of the University of Copenhagen, for her work on approximation algorithms. Professor Hochbaum was awarded in 2005 the title of INFORMS fellow.