 |
Center for Nonlinear Studies |
In this talk, I will present new results on the basic problem of using plaque motion estimation in order to predict atherosclerotic plaque rupture. I will begin with a review of optical flow methods and their application to motion estimation. I will then present motion trajectory results on actual videos, depicting periodic plaque deformations through time. We will then discuss ongoing future research on the use of amplitude-modulation frequency-modulation models for estimating the motion, as well as the use of image denoising methods for improving motion estimation. Dr. M.S. Pattichis is the director of the image and video Processing and Communications lab (ivPCL) at the department of Electrical and Computer Engineering at the University of New Mexico. He is currently the advisor to ten graduate students, including eight doctoral students pursuing research in digital image and video processing. Lab research has been supported through grants from the National Institutes of Health (NIH), Hewelett-Packard, Honeywell, Xilinx, and other industrial sponsors. His research is primarily focused on the development of new image models for non-stationary image analysis, video compression, and medical imaging. Dr. Pattichis has published over 60 conference and journal papers, and has also contributed 14 chapters to different books. He has been recently appointed as an associate editor for Pattern Recognition.
Host: DDMA Speaker Series