Faculty of Engineering and Natural Sciences
Information Theoretic Signal Processing
Departments of BME and CSEE, Oregon Health & Science University
Abstract: In this talk, we will discuss the recent applications of information theoretic optimality criteria an nonparametric estimators to a variety of problems that arise naturally in statistical signal processing an machine learning. These include system identification, clustering and segmentation, pattern classification, ad dimensionality reduction. Specific applications will include image segmentation and brain interface design. We will also present a brief overview of the current research projects that we are pursuing at the Oregon Health and Science University. These include medical laboratory test result verification and patient state visualization through statistical modeling, noninvasive brain computer interface design and cognitive state assessment using EEG, noninvasive lung tumor tracking via optimal recursive Bayesian state estimation, unobtrusive monitoring of elderly with a multimodal suite of sensors to assess cognitive capabilities and early detection of cognitive decline through continuous monitoring.
December 20, 2006, 14:40, FENS L045