Nick Chbat & Kees van Zon

Room IS 503
Tuesday, November 8, 2005
Welcome Coffee - 10:30-11:00
Talk - 11:00 -12:00Noon

Abstract: In this talk we introduce Philips, present a tool for Bayesian networks, and discuss physiological modeling. We start with a short introduction of Philips and Philips Research North America, with an emphasis on clinical decision support systems (CDSS). We then present a tool that we developed for sensitivity analysis of arbitrary Bayesian networks. This tool, called BNSA, uses SMILE.NET to implement the sensitivity analysis method described in Coupé and van der Gaag (1998). Finally, we present some ongoing efforts toward implementing a CDSS based on physiological modeling.

Speaker's Bio:Nick finished his PhD on Applied Control Systems from Columbia University in 1995. He then worked for General Electric Corporate Research &Development for 6.5 years where he received the Dushman award (highest technical team achievement award.) He then worked at the Mayo Clinic for 3.5 years on biomedical engineering projects, and was a faculty at the Mayo Graduate School where he recently obtained the Teacher of the Year award. He joined Philips Research in June 2005, and is currently working on Clinical Decision Support Systems. Nick is a co-author of a book, 14 papers, and holds 11 patents.

Kees graduated Cum Laude in 1986 from the Technical University of Eindhoven, The Netherlands, where he studied Electrical Engineering. From 1986-1989 he worked in the Video Pre-Development department of Philips Consumer Electronics in Eindhoven, leading a project that designed ASICs for digital video feature processing. In 1989 he moved to Philips Research Briarcliff, NY, where he headed a team that worked on video ghost cancellation. This team eventually received an Engineering

Emmy Award for its contributions to the field, which included the standardization of a GCR (Ghost Cancellation Reference) signal that is transmitted in the vertical blanking of every terrestrial NTSC broadcast in the US. In 1992 Kees joined Philips Research in Eindhoven, where he worked on a project investigating software implementations of a TV set's video processing functionality. In 1997 he took a sabbatical to go traveling. He re-joined Philips Research in Briarcliff in 1998 to lead projects on Video Chain Optimization, Scalable Video Algorithms, and Video Processing for LCoS. In 2005 he changed over to medical research in the area of clinical decision support. He currently works on Bayesian networks for stroke diagnosis. Kees is co-author of a forthcoming book, 7 papers, and holds 10 patents.