Cloud Computing and Machine Learning Models for Cancer Diagnostics

Wednesday, April 15, 2009 - 17:30
TH 331
Ljubomir Buturovic, Chief Scientist, Pathwork Diagnostics, Inc., and Adjunct Professor, San Francisco State University
Cloud computing is a recently introduced distributed processing concept with potentially significant business and technological implications. In this talk I will present the use of cloud computing in the development of cancer diagnostics tests at Pathwork Diagnostics. Pathwork multivariate cancer assays are based on massively parallel measurement of the genomic profile of the patient's tumor specimen. We use machine learning-based models to transform this data into clinical report suitable for interpretation by the ordering physician. I will explain how cloud computing helps us overcome some of the computational challenges inherent in this approach.

Ljubomir Buturovic received his Ph.D. degree from the School of Electrical Engineering, University of Belgrade, Serbia. He is currently Adjunct Professor of Computer Science at San Francisco State University, and Chief Scientist at Pathwork Diagnostics, in Redwood City, California. Prior to joining SFSU and Pathwork Diagnostics he was Bioinformatics Director at Incyte Corporation. His research interests include applications of machine learning in life sciences, with emphasis on predictive and prognostic models for cancer diagnostics.