Medical image analysis topics, with emphasis on imaging pattern analysis and classification for constructing diagnostic imaging biomarkers

Prof. Christos Davatzikos
University of Pennsylvania

Abstract:  medical image analysis has been a rapidly growing field, as imaging moves from its industrial age to its information age. Image analysis algorithms related to deformable image registration and shape and connectivity analysis are reviewed, with emphasis on deriving imaging patterns that have diagnostic and prognostic value. Applications to neuroimaging of Alzheimers Disease, schizophrenia and other brain disorders are presented.

Short Biography: Dr. Christos Davatzikos is Professor of Radiology at the University of Pennsylvania. He holds joint affiliations with the Departments of Bioengineering and Electrical Engineering at Penn, and the Applied Math group, and he directs the section for biomedical image analysis. Dr. Davatzikos obtained his undergraduate degrees from the National Technical University of Athens, Greece, and graduate degrees from Johns Hopkins University. Prior to coming to Penn, he held appointments as Assistant and Associate Professor at Johns Hopkins School of Medicine, where he founded the Center from Biomedical Image Computing.  His work is in the general field of computerized analysis of medical images, with emphasis on analysis of magnetic resonance images of the brain.  Dr. Davatzikos' interests are in the development of automated and quantitative computerized image analysis algorithms for image segmentation, registration, spatial normalization, and multi-variate quantitative analysis of high-dimensionality information using pattern classification methods, typically including structural and functional image measurements.  He is also interested in for methodologies involving population-based statistical atlases for surgical plan optimization.