Searching for 'hidden' patterns in medical images

Dr. Rebecca Thornhill

Physicist, Department of Medical Imaging, The Ottawa Hospital

Adjunct Professor, Department of Systems and Computer Engineering, Carleton University

October 12, 2017 9:00 - 10:30

Mackenzie Building Room 4332, Carleton University

abstract

WRadiologists produce diagnoses on the basis of a combination of their training, experience, and individual judgment. Radiologists perceive image patterns and associate or infer a diagnosis consistent with those patterns. However, there will be an inevitable degree of variability in image interpretation as long as it relies primarily on human visual perception. Pattern analysis can provide a quantitative vocabulary for the otherwise subjective characteristics of lesions. Tools for automated pattern recognition can provide objective information to support clinical decision-making and may serve to reduce variability. To date we have applied quantitative shape and texture pattern analysis to a number of cancer and cardiovascular imaging applications. Several of these will be discussed, with particular focus on how pattern analysis can supplement conventional radiologic interpretation. Finally, these applications will also provide opportunities to discuss some of the pitfalls and challenges presented by these techniques.

biography

Dr Thornhill completed her M.Sc. and Ph.D. in Medical Biophysics from the University of Western Ontario, where her work focused on cardiac magnetic resonance imaging (MRI). She continued on to do postdoctoral research fellowships at McMaster University and the University of Toronto (Hospital for Sick Children). Since 2010, Dr Thornhill has been working as an Imaging Scientist in the Department of Medical Imaging at The Ottawa Hospital, with special interests in cardiac MRI as well as quantitative pattern analysis of medical images. She is also an adjunct research professor in the Department of Systems and Computer Engineering at Carleton University.