You are here
Dr. Roger Tam
Faculty of Applied Science
Dr. Tam is an Associate Professor in the Department of Radiology at UBC and a member of the MS/MRI Research Group in the Division of Neurology. He also has membership in the Biomedical Engineering Program in the Faculty of Applied Science and the Bioinformatics Program in the Faculty of Science.
Current Research Focus
Dr. Tam’s research interests are centered on the application of computer vision and machine learning methods to the quantitative analysis of medical images. The Tam laboratory’s current primary research direction is the use of magnetic resonance images to improve the understanding of neurological disorders such as multiple sclerosis. In general, the projects in the Tam laboratory relate to the following topics in medical imaging: in vivo imaging, imaging biomarkers, machine learning (big data analytics for medical images and personalized medicine), imaging artifacts and their impact on quantitative analysis, computational shape modelling & morphometrics, and medical informatics & distributed medical imaging systems.
“Nonlinear Dimensionality Reduction of Brain MRI Data for Classification Applications”
Classification is an important application in brain image analysis. Dimensionality reduction is a key component of many classification algorithms and is essential for reducing the complexity of mathematical methods for working with brain images. Recently, ground-breaking discoveries in the field of machine learning have rekindled a strong interest in a type of neural network called the deep belief net (DBN), which has been shown to be very effective at automatically learning patterns of similarity in large groups of small images by reducing the dimensionality of the images to form a simplified virtual landscape in which similar images occupy the same valleys and dissimilar images are separated by hills. In comparison to the previous data used for DBN experiments, brain images are of much higher dimension and have much more complex shape and intensity variations. Our long-term research goal is to develop new DBN methods that can be used to form simplified landscapes of groups of brain images such that the valleys represent meaningful similarities that can be used for classification. Using a very large database of magnetic resonance images of patients with multiple sclerosis, we will develop new methods for making DBNs work with large, complex images. If successful, this work will greatly benefit researchers in neuroimaging by allowing them to make much more effective use of their data.
Biomedical Engineering, Machine Learning, Medical Image Analysis, Computer-Assisted Surgery