The Relevance Voxel Machine: Bayesian image-based prediction: Dr. Mert Rory Sabuncu
- Starts: 1:00 pm on Monday, February 4, 2013
- Ends: 2:00 pm on Monday, February 4, 2013
Abstract The Relevance Voxel Machine (RVoxM) is a dedicated Bayesian model for making predictions based on medical imaging data. In contrast to the generic machine learning algorithms that have often been used for this purpose, the method is designed to utilize a small number of spatially clustered sets of voxels that are particularly suited for clinical interpretation. RVoxM automatically tunes all its free parameters during the training phase, and offers the additional advantage of producing probabilistic prediction outcomes. We demonstrate RVoxM as a regression model by predicting age from volumetric gray matter segmentations, and as a classification model by distinguishing patients with Alzheimer’s disease from healthy controls using surface-based cortical thickness data. We further illustrate the use of RVoxM in examining subtle and complex relationships between neuro-degeneration and polygenic risk to Alzheimer's disease. Our results indicate that RVoxM yields biologically meaningful models, while providing state-of-the-art predictive accuracy. Bio Mert R. Sabuncu is Assistant Professor at the A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School and a Research Affiliate at the Computer Science and Artificial Intelligence Lab (CSAIL) of MIT. He completed his PhD in Electrical Engineering at Princeton University in 2006, after which he was a post-doctoral researcher in Prof. Polina Golland's lab at MIT. His current research interests are in statistical methods for multivariate, longitudinal and survival analysis of biomedical image data, computational imaging genetics, and general image processing algorithms, with a focus on neuroimage data.
- MCS 148