Multivariate classification methods have become an increasingly popular tool for identifying multiple regions of brain activity on an fMRI scan that together can differentiate disease physiology from control. We created a predictive model using fMRI data grouped by voxel count into the AAL atlas and were able to differentiate Chronic Fatigue Syndrome from control at a 80% accuracy rate. These results imply combining machine learning with fMRI data can be a powerful diagnostic tool.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Attendees will learn how to apply a multivariate pattern classification algorithm to fMRI data. A successfully developed algorithm lends a predictive and diagnostic capability to an already widely available data source (fMRI scans).


Destie Provenzano (Presenter)
Georgetown University

Stuart Washington, Georgetown University
James Baraniuk, Georgetown University

Presentation Materials: