Abstract:
The approach to the competition was to use the continuous valued output from support vector machine classification (SVC) models to estimate a heterogeneous set of temporal “features” as rated by three subjects in response to audio/video stimuli. The primary challenge to this approach was to provide the SVC with appropriate class labels for each feature. As part of my approach, I attempted to fully utilize the unique properties of the data and the parameters of the competition, which included the existence of interdependencies between the feature ratings, a high degree of consistency in the ratings across subjects, and access to the target data (season 3) images. Primary conclusions from this work indicate that 1) the quality of the labels used for training the SVCs is more important than both the quantity of training data and image voxel reduction/selection, and 2) human behavior (in this case feature rating responses to multimedia stimuli) can be much more consistent across individuals than brain structure or cognitive strategy, and the behavior’s temporal nature makes it inherently much easier to “align” than spatial brain data. We are excited by the potential of this work to provide a new approach to conceptualizing group analyses in the field of neuroimaging.
Stephen LaConte: http://www.bitc.bme.emory.edu/~slaconte
Stephen LaConte received his Ph.D. in biomedical engineering from the University of Minnesota in 2002. His graduate research was co-advised by Xiaoping Hu at the Center for Magnetic Resonance Research (CMRR) and Stephen Strother at the Neuroimaging, Visualization and Data Analysis group at the Minneapolis VA Medical Center (NEUROVIA). A major portion of his graduate research dealt with predictive modeling of functional magnetic resonance imaging (fMRI) data. Currently Stephen is an assistant professor, research track, at the Biomedical Imaging Technology Center (BITC) within the Georgia Institute of Technology and Emory University Department of Biomedical Engineering. He is involved in a broad range of fMRI and DTI projects and has NIH funding to research fMRI predictive modeling and prediction-based real-time fMRI.
