In single-subject fMRI such as is used in presurgical mapping, processing decisions and choice of threshold can greatly affect the activation maps. In order to provide support for making these decisions, we propose a self-similarity approach that uses comparisons across randomly-created split-halves of the data and evaluating the maps using measures that come from a gambling model - the Bookmaker Informedness, Markedness and Matthews Correlation Coefficient - using an fMRI simulation. Early results indicate that features of Informedness and Matthews Correlation Coefficient data may be useful for making pipeline and threshold decisions.
This abstract and the presentation materials are available to members only; a login is required.