Figure - PMC

Skip to main content

View full-text article in PMC

. Author manuscript; available in PMC: 2014 May 15.

Published in final edited form as: IEEE Trans Med Imaging. 2012 Oct 19;32(5):821–837. doi: 10.1109/TMI.2012.2225636

Fig. 13 — Results for the Computation-Sentences contrast obtained by the VEM and MCMC JDE versions compared to a GLM analysis. Top part, from left to right: NRL contrast maps for MCMC, VEM, and GLM, with sagittal, coronal and axial views from top to bottom lines (neurological convention: left is left). Middle part: plots of HRF estimates for VEM and MCMC in the two parcels circled in and on the maps: and , respectively. The canonical HRF shape is depicted in dashed line. Bottom part: axial maps of estimated regularization parameters β̂ for the two conditions, computation (comp.) and sentence (sent.), involved in the CS contrast. Parcels that are not activated by any condition are hidden. For all contrast maps, the input parcellation is superimposed in white contours.

Inline graphic — Results for the Computation-Sentences contrast obtained by the VEM and MCMC JDE versions compared to a GLM analysis. Top part, from left to right: NRL contrast maps for MCMC, VEM, and GLM, with sagittal, coronal and axial views from top to bottom lines (neurological convention: left is left). Middle part: plots of HRF estimates for VEM and MCMC in the two parcels circled in and on the maps: and , respectively. The canonical HRF shape is depicted in dashed line. Bottom part: axial maps of estimated regularization parameters β̂ for the two conditions, computation (comp.) and sentence (sent.), involved in the CS contrast. Parcels that are not activated by any condition are hidden. For all contrast maps, the input parcellation is superimposed in white contours.