Subject-specific analysis pipeline.

<p>This diagram summarizes the pre-processing, registration, and segmentation steps that were required to identify the seven regions-of-interest used in this work: caudate nucleus (CN), putamen (Pu), external and internal globus pallidus (GPe and GPi), substantia nigra (SN), subthalamic nucleus (STN), and thalamus (Tha). Axial SWI and T2-weighted images were linearly registered using nine degrees-of-freedom (DOF), and used to delineate GPe and GPi. Coronal SWI and T2-weighted images were also linearly registered using nine DOF, and used to delineate SN and STN. The purple arrows indicate spatial relations between structures, exploited to segment them (i.e. GPe and GPi are adjacent and simultaneously segmented). Diffusion-weighted images were corrected for motion, eddy-current, and susceptibility-induced distortions, and were then used to estimate fractional anisotropy (FA) and averaged <i>b0</i> images. Subsequently, axial and coronal high-resolution T2-weighted images were aligned with the averaged <i>b0</i> in order to resample segmentations of GPe, GPi, SN, and STN into diffusion native space, and then used for tractography seeding. CN, Pu and Tha were segmented from the FA image. Finally, resting-state fMRI (R-fMRI) data were corrected for motion and susceptibility-induced distortions. An average R-fMRI image was created and used as a template to register axial and coronal T2-weighted images, and <i>b0</i>, in order to resample all regions-of-interest into R-fMRI native space and then to generate R-fMRI activation maps.