CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Addressing partial volume effects using intra-subject locally adjusted cerebral blood flow images Authors:  Russell Shinohara - University of Pennsylvania Perelman School of Medicine (United States)
Alessandra Valcarcel - University of Pennsylvania (United States)
Simon Vandekar - University of Pennsylvania (United States)
Kristin Linn - University of Pennsylvania (United States) [presenting]
Abstract: Local cortical coupling is a subject-specific measure of the spatially varying relationship between cortical thickness and sulcal depth. Although it is a promising first step towards understanding local covariance patterns between two image-derived measurements, a more general coupling framework that can accommodate multiple volumetric imaging modalities is warranted. We first introduce Inter-Modal Coupling (IMCo), an analogue of local coupling in volumetric space that can be used to produce subject-level, spatially varying feature maps derived from two volumetric imaging modalities. We then leverage IMCo to address partial volume effects when studying localized relationships between gray matter density and cerebral blood flow (CBF) among participants in the Philadelphia Neurodevelopmental Cohort. We show that when CBF images are adjusted for partial volume effects at the subject level using our method, we have more power to detect non-linear interactions between age and sex in voxelwise analyses. We call the proposed IMCo-adjusted CBF images Intra-Subject Locally Adjusted Cerebral Blood Flow (ISLA-CBF) images.