Reconstructions and Representations of Cerebral Cortex

点击翻译按钮获取中文摘要

Project Summary This project will generate extensive new findings about cortical organization and connectivity in humans and nonhuman primates using high quality, multimodal datasets provided by our collaborators and by the young adult Human Connectome Project (YA-HCP). It will accelerate progress by freely sharing the resulting tools and experimental data with the scientific community. The first aim will provide a critically needed evaluation of non-invasive connectivity measures in relation to invasive anatomical tracers in macaque monkeys. Different methods for estimating fMRI-derived functional connectivity will be evaluated in order to determine which approach best correlates with `ground truth' tracer-based anatomical connectivity. This aim will also generate new insights about cortical evolution by comparing areal organization across humans, macaques, and marmosets using a novel approach in which areal features (myelin maps, resting-state networks, and identified homologous areas) constrain the registration between species. The second aim will focus on cortical organization in individual subjects using refined HCP-style analysis tools. It will optimize and evaluate intersubject alignment (using a recently developed Multimodal Surface Matching method) and individual-subject parcellation (using a machine learning based areal classifier). An important outcome will be recommendations of `best practice' for other projects that acquire less fMRI data than in the YA- HCP. These data will also be used to characterize individual variability of human cortical areas. For each of 180 areas, individual differences in size and topology (neighborhood relationships) will be examined for heritability and for symmetry across the two hemispheres. Additional analyses will reveal whether some areas are reproducibly absent in some individuals and whether `novel' areas are present in some subjects. The third aim is to enhance the capabilities of the Connectome Workbench visualization and analysis platform and the BALSA database that were introduced during previous grant periods. Enhancements to Connectome Workbench will: (i) enable non-invasive electrophysiological (MEG/EEG) and invasive neurophysiological data to be integrated with MRI data and atlas-based connectivity data (ii) facilitate interoperability across different atlases, (iii) improve interactive `HCP-style' analysis capabilities, and (iv) enable cortical layer-based analyses. Enhancements to the BALSA database include: (i) a WebGL-based web-viewer for interactive online visualization with special focus on the unique data generated by this project, (ii) an online spatial localization tool and (iii) support for uploading scene files to BALSA from 5 other software platforms besides Workbench.