Assessing regularity and variability of cortical folding patterns of dicccols

Abstract

In our recent studies, we identified 358 common cortical landmarks named Dense Individualized and Common Connectivity-based Cortical Landmarks (DICCCOL), each of which possesses consistent fiber connection patterns across individuals and populations and is thus predictive of brain function. However, the regularity and variability of the cortical folding shape patterns of these DICCCOLs are unknown yet. This paper aims to employ statistical shape pattern descriptors based on the concept of visual words to quantitatively examine the folding shapes of DICCCOL landmarks. Our results demonstrated that the morphological cortical folding patterns are quite variable, but their regularity and variability are correlated with those of fiber connection patterns. This study suggests that cortical folding shape features might be complementary to connectivity-based features that can be jointly used for brain image registration and other human brain mapping applications.