Consistency Research Patterns Of Brain Cortex Folds

Folding is an essential shape characteristic of the cerebral cortex of the human brain and it is very important for understanding the structure of human brain. The folding shape varies across individual and populations, a variety of studies in the neuroimaging field have demonstrated that cortical folding patterns are predictive of cortical cytoarchitecture and function and distinguish normal and diseased brains, suggesting the regularity of cortical folding patterns in certain degree. Meanwhile, many other studies demonstrated that cortical folding patterns are remarkably variable. Recently, some researchers found358Regions of Interest(ROI), 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 ROIs are unknown yet. To address this question, this paper aims to examine the regularity and variability of the local-scale and meso-scale folding patterns of the358ROIs with cortical surfaces reconstructed from DTI data. Our basic idea is that for each corresponding ROI, the statistical shape features of the ROI are extracted from a group of cortical surfaces and then their regularity and variability are assessed at the population-level. We examined these ROIs by two different perspective: local-scale and meso-scale folding patterns. We employed statistical shape pattern descriptors based on the concept of visual words to quantitatively examine the local-scale folding shapes of these ROIs, and we use parametric representation of cortical surface patches using polynomials. In the end, we use the cosine similarity to measure the similarity between feature vectors. Experimental results demonstrated that some ROIs have substantially more regular folding patterns, while others exhibit much more variabilities. Overall, this study and its experimental results demonstrated that the morphological cortical folding patterns could be complementary to the fiber connection patterns for the same group of ROIs, thus offering novel insights into MRI-based vs DTI-based brain mapping methodologies:both shape-based and connectivity-based features might be jointly used for brain image registration and mapping applications in the future to leverage the complementary information in both feature spaces.