US200 CN200

US200 CN200 provide population-matched brain templates that improve spatial normalization and enhance segmentation and registration accuracy for MRI analyses in Caucasian and Chinese cohorts.

Key Features:

• Population-Specific Templates: US200 is tailored for Caucasian subjects and CN200 is tailored for Chinese subjects to reflect population-specific anatomical characteristics.
• Quantitative Deformation Metrics: A voxel-wise index of deformation variability and a logarithmically transformed Jacobian determinant were used to quantify template construction variability.
• Cohort-Based Analysis: Construction and evaluation used subsets from the Human Connectome Project (HCP) and the Chinese Human Connectome Project (CHCP) as independent cohorts.
• Network-Specific Variability: The frontoparietal control network and dorsal attention network exhibited higher deformation variability than primary networks.
• Improved Segmentation and Registration: Population-matched templates increased segmentation and registration accuracy, whereas population-mismatched templates reduced performance.
• Sample-Size Modeling: Variability associated with volumetric template construction was modeled as a power function of sample size to assess sample-size effects.
• Language-Related Differences: Analysis revealed significant differences in deformation variability between populations in language-related areas.

Scientific Applications:

• Structural MRI Morphometry: Enables more accurate morphometric analyses by providing templates that match population-specific brain anatomy.
• Spatial Normalization in MRI Pipelines: Improves spatial normalization, segmentation, and registration steps in structural and functional MRI processing.
• Cross-Population Neuroimaging: Facilitates cross-cultural and cross-population comparisons between Caucasian and Chinese cohorts.
• Network and Language Mapping: Supports more reliable mapping of networks with high deformation variability, including the frontoparietal control, dorsal attention, and language-related regions.

Methodology:

Assessment used subsets from the HCP and CHCP; deformation variability was quantified with a voxel-wise index and a logarithmically transformed Jacobian determinant; variability was modeled as a power function of sample size; deformation variability was compared across functional networks and language-related areas.