A Method for Automated Cortical Surface Registration and Labeling

Area of Science:

• Neuroimaging
• Computational Anatomy
• Medical Image Analysis

Background:

• Accurate registration and delineation of human brain anatomical features in MRI are crucial for studying brain development and diseases.
• Challenges exist in achieving automatic, efficient, and precise cortical surface registration and landmark delineation due to anatomical variability.

Purpose of the Study:

• To present a novel method for fast and accurate registration, labeling, and sulcal delineation of brain images.
• To improve the mapping of anatomical features like regions of interest (ROIs) and sulcal curves (sulci) on subject brain surfaces.

Main Methods:

• A multiresolution, curvature-based approach for registering subject brain surface models to an atlas.
• Utilizing geodesic curvature flow formulated with a level set method on the cortical surface to refine sulcal curve locations.
• Incorporating curvature-based weighting to guide curves to sulcal fundi.

Main Results:

• The proposed method achieves fast and accurate registration and delineation of brain image features.
• Automatically delineated sulcal curves closely match manually delineated ones, indicating high accuracy.
• The technique effectively refines landmark locations to follow true sulcal fundi.

Conclusions:

• The developed method offers an accurate and efficient solution for brain image analysis.
• It holds potential for advancing research in brain development and neurological disorders.
• The approach successfully addresses the challenges of anatomical variability in brain mapping.