Automated sulcal segmentation using watersheds on the cortical surface.

Area of Science:

• Neuroimaging
• Computational Anatomy
• Medical Image Analysis

Background:

• The human cortical surface is complex and folded, with sulci defining distinct anatomical areas.
• Current methods often focus on sulcal spaces or curves, not the surrounding cortical regions.
• Accurate segmentation of the cortex is crucial for understanding brain structure and function.

Purpose of the Study:

• To develop a new technique for extracting and segmenting
• sulcal regions
• from magnetic resonance images (MRIs).
• To address oversegmentation issues common in watershed algorithms.
• To enable a complete parcellation of the cortical surface.

Main Methods:

• A watershed algorithm applied to a geodesic depth measure on the cortical surface.
• A postprocessing algorithm to merge oversegmented regions.
• Manual labeling of extracted sulcal regions via mouse click.
• Computation of a complete cortical surface parcellation scheme.

Main Results:

• Successful extraction of actual cortical regions surrounding sulci, termed
• sulcal regions
• .
• Demonstration of a postprocessing step to mitigate oversegmentation.
• Development of a manual labeling and parcellation scheme for the cortical surface.

Conclusions:

• The proposed method offers a novel approach to cortical surface segmentation by focusing on sulcal regions.
• This technique enhances anatomical parcellation and understanding of the human cortex.
• The method provides a foundation for comprehensive cortical mapping and analysis.