Sheet-like white matter fiber tracts: representation, clustering, and quantitative analysis

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Medical Image Analysis

Background:

• Accurate segmentation of white matter fiber tracts is crucial for understanding brain structure and function.
• Existing methods often lack automation and robust handling of complex, sheet-like tracts.
• Subject-specific analysis is essential for detecting subtle anatomical differences.

Purpose of the Study:

• To develop an automated, probabilistic method for segmenting subject-specific, sheet-like fiber tracts.
• To enable quantitative analysis of diffusion measures within segmented bundles.
• To identify and visualize spatial patterns of group differences in brain aging studies.

Main Methods:

• Introduced a novel automated and probabilistic approach for fiber tract segmentation.
• Developed a new method for medial surface generation to initialize tract clustering.
• Established point correspondences on medial representations for statistical analysis of diffusion measures.

Main Results:

• Successfully segmented sheet-like fiber tracts in a subject-specific manner.
• Generated statistics of diffusion measures by establishing point correspondences.
• Demonstrated the algorithm's capability in identifying spatial patterns of group differences in brain aging.

Conclusions:

• The proposed automated method offers robust segmentation of sheet-like fiber tracts.
• The approach facilitates quantitative analysis of diffusion metrics and group comparisons.
• This technique is valuable for population studies, particularly in understanding brain aging and neurological conditions.