White Matter Fiber Tracking Method with Adaptive Correction of Tracking Direction.

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Medical Image Analysis

Background:

• Deterministic fiber tracking is efficient for estimating brain structural connectivity.
• Classical deterministic methods struggle with accuracy in crossing fiber regions.
• Need for improved noninvasive methods in clinical neuroscience.

Purpose of the Study:

• To introduce an adaptive correction-based deterministic white matter fiber tracking method (FTACTD).
• To address the limitations of existing deterministic methods in complex fiber regions.
• To enhance the accuracy and reliability of brain structural connectivity mapping.

Main Methods:

• FTACTD adaptively adjusts deflection direction based on tensor matrix and adjacent voxel data.
• Mimics natural fiber deflection angles and directions using diffusion tensor shape.
• Employs both forward and reverse tracking with validation on simulated and real DWI data.

Main Results:

• FTACTD achieved the highest number of valid bundles (13) on simulated data.
• Significantly reduced incorrect fiber bundles and no connections compared to FACT and SD_Stream.
• Demonstrated superior accuracy, completeness, and continuity in in vivo experiments.

Conclusions:

• FTACTD offers superior white matter fiber tracking results.
• Provides a robust methodological basis for diagnosing and treating brain disorders.
• Enhances understanding of white matter deficits and abnormalities.