Automated white-matter tractography using a probabilistic diffusion tensor atlas: Application to temporal lobe...

Area of Science:

• Neuroimaging
• Neuroscience
• Medical Physics

Background:

• Diffusion-weighted magnetic resonance imaging (dMRI) is crucial for mapping white matter tracts.
• Manual fiber tracking is time-consuming and requires expert interpretation.
• Integrating anatomical information enhances the reliability of dMRI tractography.

Purpose of the Study:

• To develop and validate an automated method for labeling white matter fiber tracts using a probabilistic atlas.
• To compare the efficacy of atlas-based tract identification with manual methods.
• To assess white matter integrity in temporal lobe epilepsy (TLE) patients using fractional anisotropy (FA).

Main Methods:

• Construction of a probabilistic fiber atlas from 21 healthy controls and 21 TLE patients.
• Manual identification of 23 white matter tracts, requiring ~40 hours of expert analysis.
• Application of the automated atlas-based method for tract labeling and FA analysis in TLE.

Main Results:

• The automated atlas-based method successfully labeled white matter fiber tracts.
• Atlas-derived and manual fiber selection yielded similar patterns of results.
• Effect sizes for between-group comparisons (TLE vs. controls) were comparable or larger with atlas-derived fibers.

Conclusions:

• Automated labeling of white matter tracts using a probabilistic atlas is a reliable and efficient alternative to manual methods.
• This approach enhances the analysis of white matter integrity in neurological disorders.
• The atlas-based method facilitates large-scale dMRI studies and clinical applications.