Corticospinal tractography with morphological, functional and diffusion tensor MRI: a comparative study of four...

Area of Science:

• Neuroimaging
• White Matter Tractography

Background:

• Diffusion tensor imaging (DTI) enables white matter bundle study but lacks anatomical validation, especially in complex crossing fiber regions.
• Deterministic tractography algorithms are widely used in clinical settings for reconstructing white matter pathways.

Purpose of the Study:

• To compare the performance of four deterministic tractography algorithms in reconstructing the corticospinal tract.
• To evaluate the anatomical accuracy and reliability of tractography using expert neuroradiologist assessment.

Main Methods:

• 15 healthy volunteers underwent MRI, including diffusion-weighted imaging.
• The corticospinal tract was reconstructed using Euler, Runge-Kutta second order (RK2), Runge-Kutta fourth order (RK4), and tensor deflection (TEND) algorithms.
• Quantitative and qualitative evaluations were performed, including assessment by two expert neuroradiologists.

Main Results:

• RK2 and TEND algorithms showed no significant quantitative difference in fiber count.
• Qualitative analysis revealed a consistent lack of fibers in the ventrolateral regions of functional regions of interest.
• Expert neuroradiologists demonstrated low concordance, with RK2 being the preferred algorithm.

Conclusions:

• Current deterministic tractography algorithms struggle to produce anatomically accurate white matter bundles.
• Algorithm selection was not based on mathematical robustness or fiber count.
• Validating tractography-derived anatomical data remains a significant challenge in neuroimaging.