Limitations and requirements of diffusion tensor fiber tracking: an assessment using simulations

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Computational Neuroscience

Background:

• Diffusion tensor fiber tracking (DTFT) offers insights into in vivo brain connectivity.
• Validating DTFT is challenging due to the absence of a gold standard.
• Tracking reliability depends on data quality and algorithm robustness.

Purpose of the Study:

• To investigate the impact of various parameters on DTFT reliability using simulated models.
• To assess the relevance of simulation findings to real-world neuroimaging data.
• To guide the design of imaging sequences and tracking algorithms for improved DTFT.

Main Methods:

• DTFT was performed on two simulated models and volunteer data.
• Simulations analyzed the effects of SNR, anisotropy, curvature, fiber cross-section, background anisotropy, step size, and interpolation.
• Real data was used to validate simulation-derived trends.

Main Results:

• High SNR and anisotropy, coupled with interpolation and a low step size, generally yield the most reliable tracking results.
• Partial volume effects negatively impact tracking, particularly with anisotropic backgrounds and narrow fibers.
• Simulation results demonstrated similar trends to those observed in real neuroimaging data.

Conclusions:

• Simulations provide valuable insights into DTFT reliability across different parameters.
• Findings can inform the selection of imaging parameters and tracking strategies for specific brain structures.
• This research aids in determining trackable structures based on image quality.