Study of the sources of errors in fiber tracking using diffusion tensor MRI

Diffusion tensor fiber tracking provides exciting new opportunities to study central nervous system (CNS) anatomy. In this study, simulated models are used to study the effect of anisotropy, partial voluming, and signal to noise ratio (SNR) on the quality of the tracking procedure. Effect of SNR on the tract path is also studied on real data. Results show that at low SNR values, white matter (WM) areas become more anisotropic and the angles between the eigenvectors of the tract path are over-estimated, which causes a detrimental effect on the tract length. Partial volume effects are shown to have a noticed effect on the anisotropic areas, thus affecting the tracking procedure.