A new proposal for 3D fiber tracking in synthetic diffusion tensor magnetic resonance images

In this paper a new fiber tracking algorithm to be used with diffusion tensor fields acquired via magnetic resonance imaging is developed. The research effort in tracking fibers of nervous cells in the human brain affords many benefits to the medical community. The proposed method is an improvement over a previously developed algorithm that uses both geometrical and probabilistic criteria. The new scheme offers better results in regions with branching and crossing fibers, showing better computational and robustness properties. Like its predecessor, it also offers the capability of reporting an uncertainty value for the computed tracts.