We propose a new geodesic based algorithm for fiber tracking in diffusion tensor imaging data. Our algorithm computes the multi-valued solutions from the Euler-Lagrange form of the geodesic equations. Compared to other geodesic based approaches, multi-valued solutions at each grid point are computed rather than just computing the viscosity solution. This allows us to compute fibers in a region with sharp orientation, or when the correct physical solution is not the fiber computed from the first arrival time. Compared to the classical stream-line approach, our method is less sensitive to noise, since the complete tensor is used. We also compare our algorithm with the Hamilton-Jacobi equation (HJ) based approach. We show that in the cases where U-shaped bundles appear, our algorithm can capture the underlying fiber structure while other approaches may fail. The results for synthetic and real data are shown for both methods.
|Plaats van productie||Eindhoven|
|Uitgeverij||Technische Universiteit Eindhoven|
|Status||Gepubliceerd - 2010|
|ISSN van geprinte versie||0926-4507|
Sepasian, N., Vilanova, A., Thije Boonkkamp, ten, J. H. M., & Haar Romeny, ter, B. M. (2010). An innovative geodesic based multi-valued fiber-tracking algorithm for diffusion tensor imaging. (CASA-report; Vol. 1027). Eindhoven: Technische Universiteit Eindhoven.