Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz

Logo MUG-Forschungsportal

Gewählte Publikation:

SHR Neuro Krebs Kardio Lipid Stoffw Microb

Bauer, MH; Barbieri, S; Klein, J; Egger, J; Kuhnt, D; Freisleben, B; Hahn, HK; Nimsky, C.
Boundary estimation of fiber bundles derived from diffusion tensor images.
Int J Comput Assist Radiol Surg. 2011; 6(1):1-11 Doi: 10.1007/s11548-010-0423-x
Web of Science PubMed FullText FullText_MUG


Co-Autor*innen der Med Uni Graz
Egger Jan

Dimensions Citations:

Plum Analytics:

Scite (citation analytics):

Diffusion tensor imaging (DTI) is a non-invasive imaging technique that allows estimating the location of white matter tracts based on the measurement of water diffusion properties. Using DTI data, the fiber bundle boundary can be determined to gain information about eloquent structures, which is of major interest for neurosurgical interventions. In this paper, a novel approach for boundary estimation is presented. DTI in combination with diverse segmentation algorithms allows estimating the position and course of fiber tracts in the human brain. For additional information about the expansion of the fiber bundle, the introduced iterative approach uses the centerline of a tracked fiber bundle between two regions of interest (ROI). After sampling along this centerline, rays are sent out radially, discrete 2D contours are calculated, and the fiber bundle boundary is estimated in a stepwise manner. For this purpose, each ray is analyzed using several criteria, including anisotropy parameters and angle parameters, to find the boundary point. The novel method for automatically calculating the boundaries has been applied to several artificially generated DTI datasets. Multiple parameters were varied: number of rays per plane, sampling rate and sampled points along the rays. For the DTI data used in the experiments, the method yielded a dice similarity coefficient (DSC) between 74.7 and 91.5%. In this paper, a novel approach to retrieve significant information about the fiber bundle boundary from DTI data is presented. The method is a contribution to gather important knowledge about high-risk structures in neurosurgical interventions.
Find related publications in this database (using NLM MeSH Indexing)
Algorithms -
Brain - anatomy & histology
Diffusion Tensor Imaging - methods
Humans -
Nerve Fibers, Myelinated - chemistry

Find related publications in this database (Keywords)
Diffusion tensor imaging
Fiber tracking
Boundary estimation
Ray-based approach
© Med Uni Graz Impressum