Edinburgh Research Explorer

Neighbourhood tractography: a new approach to seed point placement for fibre tracking

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationProceedings of the British Chapter of the ISMRM, Guildford, UK
Publication statusPublished - 2006

Abstract

One area in which diffusion MRI (dMRI) based fibre tracking techniques have strong potential is in the segmentation of individual white matter structures (tracts) from dMRI images. The segmented areas can be used as regions of interest (ROIs) for studying tract specific effects of pathology [1]. This kind of tractography based segmentation is advantageous over more established ROI methods because the regions are calculated algorithmically, removing observer subjectivity; and because the regions can be arbitrarily shaped in three dimensions, matching the anatomy of the underlying fasciculus.
However, fibre tracking algorithms typically require as input a seed point, a location in dMRI space from which the algorithm begins to reconstruct a tract. The resultant segmentation can be very strongly dependent on the exact location of this point. This sensitivity can be problematic when trying to consistently segment a specific tract from several brain volumes. In this work we demonstrate the inconsistency of segmentations derived from registration based seed point placement, and describe an alternative approach that is based on maximising output similarity to a reference tract within a seeding neighbourhood in native image space.

ID: 18592877