3D volume segmentation of MRA data sets using level sets

Extraction of vessel networks based on multiview projection and phase field model

“…Farag et al [20] proposed an approach using level sets in the 3D volume. The classification approach depends on initializing the level sets in the 3D volume, and the level sets evolve with time to yield the blood vessels.…”

Three-dimensional reference and stereotactic atlas of human cerebrovasculature from 7Tesla

“…Numerous methods have been proposed for automatic segmentation of vasculature and their taxonomy is given in Suri et al (2002). They are based on various principles, including expectation maximization (Wilson and Noble, 1999), region growing (Chapman et al, 2004), fuzzy connectedness (Forkert et al, 2009), skeletonization (Yim et al, 2000), active contours (Yu et al, 2008), level set for CTA (Manniesing et al, 2006) and for MRA (Farag et al, 2004), mathematical morphology (Zana and Klein, 2001), and atlas-related (Passat et al, 2005). These methods usually produce incomplete segmentations with numerous false positives.…”

From MIP image to MRA segmentation using fuzzy set theory

“…From an axis model, a representation of the central vessel axis is obtained and is used as reference for extracting the vessels boundaries; Passat et al [3] have proposed vessels segmentation from vascular atlas generated from any set of phase contrast MRA. • Level sets approaches: Farag et al [4] have described a vessels segmentation based on level sets principles introducing an automatic seed initialization; Lorigo et al [5] have introduced a curves evolution scheme for segmentation extended to 3D.…”