Method and apparatus for fast automatic centerline extraction for virtual endoscopy
View Patent ↗A method for automatic centerline extraction for a virtual endoscopy image of an organ having a boundary surface includes centering on selected points of an initial path through the image, which is derived from an endoscopy dataset, respective spheres exhibiting respective maximal diameters short of contacting the boundary surface; and forming a centered path consecutively joining centers of the spheres.
1. A method for automatic centerline extraction for a virtual endoscopy image of an organ having a boundary surface, comprising:
deriving an endoscopy voxel dataset by using an endoscopy protocol;
deriving an initial path through said image between initial and final voxels of said dataset, said initial path exhibiting vertices;
centering respective spheres on selected ones of said vertices of said initial path, each sphere modeled as a polyhedron exhibiting respective vertices;
adjusting, iteratively, a center and a size of each sphere, wherein the centers are adjusted according to a translation force calculated on ones of said vertices of respective polyhedrons modeling the spheres colliding with said boundary surface to remove the spheres from collision, such that each sphere exhibits a respective maximal diameter short of contacting said boundary surface; and
forming a centered path consecutively joining centers of said spheres.
2. A method as recited in claim 1 , wherein:
said step of forming a centered path comprises smoothing said centered path to form a modified centered path.
3. A method as recited in claim 2 , wherein said step of forming a centered path comprises:
centering on selected points of said modified centered path respective spheres exhibiting respective maximal diameters short of contacting said boundary surface; and
forming a further modified centered path consecutively joining centers of said spheres.
4. A method as recited in claim 2 , wherein said step of forming a centered path comprises repetitively performing the steps of centering and forming for deriving a final centered path that has been modified to a desired degree.
5. A method for automatic centerline extraction for a virtual endoscopy image of an organ having a boundary surface, comprising:
deriving an endoscopy voxel dataset by using an endoscopy protocol;
deriving an initial path through said image between initial and final voxels of said dataset, said initial path exhibiting vertices;
centering respective spheres on selected ones of said vertices of said initial path, each sphere modeled as a polyhedron exhibiting respective vertices;
adjusting, iteratively, a center and a size of each sphere, wherein the centers are adjusted according to a translation force calculated on ones of said vertices of respective polyhedrons
modeling the spheres colliding with said boundary surface to remove the spheres from collision; and
forming a centered path consecutively joining centers of said spheres.
6. A method as recited in claim 5 , wherein:
said step of forming a centered path comprises smoothing said centered path to form a modified centered path.
7. A method as recited in claim 6 , wherein said step of forming a centered path comprises:
centering on selected points of said modified centered path respective spheres exhibiting respective maximal diameters short of contacting said boundary surface; and
forming a further modified centered path consecutively joining centers of said spheres.
8. A method as recited in claim 6 , wherein said step of forming a centered path comprises repetitively performing the steps of centering and forming for deriving a final centered path that has been modified to a desired degree.