Method for editing 3D image segmentation maps
View Patent ↗A digital image editing method includes receiving a three-dimensional volume image of an anatomical object of interest, wherein the volume image is characterized by first and second mutually exclusive segmentation classes. The method also includes deriving a two-dimensional slice image from the volume image, selecting a single point on the slice image within the second segmentation class, and defining a plane in response to the selection of the single point, the plane dividing the second segmentation class into a target portion corresponding to the object and a remainder portion.
1. A digital image editing method, comprising:
receiving a three-dimensional volume image of an anatomical object of interest, wherein the volume image is characterized by first and second mutually exclusive segmentation classes;
deriving a two-dimensional slice image from the volume image;
selecting a single point on the slice image within the second segmentation class; and
defining a plane in response to the selection of the single point, the plane dividing the second segmentation class into a target portion corresponding to the object and a remainder portion.
2. The method of claim 1 , wherein the first segmentation class corresponds to at least one of a parenchyma, a vessel, and a chest wall, and the second segmentation class corresponds to a lesion.
3. The method of claim 1 , wherein the target portion corresponds to a lesion, and the remainder portion corresponds to at least one of a parenchyma, a vessel, and a chest wall.
4. The method of claim 1 , wherein the single point lies within the plane.
5. The method of claim 1 , further including creating a three-dimensional rendering of the target and remainder portions connected at the first plane.
6. The method of claim 5 , further including rendering the remainder portion transparent such that only the target portion is displayed in the three-dimensional rendering.
7. The method of claim 1 , further including defining an additional plane in response to the selection of the single point, wherein the plane and the additional plane each define a cross section of the object thereon.
8. The method of claim 7 , further including calculating a characteristic of each of the defined cross sections, and choosing one of the plane and the additional plane based on the calculated characteristics.
9. The method of claim 8 , wherein choosing one of the plane and the additional plane comprises selecting a plane that defines a minimum cross sectional area of a contiguous portion of the object and that passes through the single point.
10. The method of claim 8 , wherein the calculated characteristic is an area of the defined cross section.
11. A digital image editing method, comprising:
viewing a two-dimensional slice image of an anatomical object of interest corresponding to a three-dimensional volume image of the object, wherein the volume image is characterized by first and second mutually exclusive segmentation classes;
selecting a single point on the slice image within the second segmentation class;
forming a plurality of planes in response to the selection of the single point, each plane of the plurality of planes passing through the single point and defining a cross section of the object thereon;
calculating a characteristic of each cross section defined by the plurality of planes;
selecting a plane of the plurality of planes based on the calculated characteristics; and
classifying the second segmentation class along the selected plane.
12. The method of claim 11 , wherein calculating a characteristic of each cross section comprises calculating an area of a contiguous portion of each cross section defined by the plurality of planes.
13. The method of claim 11 , wherein selecting a plane of the plurality of planes comprises selecting a plane defining a minimum cross sectional area of a contiguous portion of the object.
14. The method of claim 11 , wherein the first segmentation class corresponds to at least one of a parenchyma, a vessel, and a chest wall, and the second segmentation class corresponds to a lesion.
15. The method of claim 11 , further including forming a three-dimensional rendering of the object comprising a target portion and a remainder portion connected by the selected plane.
16. The method of claim 15 , further including rendering the remainder portion transparent such that only the target portion is displayed in the three-dimensional rendering.
17. A digital image editing method, comprising:
receiving a three-dimensional volume image of an anatomical object of interest, wherein the volume image is characterized by first and second mutually exclusive segmentation classes;
deriving a three-dimensional view from the volume image;
selecting a single point on the three-dimensional view within the second segmentation class; and
defining a plane in response to the selection of the single point, the plane dividing the second segmentation class into a target portion corresponding to the object and a remainder portion.
18. The method of claim 17 , wherein the first segmentation class corresponds to a parenchyma and the second segmentation class corresponds to a lesion connected to one of a vessel and a chest wall.
19. The method of claim 17 , further including identifying an additional point on the three-dimensional view, the single point and the additional point both being disposed on the plane.
20. The method of claim 19 , wherein the additional point lies along a line of sight vector passing through the single point.