Method for the 3d modeling of a tubular structure

• US 7,574,026 B2
• Filed: 01/23/2004
• Issued: 08/11/2009
• Est. Priority Date: 02/12/2003
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Movassaghi, Babak, Rasche, Volker
• Primary Examiner(s): Wu; Jingge
• Assistant Examiner(s): Tsai; Tsung-Yin
• Application Number:
  US10/545,184
• Publication Number:
  US 20060159326A1
• Time in Patent Office: 2,027 Days
• Field of Search: 382/131, 372/62, 372/19, 345/419
• US Class Current: 382/128
• CPC Class Codes: G06T 11/008 Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
• CPC Class Codes: G06T 17/10 Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
• CPC Class Codes: G06T 2211/404 Angiography
• CPC Class Codes: G06T 2211/412 Dynamic

Abstract

The invention provides a method for 3D modeling of a three-dimensional tubular structure of an examination object from 2D projection images (D) of the structure taken from different projection directions. The method has the following steps: reconstruction of a 3D image from the 2D projection images (D); selection of at least one 3D central line point (MO) in the 3D image, said 3D central line point being located in the structure; segmentation of 3D central line points (M) of the structure in the 3D image; forward projection of the 3D central line points (M), which have been segmented in the 3D image, into 2D projection images (D′); determination of border points of the structure in the 2D projection images (D′) on the basis of 3D central line points (Z) that have been projected in; and back-projection of the border points from the 2D projection image (D′) into the 3D image.

Claims

1. A method for the 3D modeling of a three-dimensional tubular structure of an examination object from a number of 2D projection images (D) of the tubular structure (H) taken from different projection directions, comprising the steps:

a) reconstruction of a 3D image (B) from the 2D projection images (D),

b) selection of at least one 3D central line point (M0) in the 3D image (B), said 3D central line point being located in the tubular structure (H),

c) segmentation of other 3D central line points (M) of the tubular structure (H) in the 3D image (B),

d) forward projection of the 3D central line points (M), which have been segmented in the 3D image (B), into the 2D projection images (D′

),

e) determination of border points of the tubular structure (H) in the 2D projection images (D′

) on the basis of 3D central line points (Z) that have been projected in, and

f) back-projection of the border points from the 2D projection images (D′

) into the 3D image (B).

2. A method as claimed in claim 1, characterized in that in step b) at least one start point and one end point of a 3D central line (M) in the tubular structure (H) is selected.

3. A method as claimed in claim 1, characterized in that the 3D image (B) is reconstructed to have a low resolution.

4. A method as claimed in claim 1, characterized in that the determination of border points of the tubular structure (H) in the 2D projection images (D) in step e) takes place automatically, in particular by taking account of the image values along a cross section through the tubular structure (H).

5. A method as claimed in claim 1, characterized in that, in parallel with the taking of the 2D projection images (D), a movement signal (E) that represents a periodic movement of the examination object is recorded, and in that the 3D image (B) is created from 2D projection images (D) taken in the same movement phases.

6. A method as claimed in claim 1, characterized in that an electrocardiogram or a respiratory movement signal is used as movement signal (E).

7. A method as claimed in claim 1, characterized in that the method is used to image the coronary vessels of a patient.

8. A device for the 3D modeling of a three-dimensional tubular structure of an examination object from a number of 2D projection images (D) of the tubular structure (H) taken from different projection directions, comprising:

a) a reconstruction unit for reconstructing a 3D image (B) from the 2D projection images (D),

b) selection means for selecting at least one 3D central line point (M0) in the 3D image (B), said 3D central line point being located in the tubular structure (H),

c) a segmentation unit for segmenting other 3D central line points (M) of the tubular structure (H) in the 3D image (B),

d) a forward projection unit for the forward projection of the 3D central line points (M), which have been segmented in the 3D image (B), into the 2D projection images (D′

),

e) a border point determination unit for determining border points of the tubular structure (H) in the 2D projection images (D′

) on the basis of the 3D central line points (Z) that have been projected in, and

f) a back-projection unit for the back-projection of the border points from the 2D projection images (D′

) into the 3D image (B).

9. A computer program embodied in a computer-readable medium and having computer programming means for instructing a computer to carry out the steps of the method:

a) reconstruction of a 3D image (B) from the 2D projection images (D),

b) selection of at least one 3D central line point (M0) in the 3D image (B), said 3D central line point being located in the tubular structure (H),

c) segmentation of other 3D central line points (M) of the tubular structure (H) in the 3D image (B),

d) forward projection of the 3D central line points (M), which have been segmented in the 3D image (B), into the 2D projection images (D′

),

e) determination of border points of the tubular structure (H) in the 2D projection images (D′

) on the basis of 3D central line points (Z) that have been projected in, and

f) back-projection of the border points from the 2D projection images (D′

) into the 3D image (B).