Method and system of affine registration of inter-operative two dimensional images and pre-operative three dimensional images

• US 20050203385A1
• Filed: 01/14/2005
• Published: 09/15/2005
• Est. Priority Date: 01/21/2004
• Status: Active Grant

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Xu, Chenyang, Sauer, Frank, Sundar, Hari
• Granted Patent: US 7,450,743 B2
• Time in Patent Office: Days
• US Class Current: 600/427
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/10121 Fluoroscopy
• CPC Class Codes: G06T 2207/20072 Graph-based image processing
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06T 7/38 Registration of image sequences

Abstract

A system and method for registering a sequence of inter-operative two dimensional (2D) medical images of a target feature with pre-operative three dimensional (3D) medical images of the target feature is disclosed. The 3D image of the target feature is converted to a first skeletal graph. The 2D image of the target feature is converted to a second skeletal graph. Graph matching of the first and second skeletal graphs is performed to obtain a rough alignment of the graphs, and the first and second skeletal graph registered.

Claims

1. A method for registering a sequence of inter-operative two dimensional (2D) medical images of a target feature with pre-operative three dimensional (3D) medical images of the target feature comprising the steps of:

converting the 3D image of the target feature to a first skeletal graph;

converting the 2D image of the target feature to a second skeletal graph;

performing graph matching of the first and second skeletal graph to obtain a rough alignment of the graphs; and

registering the first and second skeletal graph.

2. The method of claim 1 wherein the target feature is blood vessels in a particular tissue region.

3. The method of claim 1 wherein the step of converting the 3D image of the target feature to a first skeletal graph further comprises the steps of:

segmenting the target feature from the 3D image; and

extracting a centerline from the segmented 3D image.

4. The method of claim 3 wherein the centerline is extracted using a parallel thinning algorithm.

5. The method of claim 1 wherein the step of converting the 2D image of the target feature to a second skeletal graph further comprises the steps of:

segmenting the target feature from the 2D image; and

studying a flow of a contrast agent through the target feature.

6. The method of claim 5 wherein the step of studying the flow of the contrast agent further comprises the steps of:

recording an arrival time and departure time of the flow of the contrast agent at each intersection of the target feature;

based on the arrival time and departure time determining if the intersection is a connected intersection or an overlapping intersection.

7. The method of claim 1 wherein the step of performing graph matching further comprises using rooted directed acyclic graphs.

8. The method of claim 1 wherein the step of registering the first and second skeletal graphs further comprises the step of:

using an iterative closest point algorithm to refine the registration.

9. The method of claim 1 wherein registration of the first and second skeletal graph is maintained by tracking the target feature in the 2D image.

10. The method of claim 1 wherein a monoplanar fluoro image is used to perform registration of the 2D and 3D images.

11. The method of claim 1 wherein the sequence of 2D images are sequential in time.

12. The method of claim 1 wherein the target feature is associated with a human organ.

13. The method of claim 12 wherein the organ is a liver.

14. A system for registering a sequence of inter-operative two dimensional (2D) images of a target feature with pre-operative three dimensional (3D) images of the target feature, the system comprising:

a 2D imaging system for imaging a target feature;

a database for storing 3D images of the target feature;

a processor for receiving and processing the 2D images, the processor performing the following steps;

i). converting the 3D image of the target feature to a first skeletal graph;

ii). converting the 2D image of the target feature to a second skeletal graph;

iii). performing graph matching of the first and second skeletal graph to obtain a rough alignment of the graphs; and

iv). registering the first and second skeletal graph; and

a display for displaying the registered images.

15. The system of claim 14 wherein the target feature is blood vessels in a particular tissue region.

16. The system of claim 14 wherein the step of converting the 3D image of the target feature to a first skeletal graph further comprises the steps of:

segmenting the target feature from the 3D image; and

extracting a centerline from the segmented 3D image.

17. The system of claim 16 wherein the centerline is extracted using a parallel thinning algorithm.

18. The system of claim 14 wherein the step of converting the 2D image of the target feature to a second skeletal graph further comprises the steps of:

segmenting the target feature from the 2D image; and

studying a flow of a contrast agent through the target feature.

19. The system of claim 18 wherein the step of studying the flow of the contrast agent further comprises the steps of:

recording an arrival time and departure time of the flow of the contrast agent at each intersection of the target feature;

based on the arrival time and departure time determining if the intersection is a connected intersection or an overlapping intersection.

20. The system of claim 14 wherein the step of performing graph matching further comprises using rooted directed acyclic graphs.

21. The system of claim 14 wherein the step of registering the first and second skeletal graphs further comprises the step of:

using an iterative closest point algorithm to refine the registration.

22. The system of claim 14 wherein registration of the first and second skeletal graph is maintained by tracking the target feature in the 2D image.

23. The system of claim 14 wherein a monoplanar fluoro image is used to perform registration of the 2D and 3D images.

24. The system of claim 14 wherein the sequence of 2D images are sequential in time.

25. The system of claim 14 wherein the target feature is associated with a human organ.

26. The system of claim 25 wherein the organ is a liver.