Rigid body tracking for radiosurgery

• US 20070110289A1
• Filed: 11/16/2005
• Published: 05/17/2007
• Est. Priority Date: 11/16/2005
• Status: Active Grant

• Current Assignee: Accuray Incorporated
• Original Assignee: Accuray Incorporated
• Inventors: Kuduvalli, Gopinath, Fu, Dongshan
• Granted Patent: US 7,684,647 B2
• Time in Patent Office: Days
• US Class Current: 382/128
• CPC Class Codes: A61B 6/4458 the source unit or the detector unit being attached to robotic arms
• CPC Class Codes: G06V 10/24 Aligning, centring, orientation detection or correction of the image
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images

Abstract

A method and apparatus compares in-treatment X-ray images of a volume with pre-treatment reference images of the volume and determines a combined similarity measure. A combined registration search space is searched to maximize the combined similarity measure and to jointly register the in-treatment X-ray images with the pre-treatment images.

Claims

1. A method, comprising:

comparing a pair of X-ray images of a volume of interest with a corresponding pair of reference images from a 3-D scan of the volume of interest to determine a combined similarity measure between the pair of X-ray images and the corresponding pair of reference images; and

searching a combined registration search space to maximize the combined similarity measure, wherein the pair of X-ray images is jointly registered with the pair of reference images.

2. The method of claim 1, wherein the combined registration search space comprises a plurality of transformation parameters, the method further comprising converting the plurality of transformation parameters into a 3-D rigid transformation of the volume.

3. The method of claim 1, wherein searching the combined registration search space comprises:

searching for three in-plane translation parameters in two projections;

searching for two in-plane rotation parameters in the two projections; and

searching for a common out-of-plane rotation parameter in the two projections.

4. The method of claim 3, wherein searching for the three in-plane translation parameters in the two projections comprises:

searching initially for the three in-plane translation parameters using three-parameter multi-resolution matching; and

refining the three in-plane translation parameters using one of three-parameter sub-pixel matching and three-parameter gradient minimization.

5. The method of claim 3, wherein searching for the two in-plane rotation parameters in the two projections comprises:

searching initially for the two in-plane rotation parameters using two-parameter multi-resolution matching; and

refining the two in-plane rotation parameters using one of 2-D curve-fitting and gradient minimization.

6. The method of claim 3, wherein searching for the common out-of-plane rotation parameter in the two projections comprises:

searching initially for the common out-of-plane rotation parameter using a 1-D search method; and

refining the common out-of-plane rotation parameter using the 1-D search method and interpolation.

7. The method of claim 3, further comprising:

determining whether the plurality of transformation parameters have enough accuracy to achieve a desired registration accuracy; and

repeating the method of claim 3 if the plurality of transformation parameters do not have enough accuracy to achieve the desired registration accuracy;

else

applying the 3-D rigid transformation to the volume of interest.

8. An article of manufacture, including machine-accessible instructions that when accessed by a data processing system, cause the data processing system to perform a method, comprising:

comparing a pair of X-ray images of a volume of interest with a corresponding pair of reference images of the volume of interest to determine a combined similarity measure between the pair of X-ray images and the corresponding reference images; and

searching a combined registration search space to maximize the combined similarity measure, wherein the pair of X-ray images is jointly registered with the pair of reference images.

9. The article of manufacture of claim 8, wherein the combined registration search space comprises a plurality of transformation parameters, the method further comprising converting the plurality of transformation parameters into a 3-D rigid transformation of the volume.

10. The article of manufacture of claim 8, wherein searching the combined registration search space comprises:

searching for three in-plane translation parameters in two projections;

searching for two in-plane rotation parameters in the two projections; and

searching for a common out-of-plane rotation parameter in the two projections.

11. The article of manufacture of claim 10, wherein searching for the three in-plane translation parameters in the two projections comprises:

searching initially for the three in-plane translation parameters using three-parameter multi-resolution matching; and

refining the three in-plane translation parameters using one of three-parameter sub-pixel matching and three-parameter gradient minimization.

12. The article of manufacture of claim 10, wherein searching for the two in-plane rotation parameters in the two projections comprises:

searching initially for the two in-plane rotation parameters using two-parameter multi-resolution matching; and

refining the two in-plane rotation parameters using one of 2-D curve-fitting and gradient minimization.

13. The article of manufacture of claim 10, wherein searching for the common out-of-plane rotation parameter in the two projections comprises:

searching initially for the common out-of-plane rotation parameter using a 1-D search method; and

refining the common out-of-plane rotation parameter using the 1-D search method and interpolation.

14. The article of manufacture of claim 10, the method further comprising:

determining whether the plurality of transformation parameters have enough accuracy to achieve a desired registration accuracy; and

repeating the method of claim 10 if the plurality of transformation parameters do not have enough accuracy to achieve the desired registration accuracy;

else

applying the 3-D rigid transformation to the volume of interest.

15. A system comprising a processing device, wherein the processing device is configured to compare a pair of X-ray images of a volume of interest with a corresponding pair of reference images of the volume of interest to determine a combined similarity measure between the pair of X-ray images and the corresponding reference images, and wherein the processing device is further configured to search a combined registration search space to maximize the combined similarity measure, wherein the pair of X-ray images is jointly registered with the pair of reference images.

16. The system of claim 15, further comprising:

an imaging system coupled with the processing device to generate the pair of X-ray images of the volume of interest; and

means for generating a plurality of pairs of reference images from the 3-D scan of the volume of interest, the pairs of reference images comprising nominal in-plane reference images and a plurality of out-of-plane reference images in two projections.

17. The system of claim 15, wherein the combined registration search space comprises a plurality of transformation parameters, and wherein the processing device is further configured to convert the plurality of transformation parameters into a 3-D rigid transformation of the volume.

18. The system of claim 15, wherein to search the combined registration search space, the processing device is configured to search for three in-plane translation parameters in two projections, to search for two in-plane rotation parameters in the two projections, and to search for a common out-of-plane rotation parameter in the two projections

19. The system of claim 18, wherein to search for the three in-plane translation parameters in the two projections, the processing device is configured to search initially for the three in-plane translation parameters using three-parameter multi-resolution matching, and to refine the three in-plane translation parameters using one of three-parameter sub-pixel matching and three-parameter gradient minimization.

20. The system of claim 18, wherein to search for the two out-of-plane rotation parameters in the two projections, the processing device is configured to search initially for the two in-plane rotation parameters using two-parameter multi-resolution matching, and to refine the two in-plane rotation parameters using one of 2-D curve-fitting and gradient minimization.

21. The system of claim 18, wherein to search for the common out-of-plane rotation parameter in the two projections, the processing device is configured to search initially for the common out-of-plane rotation parameter using a 1-D search method, and to refine the common out-of-plane rotation parameter using the 1-D search method and interpolation.

22. The system of claim 18, wherein the processing device is further configured to determine whether the plurality of transformation parameters have enough accuracy to achieve a desired registration accuracy, to repeat the method of claim 18 if the plurality of transformation parameters do not have enough accuracy to achieve the desired registration accuracy, else to apply the 3-D rigid transformation to the volume of interest.

23. An apparatus, comprising:

means for comparing a pair of X-ray images of a volume of interest with a corresponding pair of reference images from a 3-D scan of the volume of interest to determine a combined similarity measure between the pair of X-ray images and the corresponding pair of reference images; and

means for searching a combined registration search space to maximize the combined similarity measure, wherein the pair of X-ray images is jointly registered with the pair of reference images.

24. The apparatus of claim 23, wherein the combined registration search space comprises a plurality of transformation parameters, the method further comprising converting the plurality of transformation parameters into a 3-D rigid transformation of the volume.