Rapid convolution based large deformation image matching via landmark and volume imagery

• US 6,226,418 B1
• Filed: 11/05/1998
• Issued: 05/01/2001
• Est. Priority Date: 11/07/1997
• Status: Expired due to Term

• Current Assignee: Washington University In St Louis
• Original Assignee: Washington University In St Louis
• Inventors: Miller, Michael I., Joshi, Sarang C., Christensen, Gary E.
• Primary Examiner(s): Tran, Phuoc
• Assistant Examiner(s): Mariam, Daniel G.
• Application Number:
  US09/186,359
• Time in Patent Office: 908 Days
• Field of Search: 382/131, 382/151, 382/217, 382/218, 382/276, 382/280, 382/291, 382/294, 345/425, 378/4, 600/424, 600/426
• US Class Current: 382/294
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 3/14 Transformations for image registration, e.g. adjusting or mapping for alignment of images
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06T 7/37 using transform domain methods
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 10/754 involving a deformation of the sample pattern or of the reference pattern; Elastic matching

Abstract

An apparatus and method for image registration of a template image with a target image with large deformation. The apparatus and method involve computing a large deformation transform based on landmark manifolds, image data or both. The apparatus and method are capable of registering images with a small number of landmark points. Registering the images is accomplished by applying the large deformation transform.

Claims

1. A method for registering a template image with a target image, the template image containing a plurality of points, and the target image containing a plurality of points, comprising the steps of:

defining a manifold landmark point in said template image;

identifying a point in said target image corresponding to said defined a manifold landmark point;

computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image, using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image; and

registering said template image with said target image using said large deformation transform.

7. The method of claim 1, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, non-affine transform.

8. The method of claim 1, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, higher order transform.

13. The method of either claim 1 or 5, wherein said step of defining manifold landmark points in said template image includes the substep of:

defining manifold landmark points of dimension greater than zero in said template image.

14. The method of either claim 1 or 5, wherein said step of defining manifold landmark points in said template image includes the substep of:

defining individual points in said template image.

15. The method of either claim 1 or 5, wherein said step of defining manifold landmark points in said template image includes the substep of:

defining points of a curve in said template image.

16. The method of either claim 1 or 5, wherein said step of defining manifold landmark points in said template image includes the substep of:

defining points of a surface in said template image.

17. The method of either claim 1 or 5, wherein said step of defining manifold landmark points in said template image includes the substep of:

defining points of a volume in said template image.

18. The method of either claim 1, 3 or 5, wherein said step of computing large deformation transform includes the substep of:

using a linear differentiable operator.

19. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using periodic boundary values.

20. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using infinite boundary values.

21. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using mixed Dirichlet and Neumann boundary values.

22. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using Neumann boundary values.

23. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using Dirichlet boundary values.

24. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

accessing stored pre-computed transform values.

25. The method of either claim 1, 3 or 5, wherein said step of computing said large deformation transform includes the substep of:

using a fast Fourier transform.

2. An apparatus for registering a template image with a target image, the template image containing a plurality of points, and the target image containing a plurality of points, comprising:

means for defining a manifold landmark point in said template image;

means for identifying a point in said target image corresponding to said defined a manifold landmark point;

means for computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image; and

means for registering said template image with said target image using said large deformation transform.

26. The apparatus of claim 2, wherein said large deformation transform is a diffeomorphic, non affine transform.

27. The apparatus of claim 2, wherein said large deformation transform is a diffeomorphic, higher order transform.

32. The apparatus of either claim 2 or 6, wherein said means for defining manifold landmark points in said template image comprises:

means for defining manifold landmark points of dimension greater than zero in said template image.

33. The apparatus of either claim 2 or 6, wherein said means for defining manifold landmark points in said template image comprises:

means for defining individual points in said template image.

34. The apparatus of either claim 2 or 6, wherein said means for defining manifold landmark points in said template image comprises:

means for defining points of a curve in said template image.

35. The apparatus of either claim 2 or 6, wherein said means for defining manifold landmark points in said template image comprises:

means for defining points of a surface in said template image.

36. The apparatus of either claim 2 or 6, wherein said means for defining manifold landmark points in said template image comprises:

means for defining points of a volume in said template image.

37. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using a linear differentiable operator.

38. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using periodic boundary values.

39. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using infinite boundary values.

40. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using mixed Dirichlet and Neumann boundary values.

41. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using Neumann boundary values.

42. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for using Dirichlet boundary values.

43. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

means for accessing stored pre-computed transform values.

44. The apparatus of either claim 2, 4 or 6, wherein said means for computing said large deformation transform comprises:

using a fast Fourier transform.

3. A method for registering a template image with a target image, comprising the steps of:

defining a distance measure between said target and said template image;

computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image, using an image transformation operator and an image transformation boundary value relating said template image to said target image; and

registering said template image with said target image using said large deformation transform.

9. The method of claim 3, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, non-affine transform.

10. The method of claim 3, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, higher order transform.

4. An apparatus for registering a template image with a target image, comprising:

means for defining a distance measure between said target image and said template image;

means for computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image using an image transformation operator and an image transformation boundary value relating said template image to said target image; and

means for registering said template image with said target image using said large deformation transform.

28. The apparatus of claim 4, wherein said large deformation transform is a diffeomorphic, non affine transform.

29. The apparatus of claim 4, wherein said large deformation transform is a diffeomorphic, higher order transform.

5. A method for registering a template image with a target image, wherein the template image contains a plurality of points and the target image contains a plurality of points, comprising the steps of:

defining a manifold landmark point in said template image;

identifying a point in said target image corresponding to said defined manifold landmark point;

computing a first large deformation transform using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image;

defining a distance measure between said target and said template image;

computing a second large deformation transform using an image transformation, operator and an image transformation boundary value relating said template image to said target image;

fusing said first and second large deformation transforms; and

registering said template image with said target image using said fused large deformation transforms.

11. The method of claim 5, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, non-affine transform.

12. The method of claim 5, wherein the step of computing said large deformation transform further includes the substep of computing a diffeomorphic, higher order transform.

6. An apparatus for registering a template image with a target image, wherein the template image contains a plurality of points and the target image contains a plurality of points, comprising:

means for defining a manifold landmark point in said template image;

means for identifying a point in said target image corresponding to said defined manifold landmark point;

means for computing a first large deformation transform using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image;

means for defining a distance measure between said target and said template image;

means for computing a second large deformation transform using an image transformation operator and an image transformation boundary value relating said template image to said target image;

means for fusing said first and second large deformation transforms; and

means for registering said template image with said target image using said fused large deformation transforms.

30. The apparatus of claim 6, wherein said large deformation transform is a diffeomorphic, non affine transform.

31. The apparatus of claim 6, wherein said large deformation transform is a diffeomorphic, higher order transform.

45. An apparatus for registering a template image with a target image, the template image containing a plurality of points, and the target image containing a plurality of points, comprising:

a pointing device for defining a manifold landmark point in said template image and for identifying a point in said target image corresponding to said defined manifold landmark point

a first data processing unit for computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image, using a manifold landmark transformation operator and manifold landmark transformation boundary values relating said manifold landmark points in said template image to said corresponding points in said target image; and

a second data processing unit for registering said template image with said target image using said large deformation transform.

46. An apparatus for registering a template image with a target image comprising:

a distance quantifier for defining a distance between said target image and said template image;

a first data processing unit for computing a large deformation transform that is one-to-one and onto for any point selected in the template and target image using an image transformation operator and an image transformation boundary value relating said template image to said target image; and

a second data processing unit for registering said template image with said target image using said large deformation transform.

47. An apparatus for registering a template image with a target image, wherein the template image contains a plurality of points and the target image contains a plurality of points, comprising:

a pointing device for defining a manifold landmark point in said template image and for identifying a point in said target image corresponding to said defined manifold landmark point;

a first data processing unit for computing a first large deformation transform using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image;

a distance quantifier for defining a distance between said target image and said template image;

a second data processing unit for computing a second large deformation transform using an image transformation operator and an image transformation boundary value relating said template image to said target image; and

a third data processing unit for fusing said first and second large deformation transforms; and

a fourth data processing unit for registering said template image with said target image using said fused large deformation transforms.

48. A method for registering a template image with a target image, comprising the steps of:

defining a distance measure between said target and said template image;

computing a diffeomorphic transform using said distance measure, an image transformation operator, and an image transformation boundary value relating said template image to said target image; and

registering said template image with said target image using said diffeomorphic transform.

49. An apparatus for registering a template image with a target image, comprising:

means for defining a distance measure between said target image and said template image;

means for computing a diffeomorphic transform using an image transformation operator, said distance measure, and an image transformation boundary value relating said template image to said target image; and

means for registering said template image with said target image using said diffeomorphic transform.

50. A method for registering a template image with a target image, wherein the template image contains a plurality of points and the target image contains a plurality of points, comprising the steps of:

defining a manifold landmark point in said template image;

identifying a point in said target image corresponding to said defined manifold landmark point;

computing a first diffeomorphic transform using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image;

defining a distance measure between said target and said template image;

computing a second diffeomorphic transform using an image transformation operator and an image transformation boundary value relating said template image to said target image;

fusing said first and second diffeomorphic transforms; and

registering said template image with said target image using said fused diffeomorphic transforms.

51. An apparatus for registering a template image with a target image, wherein the template image contains a plurality of points and the target image contains a plurality of points, comprising:

means for defining a manifold landmark point in said template image;

means for identifying a point in said target image corresponding to said defined manifold landmark point;

means for computing a first diffeomorphic transform using a manifold landmark transformation operator and a manifold landmark transformation boundary value relating said manifold landmark point in said template image to said corresponding point in said target image;

means for defining a distance measure between said target and said template image;

means for computing a second diffeomorphic transform using an image transformation operator and an image transformation boundary value relating said template image to said target image;

means for fusing said first and second diffeomorphic transforms; and

means for registering said template image with said target image using said fused diffeomorphic transforms.