Registration of volumetric images which are relatively elastically deformed by matching surfaces

• US 5,633,951 A
• Filed: 06/07/1995
• Issued: 05/27/1997
• Est. Priority Date: 12/18/1992
• Status: Expired due to Fees

• Current Assignee: US Philips Corporation (Koninklijke Philips N.V.)
• Original Assignee: North America Philips Corporation
• Inventors: Moshfeghi, Mehran
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): CHANG, JON CARLTON
• Application Number:
  US08/482,690
• Time in Patent Office: 720 Days
• Field of Search: 382/131, 382/154, 382/285, 382/294, 364/413.13, 364/413.22, 395/119, 395/125, 395/127
• US Class Current: 382/154
• CPC Class Codes: G06T 3/153 using elastic snapping
• CPC Class Codes: G06T 7/344 involving models
• CPC Class Codes: G06V 10/755 Deformable models or variational models, e.g. snakes or active contours
• CPC Class Codes: H04N 19/20 using video object coding
• CPC Class Codes: H04N 19/537 Motion estimation other than block-based
• CPC Class Codes: H04N 19/54 using feature points or meshes
• CPC Class Codes: H04N 19/577 Motion compensation with bidirectional frame interpolation, i.e. using B-pictures

Abstract

An 3D imaging method registers relatively deformed first and second volumetric images of an object which may be obtained from different imaging modalities, and/or at different times by performing an elastic geometric transformation on one of the volumetric images. The method includes extracting first and second corresponding surfaces from the respective first and second images, which surfaces delineate the same feature, such as a bone/tissue interface. This is done by extracting a stack of contours for each surface. The first surface is iteratively warped toward alignment with the second surface to arrive at a global translation vector and a set of residual surface distortion vectors. Volume distortion vectors, determined by applying a weighting function to the residual surface distortion vectors, are used to indicate the locations in the second volumetric image of voxel centers whose interpolated intensities are to be moved to lattice points. The weighting function includes a constant indicative of elastic stiffness of the second volume. The extraction of the first surface, surface warping, and volume distortion vector determination steps are repeated in successive iterations until a measure of mis-registration between the volumetric images is less than a predetermined value, with the elastic stiffness constant being decreased each time.

Claims

1. A 3D imaging method comprising:

a) scanning at least a region of an object with an imaging modality to obtain one of first and second relatively geometrically deformed three dimensional volumetric images containing voxels throughout said first and second volumetric images at positions corresponding to a three dimensional grid of lattice points, the other of said first and second images being available;

b) extracting a first surface from the first image by identifying spaced apart surface sample points lying on the first surface, said first surface delineating a feature of said object in said first image;

c) extracting a second surface from the second image by identifying spaced apart surface sample points lying on the second surface, said second surface delineating in said second image the same feature of said object as delineated by said first surface in said first image;

d) automatically, without input from an operator, determining a set of 3D surface point deformation vectors applicable to the respective sample points of one of the first and second surfaces which would register said sample points of said one surface with the other of said first and second surfaces;

e) automatically, without input from an operator, deriving a three dimensional volumetric set of 3D lattice point deformation vectors corresponding to said three dimensional grid of lattice points, each of said 3D lattice point deformation vectors being derived by applying a weighting function to said 3D surface point deformation vectors which are within a vicinity of said derived 3D lattice point deformation vector, said weighting function including a decay constant β

which determines the effective size of said vicinity; and

f) automatically, without input from an operator, applying said three dimensional volumetric set of 3D lattice point deformation vectors to corresponding voxels throughout the entire volume of said three dimensional second image to elastically transform said three dimensional second image into a three dimensional compensated image,

wherein the spaced apart surface sample points lying on the first surface are vertex points of polygonal patches which tile the first surface, the spaced apart surface sample points lying on the second surface are vertex points of polygonal patches which tile the second surface, and said determining a set of 3D surface point deformation vectors is in response to a 3D force field and comprises the steps of;

g) for each vertex point on the first and second surfaces, determining a 3D displacement vector directed to a nearest patch of the other surface, so as to produce first and second sets of displacement vectors applicable to the vertex points of the first and second surfaces, respectively; and

h) determining said 3D force field as a function of said first and second sets of displacement vectors.

2. The imaging method as claimed in claim 1, wherein said set of 3D surface point deformation vectors determined is applicable to the respective sample points of the first surface for warping the first surface toward the second surface.

3. The imaging method as claimed in claim 2, further comprising:

g) forming a measure of the degree of mis-registration between the compensated image and the first image; and

h) repeating steps d) through f), by substituting a last compensated image for said second image, until said measure does not exceed a predetermined maximum, said weighting function being a weighted mean interpolation function and said decay constant β

being modified for each said repeating of steps d) through f) in a direction so as to reduce the effective size of said vicinity in order to elastically deform said last compensated image in ever finer detail.

4. The imaging method as claimed in claim 2, further comprising the step of simultaneously displaying at least image portions from said first and said compensated images.

5. The method of claim 2, further comprising the step of displaying an image whose voxels are determined as a function of the intensities of corresponding voxels of said first and said compensated images.

6. The imaging method as claimed in claim 1, further comprising:

g) forming a measure of the degree of mis-registration between the compensated image and the first image; and

h) repeating steps d) through f), by substituting a last compensated image for said second image, until said measure does not exceed a predetermined maximum, said weighting function being a weighted mean interpolation function and said decay constant β

being modified for each said repeating of steps d) through f) in a direction so as to reduce the effective size of said vicinity in order to elastically deform said last compensated image in ever finer detail.

7. The imaging method as claimed in claim 6, further comprising the step of simultaneously displaying at least image portions from said first and said compensated images.

8. The method of claim 6, further comprising the step of displaying an image whose voxels are determined as a function of the intensities of corresponding voxels of said first and said compensated images.

9. The imaging method as claimed in claim 1, further comprising the step of simultaneously displaying at least image portions from said first and said compensated images.

10. The method of claim 1, further comprising the step of displaying an image whose voxels are determined as a function of the intensities of corresponding voxels of said first and said compensated images.