SYSTEM AND METHOD FOR LEARNING-BASED 2D/3D RIGID REGISTRATION FOR IMAGE-GUIDED SURGERY

• US 20070248251A1
• Filed: 03/15/2007
• Published: 10/25/2007
• Est. Priority Date: 04/24/2006
• Status: Active Grant

• Current Assignee: Siemens Healthcare GMBH (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG)
• Inventors: Sun, Yiyong, Xu, Chenyang, Liao, Rui, Sauer, Frank, Khamene, Ali, Guetter, Christoph
• Granted Patent: US 7,940,999 B2
• Time in Patent Office: Days
• US Class Current: 382/128
• 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 7/35 using statistical methods
• CPC Class Codes: G06V 10/758 Involving statistics of pixels or of feature values, e.g. histogram matching
• CPC Class Codes: G06V 20/647 by matching two-dimensional images to three-dimensional objects

Abstract

A method of registering 3-dimensional digitized images to 2-dimensional digitized images during a medical procedure includes providing a pair of correctly-registered training images L={l_r, l_f} and their joint intensity distribution p_l(i_r, i_f), wherein i_r and i_f are reference and floating images, respectively, providing a pair of observed images O={o_r, o_f} and their joint intensity distribution p_o(i_r, i_f), mapping a marginal intensity distribution of the observed pair O={o_r, o_f} to a marginal intensity distribution of the training pair L={l_r, l_f}, and estimating a set of parameters T that registers image o_f to image o_r by maximizing a weighted sum of a Jensen-Shannon divergence (JSD) of a joint intensity distribution of the observed pair and a joint intensity distribution of the training pair and a similarity measure between the observed images.

Claims

1. A method of registering 3-dimensional digitized images to 2-dimensional digitized images during a medical procedure comprising the steps of:

providing a pair of correctly-registered training images L={l_r, l_f} and their joint intensity distribution p_l(i_r, i_f), wherein i_r and i_f are reference and floating images, respectively;

providing a pair of observed images O={o_r, o_f} and their joint intensity distribution p_o(i_r, i_f);

mapping a marginal intensity distribution of the observed pair O={o_r, o_f} to a marginal intensity distribution of the training pair L={l_r, l_f}; and

estimating a set of parameters T that registers image o_f to image o_r by maximizing a weighted sum of a Jensen-Shannon divergence (JSD) of a joint intensity distribution of the observed pair and a joint intensity distribution of the training pair and a similarity measure between the observed images.

2. The method of claim 1, wherein mapping the marginal intensity distribution of the observed pair to the marginal intensity distribution of the training pair comprises mapping an intensity i for observed image o_r to intensity k for training image l_r where k minimizes an absolute difference between the marginal histograms p_or(i) and p_lr(k) of o_r and l_r respectively, where t₁<

k<

t₂ with t₁ and t₂ defined by C_1r(t₁−

1)<

C_or(i−

1)≦

C_lr(t₁) and C_lr(t₂−

1)<

C_or(i)≦

C_lr(t₂) wherein C_or( ) and C_ir( ) denote cumulative density functions of o_r and l_r respectively.

3. The method of claim 1, wherein one of said pair of observed images is a 2D digitally reconstructed radiograph (DRR) generated from a 3-dimensional (3D) image of a medical target, and the other of said pair is a live 2-dimensional (2D) X-ray image of said medical target.

4. The method of claim 3, wherein said registration parameters T comprise six parameters that characterize a pose of said 3D image with respect to said 2D X-ray image.

5. The method of claim 4, wherein maximizing said weighted sum comprises maximizing an expression equivalent to

α

(1−

JSD(p_l∥

p_o^T))+(1−

α

)[MI(o_r,o_f^T) or PI(o_r,o_f^T)],

wherein JSD(p_l∥

p_o^T) is a Jensen-Shannon divergence of the joint intensity distribution p_l of the training images and of the joint intensity distribution p_o^T of the observed images wherein one of said observed images has been registered to the other via the parameters T, MI (o_r,o_f^T) is a Mutual Information measure of the registered observed images, PI(o_r,o_f^T) is a Pattern Intensity measure of the registered observed images, and α

is a weighting factor with values between 0 and 1.

6. The method of claim 5, wherein said expression is maximized using a Hill-climbing optimization method.

7. The method of claim 5, wherein the Jensen-Shannon divergence is defined as

8. The method of claim 5, wherein the Mutual Information measure of the registered observed images is defined as

9. The method of claim 5, wherein the Pattern Intensity measure of the registered observed images is defined as

10. The method of claim 1, further comprising repeating the steps of providing a pair of observed images, mapping a marginal intensity distribution of the observed pair to a marginal intensity distribution of the training pair, and estimating a set of parameters T that registers the observed pair of images until a difference in parameter values T between iterations is sufficiently small.

11. A method of registering 3-dimensional digitized images to 2-dimensional digitized images during a medical procedure comprising the steps of:

providing a pair of correctly-registered training images L={l_r, l_f} and their joint intensity distribution p_l(i_r, i_f), wherein i_r and i_f are reference and floating images, respectively;

providing a 3-dimensional (3D) image volume of a medical target;

generating a 2D digitally reconstructed radiograph (DRR) from said 3D image volume of said medical target;

generating a live 2-dimensional (2D) X-ray image of said medical target, wherein said DRR and said Z-ray image comprise pair of observed images O={o_r, o_f} to be registered;

calculating a joint intensity distribution p_o(i_r, i_f) of the observed images; and

estimating a set of parameters T that registers image o_f to image o_r by maximizing a weighted sum of a Jensen-Shannon divergence (JSD) of a joint intensity distribution of the observed pair and a joint intensity distribution of the training pair and a similarity measure between the observed images, wherein said registration parameters T comprise six parameters that characterize a pose of said 3D image with respect to said 2D X-ray image.

12. The method of claim 11, further comprising mapping an intensity i for observed image o_r to intensity k for a training image l_r where k minimizes an absolute difference between the marginal histograms p_or(i) and p_lr(k) of o_r and l_r respectively, where t₁<

k<

t₂ with t₁ and t₂ defined by

C_1r(t₁−

1)<

C_or(i−

1)≦

C_lr(t₁) and C_lr(t₂−

1)<

C_or(i)≦

C_lr(t₂)

wherein C_or( ) and C_ir( ) denote cumulative density functions of o_r and l_r respectively.

13. A program storage device readable by a computer, tangibly embodying a program of instructions executable by the computer to perform the method steps for registering 3-dimensional digitized images to 2-dimensional digitized images during a medical procedure comprising the steps of:

providing a pair of correctly-registered training images L={l_r, l_f} and their joint intensity distribution p_l(i_r, i_f), wherein i_r and i_f are reference and floating images, respectively;

providing a pair of observed images O={o_r, o_f} and their joint intensity distribution p_o(i_r, i_f);

mapping a marginal intensity distribution of the observed pair O={o_r, o_f} to a marginal intensity distribution of the training pair L={l_r, l_f}; and

estimating a set of parameters T that registers image o_f to image o_r by maximizing a weighted sum of a Jensen-Shannon divergence (JSD) of a joint intensity distribution of the observed pair and a joint intensity distribution of the training pair and a similarity measure between the observed images.

14. The computer readable program storage device of claim 13, wherein mapping the marginal intensity distribution of the observed pair to the marginal intensity distribution of the training pair comprises mapping an intensity i for observed image o_r to intensity k for training image l_r where k minimizes an absolute difference between the marginal histograms p_or(i) and p_lr(k) of o_r and l_r respectively, where t₁<

k<

t₂ with t₁ and t₂ defined by C_1r(t₁−

1)<

C_or(i−

1)≦

C_lr(t₁) and C_lr(t₂−

1)<

C_or(i)≦

C_lr(t₂) wherein C_or( ) and C_ir( ) denote cumulative density functions of o_r and l_r respectively.

15. The computer readable program storage device of claim 13, wherein one of said pair of observed images is a 2D digitally reconstructed radiograph (DRR) generated from a 3-dimensional (3D) image of a medical target, and the other of said pair is a live 2-dimensional (2D) X-ray image of said medical target.

16. The computer readable program storage device of claim 15, wherein said registration parameters T comprise six parameters that characterize a pose of said 3D image with respect to said 2D X-ray image.

17. The computer readable program storage device of claim 16, wherein maximizing said weighted sum comprises maximizing an expression equivalent to

α

(1−

JSD(p_l∥

p_o^T))+(1−

α

)[MI(o_r,o_f^T) or PI(o_r,o_f^T)],

wherein JSD(p_l∥

p_o^T) is a Jensen-Shannon divergence of the joint intensity distribution p_l of the training images and of the joint intensity distribution p_o^T of the observed images wherein one of said observed images has been registered to the other via the parameters T, MI(o_r,o_f^T) is a Mutual Information measure of the registered observed images, PI(o_r,o_f^T) is a Pattern Intensity measure of the registered observed images, and α

is a weighting factor with values between 0 and 1.

18. The computer readable program storage device of claim 17, wherein said expression is maximized using a Hill-climbing optimization method.

19. The computer readable program storage device of claim 17, wherein the Jensen-Shannon divergence is defined as

20. The computer readable program storage device of claim 17, wherein the Mutual Information measure of the registered observed images is defined as

21. The computer readable program storage device of claim 17, wherein the Pattern Intensity measure of the registered observed images is defined as

22. The computer readable program storage device of claim 13, the method further comprising repeating the steps of providing a pair of observed images, mapping a marginal intensity distribution of the observed pair to a marginal intensity distribution of the training pair, and estimating a set of parameters T that registers the observed pair of images until a difference in parameter values T between iterations is sufficiently small.