Apparatus and method for determining measure of similarity between images

• US 7,480,399 B2
• Filed: 01/16/2007
• Issued: 01/20/2009
• Est. Priority Date: 08/29/2003
• Status: Active Grant

• Current Assignee: Accuray Incorporated
• Original Assignee: Accuray Incorporated
• Inventors: Kuduvalli, Gopinath, Fu, Dongshan
• Primary Examiner(s): Tabatabai; Abolfazl
• Application Number:
  US11/654,254
• Publication Number:
  US 20070116341A1
• Time in Patent Office: 735 Days
• Field of Search: 382/128, 382/129, 382/130, 382/131, 382/132, 382/133, 382/134, 378/4, 378/46, 378/63, 378/90, 378/92, 378/98.4, 378/98.6, 378/98.9, 378/101, 378/140, 600/407, 600/425, 600/426, 600/427
• US Class Current: 382/128
• CPC Class Codes: A61N 2005/1061 using an x-ray imaging system having a separate imaging source
• CPC Class Codes: A61N 2005/1062 using virtual X-ray images, e.g. digitally reconstructed radiographs [DRR]
• CPC Class Codes: A61N 5/1049 for verifying the position of the patient with respect to the radiation beam
• CPC Class Codes: A61N 5/1067 in real time, i.e. during treatment
• CPC Class Codes: A61N 5/107 in real time, i.e. during treatment
• CPC Class Codes: G01N 23/04 and forming images of the material
• CPC Class Codes: G06T 2207/10124 Digitally reconstructed radiograph [DRR]
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06T 7/32 using correlation-based methods
• CPC Class Codes: G06T 7/38 Registration of image sequences

Abstract

A method and system is presented in image-guided radiosurgery for determining the measure of similarity of two digital images, for example a 2D x-ray image and a 2D DRR synthesized from 3D scan data. A two-dimensional array of pixel values of a difference image may be formed by subtracting each pixel value of the second image from the corresponding pixel value of the first image. The pattern intensity function may be constructed by taking the summation of functions of the gradients of the difference image. The neighborhood R may be defined so as to allow the gradients of the difference image to be considered in at least one direction.

Claims

1. A method in image-guided radiation treatment, comprising:

determining a measure of similarity of a first image of an object and a second image of the object, wherein each of the first image and the second image has pixel values, and wherein determining the measure of similarity comprises;

forming a difference image by subtracting corresponding pixel values of the second image from each pixel value of the first image; and

forming a pattern intensity function over all the pixels in the difference image within a neighborhood.

2. The method of claim 1, wherein the first image is an x-ray image of the object and the second image is a digitally reconstructed radiograph (DRR) synthesized from a previously generated three dimensional (3D) scan data of the object.

3. The method of claim 1, wherein forming the pattern intensity function comprises summing functions of gradients of the difference image.

4. The method of claim 3, wherein the neighborhood is defined so that the gradients of the difference image can be considered in at least one direction.

5. The method of claim 3, wherein the neighborhood is defined so that the gradients of the difference image can be considered in a plurality of directions.

6. The method of claim 1, wherein said pattern intensity function is characterized by a mathematical formulation given by

$\sum _{}^{}$

⁢

i , j ⁢

⁢

∑

k , l ⋐

R ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( i + k , j + l ) ) 2 ,

wherein I_dif(i,j) represents an array of pixel values for the difference image, and wherein σ

is a weighting constant for weighting the pattern intensity function, and where the neighborhood defined around a pixel (i,j) as a center point.

7. The method of claim 6, wherein σ

is from about 1 to about 64.

8. A computer readable medium storing software that provides instructions which, if executed, performs operations comprising:

determining a measure of similarity of a first image of an object and a second image of the object, wherein each of the first image and the second image has pixel values, and wherein determining the measure of similarity comprises;

forming a difference image by subtracting corresponding pixel values of the second image from each pixel value of the first image; and

forming a pattern intensity function over all the pixels in the difference image within a neighborhood.

9. The computer readable medium of claim 8, wherein the first image is an x-ray image of the object and the second image is a digitally reconstructed radiograph (DRR) synthesized from a previously generated three dimensional (3D) scan data of the object.

10. The computer readable medium of claim 8, wherein forming the pattern intensity function comprises summing functions of gradients of the difference image.

11. The computer readable medium of claim 10, wherein the neighborhood is defined so that the gradients of the difference image can be considered in at least one direction.

12. The computer readable medium of claim 10, wherein the neighborhood is defined so that the gradients of the difference image can be considered in a plurality of directions.

13. The computer readable medium of claim 8, wherein said pattern intensity function is characterized by a mathematical formulation given by

$\sum _{}^{}$

⁢

i , j ⁢

⁢

∑

k , l ⋐

R ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( i + k , j + l ) ) 2 ,

wherein I_dif(i,j) represents an array of pixel values for the difference image, and wherein σ

is a weighting constant for weighting the pattern intensity function, and where the neighborhood defined around a pixel (i,j) as a center point.

14. The computer readable medium of claim 13, wherein σ

is from about 1 to about 64.

15. The computer readable medium of claim 8, wherein the first image is an x-ray image of the object and the second image is a digitally reconstructed radiograph (DRR) synthesized from a previously generated three dimensional (3D) scan data of the object, and wherein the pattern intensity function is formed by summing asymptotic functions of the gradients of the difference image over all the pixels within a neighborhood R, and wherein the neighborhood R is defined so that the gradients of the difference image can be considered in at least four directions.

16. A computer readable medium of claim 15, wherein the first and second images are discretized images respectively characterized by a first and a second two-dimensional (2D) arrays of pixel values;

and wherein said difference image is a discretized image characterized by a third 2D array of pixel values, and wherein each pixel value of an image is a number representative of an intensity of the image at a corresponding 2D array element.

17. A computer readable medium of claim 16, wherein the pixel value for each image represents the intensity of said image, and wherein the pixel value at an i-th row and j-th column of the third away of pixel values for the difference image is given by

I_dif(i,j)=I_Live(i,j)−

I_DRR(i,j),

wherein I_Live(i,j) represents the (i,j)-th pixel value of a real-time x-ray image of the object, and I_DRR(i,j) represents the (i,j)th pixel value of the digitally reconstructed image of the object synthesized from previously generated 3D scan data of the object.

18. The computer readable medium of claim 15, wherein the pattern intensity function is characterized by a mathematical formulation given by

$\sum _{}^{}$

⁢

i , j ⁢

⁢

∑

k , l ⋐

R ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( i + k , j + l ) ) 2 ,

where I_dif(i,j) represents the array of pixel values for the difference image,

where σ

is a weighting constant for weighting the pattern intensity function, and where R is a neighborhood defined around a pixel (i,j) as a center point, and wherein a is from about 4 to about 16.

19. The computer readable medium of claim 15, wherein the at least four directions comprise:

a substantially horizontal direction;

a substantially vertical direction;

a diagonal direction of about 45 degrees; and

a diagonal direction of about −

45 degrees.

20. The computer readable medium of claim 19, wherein the pattern intensity function is given by

$\sum _{}^{}$

i , j ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( ⅈ

- 1 , j - 1 ) ) 2 + ∑

i , j ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( ⅈ

- 1 , j ) ) 2 + ∑

i , j ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( ⅈ

- 1 , j - 1 ) ) 2 + ∑

i , j ⁢

σ

2 σ

2 + ( I dif ⁡

( ⅈ

, j ) - I dif ⁡

( ⅈ

- 1 , j + 1 ) ) 2 .