Method and system for elastic composition of medical imaging volumes

1. A computer implemented method for composing image volumes obtained using a medical scanning device, comprising the steps of:

a. using a medical scanning device to obtain data of at least two image volumes of a patient, said image volumes having an overlap region;

b. processing the data from the two image volumes to determine the overlap region and to register the two image volumes with each other using rigid registration;

c. determining a non-rigid deformation between the two image volumes using an elastic registration algorithm;

d. performing an alpha-morphing on each overlapping region of each image volume using the non-rigid deformation by computing first intermediate image volume result I m1 and second intermediate image volume result I m2 , according to the following equation:

I m1 =I 1 °(αø k )

I m2 =I 2 °((1−α)φ k )

wherein α is a parameter with a range [0.0,1.0] which linearly increases from a top to a bottom of each overlapping region, wherein ø k represents the non rigid deformation mapping I 1 to I 2 , φ k represents the non-rigid deformation mapping I 2 to I 1 , I 1 represents a top overlapping image volume, and I 2 represents a bottom overlapping image volume; and

e. obtaining a blend region of the overlap region by performing an alpha-blending which blends the two intermediate image volume results by varying the relative contribution of the two image volumes using the results of alpha-morphing,

wherein steps b, c, d, and e are performed by a computer processor.

2. The method of claim 1 , wherein the medical scanning device is selected from the group consisting of MRI, CT, Ultrasound, Radiography, and PET imagers.

3. The method of claim 1 , wherein the elastic registration algorithm is based on the maximization of an intensity-based similarity metric between the two image volumes.

4. The method of claim 3 , wherein the metric is selected from the group consisting of Local Cross Correlation and Mutual Information.

5. The method of claim 1 , wherein the elastic registration algorithm models the deformation as a smooth vector field that indicates, for each pixel in one of the image volumes, its corresponding pixel location in a second image in the other image volume.

6. The method of claim 5 , wherein in step c. the deformation and its inverse are estimated iteratively by maximizing an intensity-based similarity metric and are regularized using a low-pass filter.

7. The method according to claim 1 , wherein the step e. includes computing values of I blend according to the equation of:

I blend =(1−α) I m1 +αI m2 ,

wherein α is a parameter with a range [0.0,1.0] which linearly increases from the top to the bottom of the blend region, where I blend represents a composed overlap region, wherein I m1 represents the first intermediate image volume result, and I m2 represents the second intermediate image volume result.

8. The method of claim 1 , wherein the image volumes are partial volumes of a patient's entire anatomy.

9. The method of claim 1 , further including the step of displaying at least a portion of the two image volumes, including the overlap region.

10. The method of claim 9 , wherein the portion is a slice which extends over the two image volumes, including the overlap region.

11. An article of manufacture comprising a non-transitory computer-usable medium having computer-readable program code embodied therein for composing a composite image from at least two smaller images, the computer-readable program code having instructions for performing the following steps:

a. obtaining data of at least two image volumes of a patient, said image volumes having an overlap region;

b. processing the data from the two image volumes to determine the overlap region and to register the two image volumes with each other using rigid registration;

c. determining a non-rigid deformation between the two image volumes using an elastic registration algorithm;

d. performing an alpha-morphing on each overlapping region of each image volume using the non-rigid deformation by computing first intermediate image volume result I m1 and second intermediate image volume result I m2 , according to the following equation:

I m1 =I 1 °(αø k )

I m2 =I 2 °((1−α)φ k )

wherein α is a parameter with a range [0.0,1.0] which linearly increases from a top to a bottom of each overlapping region, wherein ø k represents the non rigid deformation mapping I 1 to 2 φ k represents the non-rigid deformation mapping I 2 to I 1 ,I 1 represents a top overlapping image volume, and I 2 represents a bottom overlapping image volume; and

e. obtaining a blend region of the overlap region by performing an alpha-blending which blends the two intermediate image volume results by varying the relative contribution of the two image volumes using the results of alpha-morphing.

12. The article of claim 11 , wherein the medical scanning device is selected from the group consisting of MRI, CT, Ultrasound, Radiography, and PET imagers.

13. The article of claim 11 , wherein the elastic registration algorithm is based on the maximization of an intensity-based similarity metric between the two image volumes.

14. The article of claim 13 , wherein the metric is selected from the group consisting of Local Cross Correlation and Mutual Information.

15. The article of claim 11 , wherein the step c. includes determining a deformation field by obtaining a smooth vector field that indicates for each pixel in one of the image volumes, its corresponding pixel location in a second image in the other image volume.

16. The article of claim 15 , wherein in step c., the deformation and its inverse are estimated iteratively by maximizing an intensity-based similarity matrix and are regularized using a low-pass filter.

17. The article according to claim 11 , wherein the step e. includes computing values of I blend according to the equation of:

I blend =(1−α) I m1 +αI m2 ,

wherein α is a parameter with a range [0.0,1.0] which linearly increases from the top to the bottom of the blend region, where I blend represents the composed overlap region, wherein I m1 represents the first intermediate image volume result, and I m2 represents the second intermediate image volume result.

18. The article of claim 11 , wherein the image volumes are partial volumes of a patient's entire anatomy.

19. The article of claim 11 , further including the step of displaying at least a portion of the two image volumes, including the overlap region.

20. The article of claim 19 , wherein the portion is a slice which extends over the two image volumes, including the overlap region.

21. A system for composing an image from two smaller image volumes of data obtained using a medical scanning device, comprising:

a scanner for obtaining image data for at least two image volumes of a patient, wherein the image volumes have an overlap region;

a processor for:

a. processing the data from the two image volumes to determine the overlap region and to register the two image volumes with each other using rigid registration;

b. determining a non-rigid deformation between the two image volumes using an elastic registration algorithm;

c. performing an alpha-morphing on each overlapping region of each image volume using the non-rigid deformation by computing first intermediate image volume result I m1 and second intermediate image volume result I m2 , according to the following equation:

I m1 =I 1 °(αø k )

I m2 =I 2 °((1−α)φ k )

wherein α is a parameter with a range [0.0,1.0] which linearly increases from a top to a bottom of each overlapping region, wherein ø k represents the non rigid deformation mapping I 1 to I 2 , φ k represents the non-rigid deformation mapping I 2 to I 1 ,I 1 represents a top overlapping image volume, and I 2 represents a bottom overlapping image volume; and

d. obtaining a blend region of the overlap region by performing an alpha-blending which blends the two intermediate image volume results by varying the relative contribution of the two image volumes using the results of alpha-morphing.

22. The system of claim 21 , wherein the medical scanning device is selected from the group consisting of MRI, CT, Ultrasound, Radiography, and PET imagers.

23. The system of claim 21 , wherein the elastic registration algorithm is based on the maximization of an intensity-based similarity metric between the two image volumes.

24. The system of claim 21 , wherein the metric is selected from the group consisting of Local Cross Correlation and Mutual Information.

25. The system of claim 21 , wherein the processor determines a deformation field by obtaining a smooth vector field that indicates for each pixel in one of the image volumes, its corresponding pixel location in a second image in the other image volume.

26. The system of claim 25 , wherein the deformation and its inverse are estimated iteratively by maximizing an intensity-based similarity matrix and are regularized using a low-pass filter.

27. The system according to claim 21 , wherein the processor obtains a blend region by computing values of /blend according to the equation of:

I blend =(1−α) I m1 +αI m2 ,

wherein α is a parameter with a range [0.0,1.0] which linearly increases from the top to the bottom of the blend region, where I blend represents a composed overlap region,

wherein I m1 represents the first intermediate image volume result, and I m2 represents the second intermediate image volume result.

28. The system of claim 21 , wherein the image volumes are partial volumes of a patient's entire anatomy.

29. The system of claim 21 , further including a display for displaying at least a portion of the two image volumes, including the overlap region.

30. The system of claim 29 , wherein the portion is a slice which extends over the two image volumes, including the overlap region.