Deformable image registration based on masked computed tomography (CT) image

1. A method to improve computed tomography (CT) to cone beam computed tomography (CBCT) registration, comprising:

receiving, by a deformable image registration (DIR) system, a two-dimensional (2D) CT image generated by CT-scanning of an object;

receiving, by the DIR system, a 2D CBCT image generated by CBCT-scanning of the object;

generating, by the DIR system, an image mask based on Digital Imaging and Communications in Medicine (DICOM) information extracted from the 2D CBCT image, wherein for a specific pixel in the 2D CBCT image, the image mask contains a corresponding data-field indicating whether the specific pixel contains actual image data that are generated based on the CBCT-scanning of the object; and

generating, by the DIR system, a registered image by utilizing the image mask to perform a DIR between the 2D CT image and the 2D CBCT image.

2. The method as recited in claim 1 , further comprising:

performing an automatic dose estimation or a contour propagation on the object based on the registered image.

3. The method as recited in claim 1 , wherein the DICOM information further comprises DICOM Target Center and DICOM Reconstruction Diameter.

4. The method as recited in claim 3 , wherein the generating of the image mask further comprises:

constructing a reconstruction area for the 2D CBCT image based on the DICOM Target Center and DICOM Reconstruction Diameter extracted from the 2D CBCT image; and

upon a determination that the specific pixel is outside the reconstruction area for the 2D CBCT image, assigning a no-data value to the corresponding data-field in the image mask that is associated with the specific pixel.

5. The method as recited in claim 3 , wherein the generating of the image mask further comprises:

upon a determination that the specific pixel is inside the reconstruction area for the 2D CBCT image, assigning a data-existence value to the corresponding data-field in the image mask that is associated with the specific pixel.

6. The method as recited in claim 1 , wherein the generating of the registered image further comprises:

selecting a first pixel from the 2D CBCT image and a second pixel from the 2D CT image; extracting from the image mask a first data-field associated with the first pixel; and

upon a determination that the first data-field contains a data-existence value, using the first pixel during the performing of the DIR.

7. The method as recited in claim 6 , wherein the generating of the registered image further comprises:

selecting a third pixel from the 2D CBCT image and a fourth pixel from the 2D CT image;

extracting from the image mask a second data-field associated with the third pixel; and

upon a determination that the second data-field contains a no-data value, ignoring the third pixel during the performing of the DIR.

8. A method to improve computed tomography (CT) to cone beam computed tomography (CBCT) registration, comprising:

receiving, by a deformable image registration (DIR) system, a plurality of two-dimensional (2D) CT images generated by CT-scanning of an object;

receiving, by the DIR system, a plurality of 2D CBCT images generated by CBCT-scanning of the object;

generating, by the DIR system, an image mask based on Digital Imaging and Communications in Medicine (DICOM) information extracted from a specific 2D CBCT image selected from the plurality of 2D CBCT images, wherein for a specific pixel in the specific 2D CBCT image, the image mask contains a corresponding data-field indicating whether the specific pixel contains actual image data that are generated based on the CBCT-scanning of the object; and

generating, by the DIR system, a plurality of registered images by performing a DIR between the plurality of 2D CT images and the plurality of 2D CBCT images utilizing the image mask.

9. The method as recited in claim 8 , wherein the DICOM information extracted from each of the plurality of 2D CBCT images further comprises having identical DICOM Target Center and DICOM Reconstruction Diameter.

10. The method as recited in claim 9 , wherein the generating of the image mask further comprises:

constructing a reconstruction area for the specific 2D CBCT image based on the DICOM Target Center and DICOM Reconstruction Diameter; and

upon a determination that the specific pixel is outside the reconstruction area for the specific 2D CBCT image, assigning a no-data value to the corresponding data-field in the image mask that is associated with the specific pixel.

11. The method as recited in claim 10 , wherein the generating of the image mask further comprises:

upon a determination that the specific pixel is inside the reconstruction area of the specific 2D CBCT image, assigning a data-existence value to the corresponding data-field in the image mask that is associated with the specific pixel.

12. The method as recited in claim 8 , wherein the generating of the plurality of registered images further comprises:

for a 2D CBCT image selected from the plurality of 2D CBCT images and a 2D CT image selected from the plurality of 2D CT images, selecting a first pixel from the 2D CBCT image and a second pixel from the 2D CT image;

extracting from the image mask a first data-field associated with the first pixel; and

upon a determination that the first data-field contains a data-existence value, using the first pixel during the performing of the DIR.

13. The method as recited in claim 12 , wherein the generating of the plurality of registered images further comprises:

selecting a third pixel from the 2D CBCT image and a fourth pixel from the 2D CT image;

extracting from the image mask a second data-field associated with the third pixel; and

upon a determination that the second data-field contains a no-data value, ignoring the third pixel during the performing of the DIR.

14. A registration-enhancement system configured to improve computed tomography (CT) to cone beam computed tomography (CBCT) registration, comprising:

an image-receiving module configured to

receive a two-dimensional (2D) CT image generated by CT-scanning of an object, and

receive a 2D CBCT image generated by CBCT-scanning of the object;

a mask-generating module configured to

generate an image mask based on Digital Imaging and Communications in Medicine (DICOM) information extracted from the 2D CBCT image, wherein for a specific pixel in the 2D CBCT image, the image mask contains a corresponding data-field indicating whether the specific pixel contains actual image data that are generated based on the CBCT-scanning of the object; and

a deformable image registration (DIR) module configured to

generate a registered image by utilizing the image mask to perform a DIR between the 2D CT image and the 2D CBCT image.

15. The registration-enhancement system as recited in claim 14 , wherein the DICOM information further comprises DICOM Target Center and DICOM Reconstruction Diameter.

16. The registration-enhancement system as recited in claim 14 , wherein the mask-generating module is further configured to:

construct a reconstruction area for the 2D CBCT image based on the DICOM Target Center and DICOM Reconstruction Diameter extracted from the 2D CBCT image; and

upon a determination that the specific pixel is outside the reconstruction area for the 2D CBCT image, assigning a no-data value to the corresponding data-field in the image mask that is associated with the specific pixel.

17. The registration-enhancement system as recited in claim 16 , wherein the mask-generating module is further configured to:

determine if the specific pixel is inside the reconstruction area for the 2D CBCT image; and

assign a data-existence value to the corresponding data-field in the image mask that is associated with the specific pixel.

18. The registration-enhancement system as recited in claim 14 , wherein the DIR module is further configured to:

select a first pixel from the 2D CBCT image and a second pixel from the 2D CT image during the performing of the DIR;

extract from the image mask a first data-field associated with the first pixel;

determine if the first data-field contains a data-existence value; and

use the first pixel during the performing of the DIR.

19. The registration-enhancement system as recited in claim 14 , wherein the DIR module is further configured to:

select a third pixel from the 2D CBCT image and a fourth pixel from the 2D CT image;

extract from the image mask a second data-field associated with the third pixel;

determine if the second data-field contains a no-data value; and

ignore the third pixel during the performing of the DIR.

20. The method as recited in claim 1 , wherein the specific pixel that contains actual image data is disposed within an illuminated region of the 2D CBCT image.