Motion-blurred imaging enhancement method and system

1. A method to enhance a motion-blurred image, comprising:

receiving a plurality of projections, wherein the plurality of projections contain computed tomography (CT) data obtained in multiple motion phases;

generating a multi-phase image based on the plurality of projections;

generating an enhanced multi-phase image by performing a destreaking operation on the multi-phase image to reduce one or more motion-induced artifacts in the multi-phase image;

generating a plurality of re-projections by forward-projecting the enhanced multi-phase image;

generating a plurality of difference projections by comparing the plurality of projections and the plurality of re-projections;

generating a plurality of phase-correlated difference images by performing a reconstruction operation based on the plurality of difference projections; and

generating a plurality of phase-correlated images based on the plurality of phase-correlated difference images and the enhanced multi-phase image.

2. The method as recited in claim 1 , wherein the generating of the enhanced multi-phase image comprises:

generating a plurality of motion projections based on identifying moving pixels that contribute to the one or more motion-induced artifacts in the multi-phase image, being a first multi-phase image;

generating a second multi-phase image based on the plurality of motion projections; and

generating the enhanced multi-phase image by combining the second multi-phase image with the first multi-phase image.

3. The method as recited in claim 2 , wherein the generating of the plurality of motion projections comprises:

reconstructing a McKinnon-Bates (MKB) phase-correlated image based on the plurality of projections;

generating a motion map based on the MKB phase-correlated image; and

generating one of the plurality of motion projections by performing a forward-projection operation based on the motion map and the MKB phase-correlated image.

4. The method as recited in claim 1 , wherein the generating of the enhanced multi-phase image comprises:

reducing streak intensities in the multi-phase image with a filter.

5. The method as recited in claim 4 , wherein the filter is a median filter, a bilateral filter, a neighborhood filter, or a non-local means filter with edge-preserving properties.

6. The method as recited in claim 1 , wherein the generating of the enhanced multi-phase image comprises:

segmenting the multi-phase image into a plurality of segments representing anatomical structures;

smoothing boundaries among the plurality of segments; and

generating the enhanced multi-phase image based on the plurality of segments with smoothed boundaries.

7. The method as recited in claim 6 , wherein the segmenting of the multi-phase image comprises:

identifying a plurality of pixels from the multi-phase image based on an intensity threshold range for a body part; and

reassigning the plurality of pixels with a mean pixel value for the body part, wherein the reassigned plurality of pixels are deemed a segment of the plurality of segments.

8. The method as recited in claim 6 , wherein the smoothing of the boundaries comprises applying a filter to the plurality of segments.

9. The method as recited in claim 6 , wherein the smoothing of the boundaries is not performed on stationary or near stationary segments.

10. The method as recited in claim 6 , wherein the smoothing of the boundaries comprises:

replacing pixel values for a segment selected from the plurality of segments with a mean value corresponding to a body part associated with the segment;

eroding boundary regions among the plurality of segments; and

replacing the eroded boundary regions of the plurality of segments with their corresponding original pixel values.

11. A method to enhance a motion-blurred image, comprising:

generating a first multi-phase image based on a plurality of projections, wherein the plurality of projections contain computed tomography (CT) data obtained in multiple motion phases;

generating an enhanced multi-phase image by reducing streaks in the first multi-phase image;

generating a plurality of re-projections by forward-projecting the enhanced multi-phase image;

generating a plurality of difference projections by comparing each projection selected from the plurality of projections with a corresponding re-projection selected from the plurality of re-projections;

generating a plurality of phase-correlated difference images based on the plurality of difference projections; and

generating a plurality of phase-correlated images based on the plurality of phase-correlated difference images and the enhanced multi-phase image.

12. The method as recited in claim 11 , wherein the reducing of the streaks in the first multi-phase image comprising:

extracting the streaks from the first multi-phase image; and

removing the streaks from the first multi-phase image.

13. The method as recited in claim 11 , wherein the reducing of the streaks in the first multi-phase image comprises applying a filter to the first multi-phase image.

14. The method as recited in claim 11 , wherein the reducing of the streaks in the first multi-phase image comprising:

segmenting the first multi-phase image into a plurality of segments; and

adjusting boundary regions among the plurality of segments.

15. A non-transitory machine-readable medium containing a set of instructions to enhance a motion-blurred image which, in response to execution by a processor, causes the processor to:

receive a plurality of projections, wherein the plurality of projections contain computed tomography (CT) data obtained in multiple motion phases;

generate a multi-phase image based on the plurality of projections;

generate an enhanced multi-phase image by performing a destreaking operation on the multi-phase image to reduce one or more motion-induced artifacts in the multi-phase image;

generate a plurality of re-projections by forward-projecting the enhanced multi-phase image;

generate a plurality of difference projections by comparing the plurality of projections and the plurality of re-projections;

generate a plurality of phase-correlated difference images by performing a reconstruction operation based on the plurality of difference projections; and

generate a plurality of phase-correlated images based on the plurality of phase-correlated difference images and the enhanced multi-phase image.

16. The non-transitory machine-readable medium as recited phase image, which in response to execution by the processor, causes the processor to:

generate a plurality of motion projections based on identifying moving pixels that contribute to the one or more motion-induced artifacts in the multi-phase image, being a first multi-phase image;

generate a second multi-phase image based on the plurality of motion projections; and

generate the enhanced multi-phase image by combining the second multi-phase image with the first multi-phase image.

17. The non-transitory machine-readable medium as recited in claim 16 , containing additional instructions for generating the plurality of motion projections, which in response to execution by the processor, causes the processor to:

reconstruct a McKinnon-Bates (MKB) phase-correlated image based on the plurality of projections;

generate a motion map based on the MKB phase-correlated image; and

generate one of the plurality of motion projections by performing a forward-projection operation based on the motion map and the MKB phase-correlated image.

18. The non-transitory machine-readable medium as recited in claim 15 , containing additional instructions for generating the enhanced multi-phase image, which in response to execution by the processor, causes the processor to:

reduce streak intensities in the multi-phase image with a filter.

19. The non-transitory machine-readable medium as recited in claim 15 , containing additional instructions for generating the enhanced multi-phase image, which in response to execution by the processor, causes the processor to:

segment the multi-phase image into a plurality of segments representing anatomical structures;

smooth boundaries among the plurality of segments; and

generate the enhanced multi-phase image based on the plurality of segments with smoothed boundaries.

20. The non-transitory machine-readable medium as recited in claim 19 , containing additional instructions for segmenting the multi-phase image, which in response to execution by the processor, causes the processor to:

identify a plurality of pixels from the multi-phase image based on an intensity threshold range for a body part; and

reassign the plurality of pixels with a mean pixel value for the body part, wherein the reassigned plurality of pixels are deemed a segment of the plurality of segments.

21. The non-transitory machine-readable medium as recited in claim 19 , containing additional instructions for smoothing the boundaries, which in response to execution by the processor, causes the processor to:

replace pixel values for a segment selected from the plurality of segments with a mean value corresponding to a body part associated with the segment;

erode boundary regions among the plurality of segments; and

replace the eroded boundary regions of the plurality of segments with their corresponding original pixel values.

22. A non-transitory machine-readable medium containing a set of instructions to enhance a motion-blurred image which, in response to execution by a processor, causes the processor to:

generate a first multi-phase image based on a plurality of projections, wherein the plurality of projections contain computed tomography (CT) data obtained in multiple motion phases;

generate an enhanced multi-phase image by reducing streaks in the first multi-phase image;

generate a plurality of re-projections by forward-projecting the enhanced multi-phase image;

generate a plurality of difference projections by comparing each projection selected from the plurality of projections with a corresponding re-projection selected from the plurality of re-projections;

generate a plurality of phase-correlated difference images based on the plurality of difference projections; and

generate a plurality of phase-correlated images based on the plurality of phase-correlated difference images and the enhanced multi-phase image.