METHODS AND SYSTEMS FOR METAL ARTIFACT REDUCTION IN SPECTRAL CT IMAGING

• US 20160324499A1
• Filed: 05/07/2015
• Published: 11/10/2016
• Est. Priority Date: 05/07/2015
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Sen Sharma, Kriti, Pal, Debashish, Gao, Hewei
• Granted Patent: US 9,498,179 B1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/461 Displaying means of special interest
• CPC Class Codes: A61B 6/482 involving multiple energy imaging
• CPC Class Codes: A61B 6/5258 involving detection or reduction of artifacts or noise
• CPC Class Codes: G06T 11/005 Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
• CPC Class Codes: G06T 11/008 Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2211/408 Dual energy

Abstract

Various methods and systems for spectral computed tomography imaging are provided. In one embodiment, a method comprises acquiring a first projection dataset and a second projection dataset, detecting a location of metal in the first projection dataset, applying corrections to the first and second projection datasets based on the location of the metal, and displaying an image reconstructed from the corrected first and second projection datasets. In this way, metal artifacts may be substantially reduced in dual or multi-energy CT imaging.

Claims

1. A method, comprising:

acquiring a first projection dataset and a second projection dataset;

detecting a location of metal in the first projection dataset based on a metal threshold;

applying corrections to the first and second projection datasets based on the location of the metal; and

displaying an image reconstructed from the corrected first and second projection datasets.

2. The method of claim 1, wherein detecting the location of the metal in the first projection dataset comprises generating a first metal mask comprising voxels above the metal threshold in a first backprojection of the first projection dataset.

3. The method of claim 2, further comprising generating a second metal mask comprising voxels in a second backprojection of the second projection dataset at a same location of the voxels in the first metal mask.

4. The method of claim 1, further comprising generating a first prior image based on the first projection dataset and the location of the metal, and generating a second prior image based on the second projection dataset and the first prior image.

5. The method of claim 4, further comprising segmenting the first prior image into at least two regions based on selected threshold values.

6. The method of claim 5, wherein generating the second prior image based on the first prior image comprises generating the second prior image based on the segmentation of the first prior image.

7. The method of claim 4, wherein generating the second prior image based on the first prior image comprises transforming pixel values of the first prior image based on an acquisition energy of the second projection dataset.

8. The method of claim 4, wherein applying the corrections comprises generating a first interpolated projection dataset based on a forward projection of the first prior image, and generating a second interpolated projection dataset based on a forward projection of the second prior image.

9. The method of claim 8, wherein applying the corrections further comprises:

calculating a weighting function based on the first projection dataset and the location of the metal;

blending the first interpolated projection dataset with the first projection dataset based on the weighting function; and

blending the second interpolated projection dataset with the second projection dataset based on the weighting function.

10. The method of claim 1, further comprising generating a first metal image and a second metal image based on the metal threshold.

11. The method of claim 1, wherein the first projection dataset comprises a higher energy projection dataset and the second projection dataset comprises a lower energy projection dataset.

12. A method, comprising:

acquiring a first projection dataset and a second projection dataset;

generating a first metal-reduced projection dataset based on the first projection dataset and a second metal-reduced projection dataset based on the second projection dataset and the first projection dataset;

decomposing the first and the second metal-reduced projection datasets into a first metal-reduced material basis and a second metal-reduced material basis, and the first and the second projection datasets into a first material basis and a second material basis;

reconstructing a first metal-reduced material image based on the first metal-reduced material basis, a second metal-reduced material image based on the second metal-reduced material basis, a first material image based on the first material basis, and a second material image based on the second material basis;

generating a first metal image and a second metal image based on the first material image and the second material image;

generating a first metal-corrected image by combining the first metal image and the first metal-reduced material image;

generating a second metal-corrected image by combining the second metal image and the second metal-reduced material image; and

outputting at least one of the first metal-corrected image, the second metal-corrected image, and a mono-energetic image comprising a combination of the first metal-corrected image and the second metal-corrected image.

13. The method of claim 12, wherein generating the first metal-reduced projection dataset and the second metal-reduced projection dataset comprises:

generating a first metal mask based on the first projection dataset, and a second metal mask based on the first metal mask and the second projection dataset;

generating a first prior image based on the first projection dataset and the first metal mask, and a second prior image based on the first prior image;

interpolating the first projection dataset based on the first metal mask and the first prior image, and the second projection dataset based on the second metal mask and the second prior image; and

blending the first interpolated projection dataset with the first projection dataset to generate the first metal-reduced projection dataset, and the second interpolated projection dataset with the second projection dataset to generate the second metal-reduced projection dataset.

14. The method of claim 13, wherein generating the second metal mask based on the first metal mask and the second projection dataset comprises selecting values of the second projection dataset based on the first metal mask.

15. The method of claim 13, wherein generating the second prior image based on at least the first prior image comprises transforming the values of the first prior image to form the second prior image.

16. The method of claim 13, wherein the first interpolated projection dataset and the first projection dataset are blended using a weighting function, the weighting function calculated based on the first projection dataset and the first metal mask, and wherein the second interpolated projection dataset and the second projection dataset are blended using the weighting function.

17. The method of claim 13, wherein generating the first metal image and the second metal image based on the first material image and the second material image comprises:

generating a mono-energetic image based on the first material image and the second material image;

generating a binary metal mask by thresholding the mono-energetic image with a metal threshold calculated based on the first projection dataset; and

generating the first metal image by multiplying the first material image with the binary metal mask, and the second metal image by multiplying the second material image with the binary metal mask.

18. The method of claim 12, wherein the first projection dataset comprises a higher energy projection dataset and the second projection dataset comprises a lower energy projection dataset.

19. An imaging system, comprising:

an x-ray source that emits a beam of x-rays toward an object to be imaged, the x-ray source configured to emit x-rays with a high energy and a low energy;

a detector that receives the x-rays attenuated by the object;

a data acquisition system (DAS) operably connected to the detector; and

a computer operably connected to the DAS and programmed with instructions in non-transitory memory that when executed cause the computer to;

acquire, via the DAS, a first projection dataset and a second projection dataset;

detect a location of metal in the first projection dataset;

generate a first metal-corrected projection dataset based on the first projection dataset and the location of the metal; and

generate a second metal-corrected projection dataset based on the second projection dataset and the location of the metal.

20. The system of claim 19, further comprising a display, and wherein the computer is further programmed with instructions in the non-transitory memory that when executed cause the computer to output an image to the display, the image reconstructed based on the first metal-corrected projection dataset and the second metal-corrected projection dataset.