Automatic estimating and reducing scattering in computed tomography scans

Patent metadata

Field	Value
Publication number	US-10327727-B2
Application number	US-201715593342-A
Country	US
Kind code	B2
Filing date	May 12, 2017
Priority date	May 12, 2017
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

In accordance with at least some embodiments of the present disclosure, a process to estimate scattered radiation contained in x-ray projections for computed tomography (CT) reconstruction is provided. The process may construct an object model based on a plurality of projection images generated by CT scanning of an object using an x-ray radiation source and a detector panel. The process may construct a virtual radiation source based on the x-ray radiation source, and a virtual detector panel based on the detector panel. The process may perform a simulated CT scanning of the object model by simulating macroscopic behavior of particles being emitted from the virtual radiation source, passing through the object model, and being detected by the virtual detector panel. And the process may generate a simulated scatter image based on a first subset of particles scattered during the simulated CT scanning of the object model.

First claim

We claim:

1. A method to estimate scattered radiation contained in x-ray projections for computed tomography (CT) reconstruction, comprising:
constructing an object model based on a plurality of projection images generated by CT scanning of an object using an x-ray radiation source and a detector panel, wherein the object model contains a plurality of voxels, and the constructing of the object model comprises assigning each of the plurality of voxels to at least one material type with density based on each of the plurality of voxels' Hounsfield Units (HU) value;
constructing a virtual radiation source based on the x-ray radiation source;
constructing a virtual detector panel based on the detector panel;
performing a simulated CT scanning of the object model by simulating macroscopic behavior of particles being emitted from the virtual radiation source, passing through the object model, and being detected by the virtual detector panel;
constructing a Boltzmann Transport Equation (BTE) for a first subset of particles scattered during the simulated CT scanning of the object model;
using a deterministic method to solve the BTE and calculate the corresponding particle fluence distribution values in the plurality of voxels; and
generating a simulated scatter image based on the corresponding particle fluence distribution values in the plurality of voxels.

2. The method as recited in claim 1 , further comprising:
estimating a scatter-reduced image corresponding to one of the plurality of projection images based on the simulated scatter image; and
reconstructing a CT object volume with reduced scatter artifacts based on the scatter-reduced image.

3. The method as recited in claim 2 , wherein the estimating of the scatter-reduced image comprising:
generating a simulated primary image based on a second subset of particles attenuated but not scattered during the simulated CT scanning of the object model;
generating a gain map based on the simulated primary image and the simulated scatter image; and
estimating the scatter-reduced image by adjusting the plurality of projection images based on the simulated scatter image and the gain map.

4. The method as recited in claim 2 , wherein the estimating of the scatter-reduced image comprising:
estimating the scatter-reduced image by adjusting one of the plurality of projection images using subtraction or perturbation based on the simulated scatter image.

5. The method as recited in claim 1 , wherein the plurality of projection images are generated using a bowtie filter, and the performing of the simulated CT scanning of the object model further comprising:
constructing a virtual bowtie filter based on the bowtie filter; and
simulating the macroscopic behavior of the particles passing through the virtual bowtie filter after being emitted from the virtual radiation source.

6. The method as recited in claim 1 , wherein the plurality of projection images are generated using an anti-scatter grid, and the performing of the simulated CT scanning of the object model further comprising:
constructing a virtual anti-scatter grid based on the anti-scatter grid; and
simulating the macroscopic behavior of the particles passing through the virtual anti-scatter grid before being detected by the virtual detector panel.

7. The method as recited in claim 1 , wherein the constructing of the object model comprising:
constructing a virtual patient table into the object model.

8. The method as recited in claim 7 , wherein the constructing of the object model further comprising:
extending the object model in its axial direction or longitudinal direction to account for truncation occurred during the CT scanning.

9. The method as recited in claim 1 , wherein the performing of the simulated CT scanning of the object model by simulating of the macroscopic behavior of particles comprising:
simulating the particles emitted from the virtual radiation source, passed through the plurality of voxels, and detected by a plurality of virtual pixels in the virtual detector panel; and
identifying the first subset of particles scattered during the simulating of the particles.

10. The method as recited in claim 9 , wherein the simulating of the particles comprising:
transporting the particles emitted from the virtual radiation source through the plurality of voxels to calculate a set of scattering sources;
transporting the particles from the set of scattering sources across the plurality of voxels;
iterating through the calculation of scattering sources and transporting of the particles; and
transporting the particles from the plurality of voxels to the plurality of pixels in the virtual detector panel.

11. The method as recited in claim 1 , further comprising:
ray-tracing the first subset of particles from the virtual radiation source to the object model prior to the constructing of the BTE; and
ray-tracing the first subset of particles from the object model to the virtual detector panel after the constructing of the BTE.

12. The method as recited in claim 1 , wherein the spatial resolution of the plurality of pixels in the virtual detector panel is coarser than pixels in the detector panel, and the generating the simulated scatter image based on the corresponding particle fluence distribution values in the plurality of pixels further comprising:
using interpolation to up-sample the simulated scatter image to match spatial resolution of one of the plurality of the projection images.

13. A method to determine scattered radiation contained in x-ray projections for computed tomography (CT) reconstruction, comprising:
reconstructing a first CT volume based on a plurality of projection images generated by CT scanning of an object using an x-ray radiation source and a detector panel;
constructing an object model based on the CT volume, wherein the object model contains a plurality of voxels, and the constructing of the object model comprises assigning each of the plurality of voxels to at least one material type with density based on each of the plurality of voxels' Hounsfield Units (HU) value;
constructing a virtual CT scanning environment having a virtual radiation source based on the x-ray radiation source and having a virtual detector panel based on the detector panel, wherein the virtual detector panel contains a plurality of virtual pixels;
performing a simulated CT scanning of the object model by simulating macroscopic behavior of particles being emitted from the virtual radiation source, passing through the object model, and being detected by the virtual detector panel;
generating a plurality of simulated scatter images corresponding to the plurality of projection images by estimating a first subset of particles scattered during the simulated CT scanning of the object model, wherein each simulated scatter image in the plurality of simulated scatter images is generated by
constructing a Boltzmann Transport Equation (BTE) for the first subset of particles scattered during the simulated CT scanning of the object model,
using a deterministic method to solve the BTE and calculate corresponding particle fluence distribution values in the plurality of voxels, and
generating the simulated scatter image based on the corresponding particle fluence distribution values in the plurality of voxels; and
reconstructing a second CT volume with reduced scatter artifacts based on the plurality of projection images and the plurality of simulated scatter images.

14. The method as recited in claim 13 , wherein a first subset of the plurality of simulated scatter images are calculated based on a subset of the pl

Assignees

• Varian Med Sys Inc

Inventors

• Maslowski Alexander E
• Wang Adam
• Star-Lack Josh
• Sun Mingshan
• Wareing Todd

Classifications

• G16H20/40 Primary

  relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture · CPC title

• G01N23/046

  using tomography, e.g. computed tomography [CT] · CPC title

• A61B6/5282 Primary

  due to scatter · CPC title

• A61B6/483

  involving scattered radiation · CPC title

• A61B6/032

  Transmission computed tomography [CT] · CPC title