Method for the artifact correction of three-dimensional volume image data

Patent metadata

Field	Value
Publication number	US-11568585-B2
Application number	US-201916710775-A
Country	US
Kind code	B2
Filing date	Dec 11, 2019
Priority date	Dec 17, 2018
Publication date	Jan 31, 2023
Grant date	Jan 31, 2023

Abstract

A method for the artifact correction of three-dimensional volume image data of an object is disclosed. In an embodiment, the method includes receiving first volume image data via a first interface, the first volume image data being based on projection measurement data acquired via a computed tomography device, the computed tomography device including a system axis, and the first volume image data including an artifact including high-frequency first portions in a direction of a system axis and including second portions, being low-frequency relative to the high-frequency first portions, in a plane perpendicular to the system axis; ascertaining, via a computing unit, artifact-corrected second volume image data by applying a trained function to the first volume image data received; and outputting the artifact-corrected second volume image data via a second interface.

First claim

The invention claimed is:

1. A method for artifact correction of three-dimensional volume image data of an object, comprising:
receiving first volume image data via a first interface, the first volume image data being based on projection measurement data acquired via a computed tomography device, the first volume image data including an artifact, the artifact including high-frequency first portions in a direction of a first system axis of the computed tomography device, the artifact including second portions in a plane perpendicular to the first system axis, and the second portions being lower frequency than the first portions;
ascertaining, via a computing unit, artifact-corrected second volume image data by applying a trained function to the first volume image data, the trained function being based on first artifact-affected training volume image data of a training object and second substantially artifact-free training volume image data of the training object, the first artifact-affected training volume image data including a training artifact, the training artifact having high-frequency third portions in a direction of a first axis and fourth portions in a plane perpendicular to the first axis, and the fourth portions being lower frequency than the third portions; and
outputting the artifact-corrected second volume image data via a second interface.

2. The method of claim 1 , wherein the projection measurement data is based on a sub-sampling in the direction of the first system axis.

3. The method of claim 2 , wherein the artifact is a windmill artifact.

4. The method of claim 2 , wherein the trained function is based on a neural network.

5. The method of claim 2 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes:
simulated training volume image data; or
measured training volume image data.

6. The method of claim 1 , wherein the artifact is a windmill artifact.

7. The method of claim 1 , wherein the artifact is a cone beam artifact.

8. The method of claim 1 , wherein the trained function is based on a neural network.

9. The method of claim 1 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes:
simulated training volume image data; or
measured training volume image data.

10. The method of claim 9 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes at least one of:
first image data based on first training projection measurement data acquired via a first training computed tomography device, the first training computed tomography device including a spring focus in a direction of a first training system axis of the first training computing tomography apparatus;
second image data based on second training projection measurement data acquired via a second training computed tomography device, the second training projection measurement data being combined in a direction of a second training system axis of the second training computed tomography device; or
third image data reconstructed by way of an iterative reconstruction algorithm.

11. The method of claim 1 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes at least one of:
first image data based on first training projection measurement data acquired via a first training computed tomography device, the first training computed tomography device including a spring focus in a direction of a first training system axis of the first training computed tomography device;
second image data based on second training projection measurement data acquired via a second training computed tomography device, the second training projection measurement data being combined in a direction of a second training system axis of the second training computed tomography device; or
third image data reconstructed by way of an iterative reconstruction algorithm.

12. A non-transitory computer readable medium storing a computer program, directly loadable into a memory of a system, including program sections to carry out the method for artifact correction of claim 1 , when the program sections are executed by at least one processor of the system.

13. A non-transitory computer-readable storage medium, storing program sections readable and executable by at least one of a correction system or a training system, to carry out the method for the artifact correction of claim 1 , when the program sections are executed by the at least one of the correction system or the training system.

14. A correction system for artifact correction of three-dimensional volume image data of an object, comprising:
a first interface to receive first three-dimensional volume image data, the first three-dimensional volume image data being based on projection measurement data acquired via a computed tomography device, the three-dimensional first volume image data including an artifact, the artifact including high-frequency first portions in a direction of a system axis of the computed tomography device, the artifact including second portions in a plane perpendicular to the system axis, and the second portions being lower frequency than the first portions;
processing circuitry to ascertain artifact-corrected second volume image data by applying a function to the first three-dimensional volume image data, the function being trained by way of a machine learning method based on first artifact-affected training volume image data of a training object and second substantially artifact-free training volume image data of the training object, the first artifact-affected training volume image data including a training artifact, the training artifact having high-frequency third portions in a direction of a first axis and fourth portions in a plane perpendicular to the first axis, and the fourth portions being lower frequency than the third portions; and
a second interface to output the artifact-corrected second volume image data.

15. The correction system of claim 14 , wherein the projection measurement data is based on a sub-sampling in the direction of the system axis.

16. A computed tomography device, embodied for acquiring projection measurement data of an object, the computed tomography device comprising the correction system of claim 14 .

17. The correction system of claim 14 , wherein the artifact is a windmill artifact.

18. The correction system of claim 14 , wherein the artifact is a cone beam artifact.

19. The correction system of claim 14 , wherein the machine learning method is based on a neural network.

20. The correction system of claim 14 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes:
simulated training volume image data; or
measured training volume image data.

21. The correction system of claim 14 , wherein the first artifact-affected training volume image data or the second substantially artifact-free training volume image data includes at least one of:
first image data based on first training projection measurement data acquired via a first training computed tomography device, the first training computed tomography device including a spring focus in a direction of a first training system axis of the first training computed tomography devic

Assignees

• Siemens Healthcare Gmbh

Inventors

• Schoendube Harald
• Sunnegaardh Johan

Classifications

• G06T2211/424

  Iterative · CPC title

• A61B6/4078

  Fan-beams · CPC title

• A61B6/5258

  involving detection or reduction of artifacts or noise · CPC title

• G06T11/008 Primary

  Physics · mapped topic

• A61B6/4085

  Cone-beams · CPC title