Breast tomosynthesis with flexible compression paddle

• US 9,943,280 B2
• Filed: 03/07/2016
• Issued: 04/17/2018
• Est. Priority Date: 03/07/2016
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Wang, Xin, Claus, Bernhard Erich Hermann, Davis, Cynthia Elizabeth Landberg, Iordache, Razvan Gabriel
• Primary Examiner(s): Gaworecki, Mark R
• Application Number:
  US15/062,838
• Publication Number:
  US 20170251991A1
• Time in Patent Office: 771 Days
• CPC Class Codes: A61B 2090/3908 Soft tissue, e.g. breast tissue
• CPC Class Codes: A61B 2090/3937 Visible markers
• CPC Class Codes: A61B 2090/3966 Radiopaque markers visible in an X-ray image
• CPC Class Codes: A61B 5/1075 for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer A61B8/0858 takes precedence
• CPC Class Codes: A61B 6/025 Tomosynthesis
• CPC Class Codes: A61B 6/0414 with compression means
• CPC Class Codes: A61B 6/4417 related to combined acquisition of different diagnostic modalities
• CPC Class Codes: A61B 6/461 Displaying means of special interest
• CPC Class Codes: A61B 6/502 for diagnosis of breast, i.e. mammography
• CPC Class Codes: A61B 6/5205 involving processing of raw data to produce diagnostic data
• CPC Class Codes: A61B 6/5217 extracting a diagnostic or physiological parameter from medical diagnostic data
• CPC Class Codes: A61B 6/5258 involving detection or reduction of artifacts or noise
• CPC Class Codes: A61B 8/0825 for diagnosis of the breast, e.g. mammography
• CPC Class Codes: G06T 11/003 Reconstruction from projections, e.g. tomography
• CPC Class Codes: G06T 11/006 Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
• CPC Class Codes: G06T 2207/10112 Digital tomosynthesis [DTS]
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method of breast image reconstruction includes positioning a breast on an imaging system support plate, compressing the breast with a flexible paddle, obtaining imaging data, estimating a breast thickness profile by at least one of placing markers on the breast, performing an image-based analysis of the obtained data, using an auxiliary system, and performing a model-based computation. The three dimensional reconstruction including using a thickness profile of the breast surface in at least one of an iterative reconstruction, a filtered back-projection reconstruction, and a joint reconstruction performed using information obtained from an ultrasound scan. A non-transitory medium having executable instructions to cause a processor to perform the method is also disclosed.

Claims

1. A method of breast image reconstruction, the method comprising:

positioning a breast on a support plate of an imaging system;

compressing the breast with a flexible paddle;

obtaining imaging data of the breast;

estimating a breast thickness profile by at least one of placing markers on the breast, performing an image-based analysis of the obtained data, using an auxiliary system, and performing a model-based computation; and

performing a three dimensional reconstruction of the breast using a thickness profile of the breast surface in at least one of an iterative reconstruction, a filtered back-projection reconstruction, and a joint reconstruction performed using information obtained from an ultrasound scan, the reconstruction including mitigation of effects associated with compression by the flexible compression paddle.

2. The method of claim 1, where performing a three dimensional reconstruction comprises a pre-processing step that utilizes the estimated breast thickness profile and includes at least one of a shading correction including a thickness compensation, a step to enforce consistencies between multiple scan views, and a conversion of pixel values so that pixel values represent an average attenuation along a path length.

3. The method of claim 2, the shading correction thickness compensation including modifying gray scale values in the image such that the resulting image looks like the breast is of uniform constant thickness.

4. The method of claim 1, the three-dimensional reconstruction including an iterative calculation using the breast thickness profile to differentiate between voxels within the breast volume, and voxels external to the breast volume.

5. The method of claim 4, including eliminating from the reconstructed image voxels external to the breast volume.

6. The method of claim 4, including:

designating a thin layer of external voxels which are neighboring to the top surface and a thin layer of internal voxels which are neighboring to the top surface as a buffer layer to the model, performing back-projection to the model, applying a smoothing function over the buffer layer, and using the output as initial input for iterative reconstruction,

wherein artifacts in a forward projection of the image data are removed, and a geometry prior is integrated into the iterative reconstruction.

7. The method of claim 6, the smoothing function being greater at an area closer to a top surface of the breast volume and decreasing at areas closer to a boundary of the breast.

8. The method of claim 1, the estimating a breast thickness profile further including at least one of:

placing markers on the flexible mesh paddle;

embedding x-ray attenuating material in gel of a gel-filled pad;

identifying anatomical markers on a patient'"'"'s skin;

using a camera-based vision system;

performing a gray scale quantification of the imaging data, computing a scatter estimation on the gray scale quantification, and deconvolving the breast thickness from the scatter estimation and the gray scale quantification;

utilizing height readout of an ultrasound probe scanning mechanism;

calculating the scatter estimation by interpolating between scan views of the imaging data using information regarding the resemblance between adjacent scatter maps across view angles;

utilizing the curvature of the compression paddle and compression paddle height reported by the system; and

generating a dataset containing nominal breast thickness data, a skin line profile, compression force data, and a breast thickness profile, calculating a deformation of the breast based on an impact of the compression force data and mechanical properties of the breast, and calculating the breast thickness estimation using the breast deformation result.

9. The method of claim 1, including adjusting images of a reconstructed volume to account for the estimated breast thickness profile;

and

displaying the adjusted images on a display device.

10. The method of claim 9, including displaying anatomical context extracted from the reconstructed volume.

11. The method of claim 9, including displaying the images relative to a coordinate grid that is adapted to the estimated thickness profile.

12. A method of breast image reconstruction, the method comprising:

positioning a breast on a compression plate of an imaging system;

compressing the breast with a flexible paddle;

obtaining imaging data of the breast;

estimating a breast thickness profile by at least one of placing markers on the breast, performing an image-based analysis of the obtained data, using an auxiliary system, performing a model-based computation; and

;

computing a volumetric breast density using the breast thickness profile in at least one of an iterative reconstruction, a filtered back-projection reconstruction, and a joint reconstruction performed using information obtained from an ultrasound scan, the computed volumetric breast density including mitigation of effects associated with compression by the flexible compression paddle.

13. The method of claim 12, wherein computing a volumetric breast density includes at least one of:

reconstructing a three dimensional image of the breast; and

extracting density information from two or more projection images.

14. A non-transitory computer-readable medium having stored thereon instructions which when executed by a processor of a mammography imaging system cause the processor to perform a method of breast image reconstruction, the method comprising:

positioning a breast on a compression plate of an imaging system;

compressing the breast with a flexible paddle;

obtaining imaging data of the breast;

estimating a breast thickness profile by at least one of placing markers on the breast, performing an image-based analysis of the obtained data, using an auxiliary system, and performing a model-based computation; and

performing a three dimensional reconstruction of the breast by including a pre-processing step that utilizes the estimated breast thickness profile and includes at least one of a shading correction including a thickness compensation, a step to enforce consistencies between multiple scan views, and a conversion of pixel values so that pixel values represent an average attenuation along a path length, the reconstruction including mitigation of effects associated with compression by the flexible compression paddle.

15. The medium of claim 14, having stored thereon instructions which when executed cause the processor to perform the step of three dimensional reconstruction by using a thickness profile of the breast surface in at least one of an iterative reconstruction, a filtered back-projection reconstruction, and a joint reconstruction performed using information obtained from an ultrasound scan.

16. The medium of claim 15, having stored thereon instructions which when executed cause the processor to perform the steps of:

designating a thin layer of external voxels which are neighboring to the top surface and a thin layer of internal voxels which are neighboring to the top surface as a buffer layer to the model, performing back-projection to the model, applying a smoothing function over the buffer layer, and using the output as initial input for iterative reconstruction,

wherein artifacts in a forward projection of the image data are removed, and a geometry prior is integrated into the iterative reconstruction.

17. The medium of claim 14, having stored thereon instructions which when executed cause the processor to perform the step of three dimensional reconstruction by including an iterative calculation using the breast thickness profile to differentiate between voxels within the breast volume, and voxels external to the breast volume.

18. The medium of claim 14, having stored thereon instructions which when executed cause the processor to perform at least one of:

placing markers on the flexible mesh paddle;

embedding x-ray attenuating material in gel of a gel-filled pad;

identifying anatomical markers on a patient'"'"'s skin;

using a camera-based vision system;

performing a gray scale quantification of the imaging data, computing a scatter estimation on the gray scale quantification, and deconvolving the breast thickness from the scatter estimation and the gray scale quantification;

utilizing height readout of an ultrasound probe scanning mechanism;

calculating the scatter estimation by interpolating between scan views of the imaging data using information regarding the resemblance between adjacent scatter maps across view angles;

utilizing the curvature of the compression paddle and compression paddle height reported by the system; and

generating a dataset containing nominal breast thickness data, a skin line profile, compression force data, and a breast thickness profile, calculating a deformation of the breast based on an impact of the compression force data and mechanical properties of the breast, and calculating the breast thickness estimation using the breast deformation result.