Method and apparatus for cone beam breast CT image-based computer-aided detection and diagnosis

Patent metadata

Field	Value
Publication number	US-9392986-B2
Application number	US-201213985517-A
Country	US
Kind code	B2
Filing date	Feb 14, 2012
Priority date	Feb 14, 2011
Publication date	Jul 19, 2016
Grant date	Jul 19, 2016

Abstract

Cone Beam Breast CT (CBBCT) is a three-dimensional breast imaging modality with high soft tissue contrast, high spatial resolution and no tissue overlap. CBBCT-based computer aided diagnosis (CBBCT-CAD) technology is a clinically useful tool for breast cancer detection and diagnosis that will help radiologists to make more efficient and accurate decisions. The CBBCT-CAD is able to: 1) use 3D algorithms for image artifact correction, mass and calcification detection and characterization, duct imaging and segmentation, vessel imaging and segmentation, and breast density measurement, 2) present composite information of the breast including mass and calcifications, duct structure, vascular structure and breast density to the radiologists to aid them in determining the probability of malignancy of a breast lesion.

First claim

What is claimed is:

1. A method for imaging a breast of a patient, the method comprising:
(a) taking three-dimensional (3D) cone-beam breast computed tomography (CBBCT) projection image data of the breast;
(b) processing the projection image data taken in step (a) in a computing device for pre-reconstruction correction of the projection image data;
(c) performing cone beam reconstruction of a matrix of 3D distribution of a linear attenuation coefficient of the breast in the computer device;
(d) performing correction and processing after the reconstruction of the matrix of the breast in the computer device;
(e) storing the corrected reconstruction matrix into an image archive device;
(f) using a CPU-based computer device and a GPU-based computer device to perform the following plurality of tasks: detecting and characterizing mass, detecting calcifications, assessing breast density and detecting and characterizing duct system of the breast, based on the reconstructed 3D data;
(g) outputting the composite breast information including locations, shapes and size(s) of mass(es) and calcifications, number of calcifications in volume of interest, breast density measurement and shapes of duct system of the breast; and
(h) outputting the probability of malignancy of the breast based on the composite breast information determined in (g);
wherein step (d) comprises: (a) detecting the skin of the breast and to remove the skin of the breast from the image data; (b) applying a stepwise fuzzy clustering algorithm in the computing device to the image data with the skin removed to adaptively cluster image voxels in the image data; (c) dividing the image data in the computing device into a plurality of categories; and (d) forming a three-dimensional image in the computing device from the image data divided into the plurality of categories.

2. The method of claim 1 , wherein the plurality of categories comprise skin, fat, and glands.

3. The method of claim 1 , wherein step (b) comprises: (i) separating the image data into air, fat, and tissues, using a histogram thresholding method; (ii) eroding the tissue between the air and the fat, using a morphological three-dimensional erosion operation with an erosion kernel; and (iii) stopping step (b)(ii) when the erosion kernel reaches a fat area in an inner portion of the breast.

4. The method of claim 1 , wherein step (c) comprises: (A) separating the image data into air, fat, tissues, and a bias area; and (B) re-clustering the bias area to classify the bias area as air, fat, or tissues.

5. The method of claim 4 , wherein the bias area arises from at least one of low-density glandular tissue and imaging artifacts.

6. The method of claim 1 , wherein step (b) comprises: (i) recording a global ratio of a fat area to a total edge area along the edge of the breast; and (ii) stopping step (b) when the ratio reaches a predetermined value.

7. The method of claim 1 , wherein the fuzzy clustering algorithm is a fuzzy c-means algorithm.

8. The method of claim 1 , wherein (f), breast density is measured.

9. The method of claim 1 , wherein (f) breast masses are detected.

10. The method of claim 1 , wherein (f), breast calcifications are detected.

11. The method of claim 1 , wherein (f) further comprising automatically or semi-automatically detecting and characterizing at least one of mass, calcifications, duct structure, vascular structure, and breast density.

12. A system for imaging a breast of a patient, the system comprising:
an imaging apparatus for taking three-dimensional (3D) cone-beam breast computed tomography (CBBCT) projection image data of the breast;
at least a computer device configured to process the projection image data taken in step (a) in a computing device for pre-reconstruction correction of the projection image data;
to perform cone beam reconstruction of a matrix of 3D distribution of linear attenuation coefficient of the breast;
to perform correction and processing after the reconstruction of the matrix of the breast in the computer device;
a massive data storage/archive device to store the corrected reconstruction matrix into image archive device;
at least a computer device comprising a CPU-based computer device and a GPU-based computer device to perform following plurality of tasks: detecting and characterizing mass, detecting calcifications, assessing breast density and detecting and characterizing duct system of the breast, based on the reconstructed 3D data;
outputting the composite breast information including locations, shapes and size(s) of mass(es) and calcifications, number of calcifications in volume of interest, breast density measurement and shapes of duct system of the breast;
outputting the probability of malignancy of the breast based on the composite breast information determined in previous step;
said system further comprising: a processor, in communication with the imaging apparatus, configured for processing the image data taken by the imaging apparatus to detect the skin of the breast and to remove the skin of the breast from the image data; applying a stepwise fuzzy clustering algorithm to the image data with the skin removed to adaptively cluster image voxels in the image data;
dividing the image data into a plurality of categories; and forming a three-dimensional image from the image data divided into the three categories; and an output, in communication with the processor, for outputting the image.

13. The system of claim 12 , wherein the plurality of categories comprise skin, fat, and glands.

14. The system of claim 13 , wherein the processor is configured for: (i) separating the image data into air, fat, and tissues, using a histogram thresholding method; (ii) eroding the tissue between the air and the fat, using a morphological three-dimensional erosion operation with an erosion kernel; and (iii) stopping step (ii) when the erosion kernel reaches a fat area in an inner portion of the breast.

15. The system of claim 14 , wherein the processor is configured such that step (i) comprises: (A) separating the image data into air, fat, tissues, and a bias area; and (B) re-clustering the bias area to classify the bias area as air, fat, or tissues.

16. The system of claim 12 , wherein the processor is configured for a situation in which the bias area arises from at least one of low-density glandular tissue and imaging artifacts.

17. The system of claim 12 , wherein the processor is configured for: (i) recording a global ratio of a fat area to a total edge area along the edge of the breast; and (ii) stopping step (b) when the ratio reaches a predetermined value.

18. The system of claim 12 , wherein the processor is configured such that the fuzzy clustering algorithm is a fuzzy c-means algorithm.

19. The system of claim 12 , wherein the processor is configured to measure breast density.

20. The system of claim 12 , wherein the processor is configured to detect breast masses.

21. The system of claim 12 , wherein the processor is configured to detect breast calcifications.

22. The system of claim 12 , wherein the processor is configured for automatically or semi-automatically detecting and characterizing at least one of mass, calcifications, duct structure, vascular structure, and breast density.

Assignees

• Ning Ruola
• Zhang Xiaohua
• Conover David L
• Liu Shaohua
• Univ Rochester
• Koning Corp

Inventors

• Ning Ruola
• Zhang Xiaohua
• Conover David L
• Liu Shaohua

Classifications

• G06T2207/10081

  Computed x-ray tomography [CT] · CPC title

• G06T7/0012

  Biomedical image inspection · CPC title

• A61B6/4435

  the source unit and the detector unit being coupled by a rigid structure · CPC title

• A61B6/502 Primary

  for diagnosis of breast, i.e. mammography · CPC title

• A61B6/5217

  extracting a diagnostic or physiological parameter from medical diagnostic data · CPC title