Method for classifying breast tissue density using computed image features

• US 7,903,861 B2
• Filed: 12/28/2006
• Issued: 03/08/2011
• Est. Priority Date: 12/28/2006
• Status: Active Grant

• Current Assignee: CareStream Health Incorporated
• Original Assignee: CareStream Health Incorporated
• Inventors: Fletcher-Heath, Lynn M., Barski, Lori L., Luo, Hui
• Primary Examiner(s): Mehta; Bhavesh M
• Assistant Examiner(s): Koziol; Stephen R
• Application Number:
  US11/616,953
• Publication Number:
  US 20080159613A1
• Time in Patent Office: 1,531 Days
• Field of Search: 382/128, 382/132
• US Class Current: 382/132
• CPC Class Codes: A61B 6/502 for diagnosis of breast, i.e. mammography
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/12 Edge-based segmentation

Abstract

A method for classifying tissue density of a breast includes obtaining mammography image data and segmenting the mammography image to identify the region representing the breast tissue. A plurality of regions within the breast tissue region are identified for obtaining image features therefrom. A plurality of image features are computed from the identified plurality of regions. The breast tissue density is classified using the computed plurality of image features.

Claims

1. A method for classifying tissue density of a breast, comprising:

accessing mammography image data of the breast;

segmenting the mammography image to identify a breast tissue region representing the breast tissue;

identifying a plurality of regions of interest within the breast tissue region for obtaining image features therefrom;

computing a plurality of image features from the identified plurality of regions of interest; and

classifying the tissue density of the breast using the computed plurality of image features,

wherein the step of identifying a plurality of regions comprises;

extracting a contour of the breast tissue region;

generating a plurality of contour segments substantially parallel to the breast tissue region contour and inside the breast tissue region; and

defining a plurality of regions along the generated plurality of contour segments.

2. The method of claim 1 further comprising normalizing the breast tissue image content.

3. The method of claim 1 wherein computing a plurality of image features includes forming one or more of the following:

a gray-value histogram, a gradient histogram, and a texture value histogram.

4. The method of claim 1 wherein classifying the tissue density further comprises generating a probability vector for the image, wherein each element in the probability vector represents a confidence level of a tissue density from each of said plurality of tissue density densities.

5. The method of claim 4 wherein the tissue density of a patient is determined according to the combination of probability vectors of a plurality of mammograms obtained for the patient in an examination.

6. The method of claim 1 further comprising displaying the classification results along with a mammography image.

7. The method of claim 1 further comprising storing the classification results along with the image data.

8. The method of claim 1 further comprising rendering one or more of the mammography images according to the classification of breast density type.

9. The method of claim 1 further comprising:

determining tissue density classification for a patient at a first and second time period; and

plotting the tissue density classification as a function of time.

10. The method of claim 1 wherein said generated plurality of contour segments are at pre-defined distances from the breast tissue region contour.

11. The method of claim 1 wherein each of said plurality of regions has a center and wherein the plurality of regions are arranged by setting the center of each region along a contour with a preset distance between neighboring regions.

12. The method of claim 1 including determining a shape and size of a particular region of interest based on the characteristics of the breast tissue within that particular region of interest.

13. The method of claim 1 wherein the step of computing the plurality of image features includes computing a feature vector from the statistics of image features obtained from the portion of image formed by combining all ROIs.

14. The method of claim 1 wherein the plurality of image features includes one or more of:

pixel image intensity, gradient magnitude per pixel, and texture value per pixel.