Method and system for segmenting desired regions in digital mammograms

• US 6,091,841 A
• Filed: 10/14/1999
• Issued: 07/18/2000
• Est. Priority Date: 09/04/1997
• Status: Expired due to Term

• Current Assignee: iCAD Incorporated
• Original Assignee: Qualia Computing Inc. (iCAD Incorporated)
• Inventors: Rosenstengel, John E., Rogers, Steven K., Amburn, Philip, DeSimio, Martin P., Broussard, Randy P., Ochoa, Edward M., Berkey, Telford S., Hoffmeister, Jeffrey W., Rathbun, Thomas F.
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US09/418,383
• Time in Patent Office: 278 Days
• Field of Search: 382/128, 382/132, 382/168, 382/169, 382/171, 382/173, 382/180, 382/256, 382/257, 128/922, 378/37
• US Class Current: 382/132
• CPC Class Codes: A61B 6/502 for diagnosis of breast, i.e. mammography
• CPC Class Codes: B25J 15/04 with provision for the remote detachment or exchange of the head or parts thereof
• CPC Class Codes: B29C 2791/006 Using vacuum
• CPC Class Codes: B29C 2791/007 Using fluid under pressure
• CPC Class Codes: B29C 49/0021 using moulds or mould parts movable in a closed path, e.g. mounted on movable endless supports
• CPC Class Codes: B29L 2023/18 Pleated or corrugated hoses
• CPC Class Codes: G06F 18/41 Interactive pattern learning with a human teacher
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 10/443 by matching or filtering
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images
• CPC Class Codes: Y10S 128/922 including image analysis

Abstract

A method and system for detecting and displaying clustered microcalcifications in a digital mammogram, wherein a single digital mammogram is first automatically cropped to a breast area sub-image which is then processed by means of an optimized Difference of Gaussians filter to enhance the appearance of potential microcalcifications in the sub-image. The potential microcalcifications are thresholded, clusters are detected, features are computed for the detected clusters, and the clusters are classified as either suspicious or not suspicious by means of a neural network. Thresholding is preferably by sloping local thresholding, but may also be performed by global and dual-local thresholding. The locations in the original digital mammogram of the suspicious detected clustered microcalcifications are indicated. Parameters for use in the detection and thresholding portions of the system are computer-optimized by means of a genetic algorithm. The results of the system are optimally combined with a radiologist'"'"'s observation of the original mammogram by combining the observations with the results, after the radiologist has first accepted or rejected individual detections reported by the system.

Claims

1. A method for automatically segmenting a desired region in a digital mammogram image from undesired regions, comprising the steps of:

normalizing the brightness values of the pixels in said digital mammogram image to produce a normalized image;

selecting a seed pixel in said normalized image;

region growing said normalized image starting from said seed pixel to produce a region-grown mask;

closing holes in the region-grown mask to produce a closed mask;

eroding the closed mask to produce an eroded mask; and

dilating the eroded mask to produce a dilated mask.

2. The method according to claim 1 wherein said step of normalizing comprises the steps of:

obtaining the histogram for the digital mammogram image;

determining the range of histogram bins containing substantially all of the digital mammogram pixels corresponding to said desired region;

setting the pixels below said range to a minimum brightness level;

setting the pixels above said range to a maximum brightness level; and

linearly mapping the brightness values of the pixels in said range between said minimum and maximum brightness levels.

3. The method according to claim 1 wherein said step of selecting comprises the steps of:

dividing said digital mammogram into a plurality of substantially equally-sized regions;

determining which of said equally-sized regions is brightest; and

selecting a pixel in said brightest equally-sized region which has a maximum brightness for that region.

4. The method according to claim 1 wherein said region growing step comprises the steps of:

computing the contrast ratio with respect to said seed pixel of one of the four nearest neighbor pixels of said seed pixel;

comparing said contrast ratio with a threshold value; and

setting the corresponding pixel in said region-grown mask to a predetermined value based on the outcome of said comparing step.

5. The method according to claim 1 further comprising the step of:

cropping the digital mammogram image to the largest rectangle that just encloses the digital mammogram pixels corresponding to the dilated mask.

6. A method for segmenting an area of a digital mammogram image corresponding to breast tissue from the remainder of the image, comprising the steps of:

storing a digital representation of said digital mammogram image;

enhancing the digital representation to produce an enhanced image in which the contrast of the area of the mammogram image corresponding to breast tissue is increased;

thresholding the enhanced image to produce a binary image comprising a seed pixel;

region growing said seed pixel in said binary image to produce a mask;

closing holes in said mask;

eroding said mask; and

dilating said mask.

7. The method according to claim 6 further comprising the step of:

cropping said digital representation to the size of the largest object in said mask.

8. A system for automatically segmenting a desired region in a digital mammogram image from undesired regions, comprising:

normalizing means for normalizing the brightness values of the pixels in said digital mammogram image to produce a normalized image;

seed-selecting means for selecting a seed pixel in said normalized image;

region-growing means for region growing said normalized image starting from said seed pixel to produce a region-grown mask;

hole-closing means for closing holes in the region-grown mask to produce a closed mask;

erosion means for eroding the closed mask to produce an eroded mask; and

dilation means for dilating the eroded mask to produce a dilated mask.

9. The system according to claim 8 wherein said normalizing means comprises:

histogram means for obtaining the histogram for the digital mammogram image;

range determining means for determining the range of histogram bins containing substantially all of the digital mammogram pixels corresponding to said desired region;

means for setting the pixels below said range to a minimum brightness level;

means for setting the pixels above said range to a maximum brightness level; and

linear mapping means for linearly mapping the brightness values of the pixels in said range between said minimum and maximum brightness levels.

10. The system according to claim 8 wherein said seed-selecting means comprises:

means for dividing said digital mammogram into a plurality of substantially equally-sized regions;

means for determining which of said equally-sized regions is brightest; and

means for selecting a pixel in said brightest equally-sized region which has a maximum brightness for that region.

11. The system according to claim 8 wherein said region-growing means comprises:

means for computing the contrast ratio with respect to said seed pixel of one of the four nearest neighbor pixels of said seed pixel;

means for comparing said contrast ratio with a threshold value; and

means for setting the corresponding pixel in said region-grown mask to a predetermined value based on the outcome of said comparing step.

12. The system according to claim 8 further comprising:

cropping means for cropping the digital mammogram image to the largest rectangle that just encloses the digital mammogram pixels corresponding to the dilated mask.

13. A system for segmenting an area of a digital mammogram image corresponding to breast tissue from the remainder of the image, comprising:

storage means for storing a digital representation of said digital mammogram image;

image enhancement means for enhancing the digital representation to produce an enhanced image in which the contrast of the area of the mammogram image corresponding to breast tissue is increased;

thresholding means for thresholding the enhanced image to produce a binary image comprising a seed pixel;

region growing means for region growing said seed pixel in said binary image to produce a mask;

hole closing means for closing holes in said mask;

erosion means for eroding said mask; and

dilation means for dilating said mask.

14. The system according to claim 13 further comprising:

cropping means for cropping said digital representation to the size of the largest object in said mask.