Abnormal shadow detecting system

• US 7,248,728 B2
• Filed: 03/11/2003
• Issued: 07/24/2007
• Est. Priority Date: 03/11/2002
• Status: Expired due to Fees

• Current Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Original Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Inventors: Takeo, Hideya
• Primary Examiner(s): Johns; Andrew W.
• Application Number:
  US10/384,737
• Publication Number:
  US 20030169915A1
• Time in Patent Office: 1,596 Days
• Field of Search: 382/128, 382/132, 382/173, 382/180, 378/37, 128/922
• US Class Current: 382/132
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

In an abnormal shadow detecting system, a mammary gland distribution map generating system generates a mammary gland distribution map of each of left and right breasts by dividing an image of each of the left and right breasts into a plurality of ranges according to the density of the image on the basis of image data representing the image, and an extracting system extracts a bilaterally asymmetric range by comparing the ranges of the mammary gland distribution map for one of the left and right breasts with those of the mammary gland distribution map for the other breast.

Claims

1. An abnormal shadow detecting system comprising:

a mammary gland distribution map generating means which generates a mammary gland distribution map of each of left and right breasts by dividing an image of each of the left and right breasts into a plurality of ranges according to the density

of the image on the basis of image data representing the image,

and an extracting means which extracts a bilaterally asymmetric range by comparing the ranges of the mammary gland distribution map for one of the left and right breasts with those of the mammary gland distribution map for the other breast.

2. An abnormal shadow detecting system as defined in claim 1 in which the mammary gland distribution map generating means divides an object range of a side view of each breast into a pectoral muscle range, a mammary gland range and a fat range.

3. The system of claim 2, wherein the pectoral muscle range comprises data of pectoral muscle tissue.

4. An abnormal shadow detecting system as defined in claim 1 in which the mammary gland distribution map generating means divides an object range of a front view of each breast into a mammary gland range and a fat range.

5. An abnormal shadow detecting system as defined in claim 1 further comprising a bilaterally asymmetric shadow detecting means which detects a prospective range of the bilaterally asymmetric shadow from the bilaterally asymmetric ranges, extracted by the extracting means, on the basis of information on the bilaterally asymmetric range.

6. An abnormal shadow detecting system as defined in claim 5 which the information on the bilaterally asymmetric range comprises at least one of information on the position in the image of the range, information on the area of the range, information on the variance, and information on the density.

7. The system of claim 5, wherein the prospective range rejects bilaterally asymmetric ranges as a shadow depending on at least one of:

position of the fat range and mammary gland range, a small area of the extracted region, a small variance of the extracted region and a high density (low brightness) of the extracted region.

8. The system of claim 1, further comprising a transform means for aligning points of data for the left and right breasts and wherein the mammary gland distribution map generating means operates on an output of the transform means.

9. The system of claim 1, wherein the extracting means determines a degree of bilateral asymmetry by applying respective weighting coefficients to image data of the left and right breasts based on average density ranges of the data of the left and right breasts and calculates a probability of the shadow region based on the weighting coefficients.

10. An abnormal shadow detecting system comprising a mammary gland distribution map generating means which generates a pair of mammary gland distribution maps of a breast by dividing each of two images of the breast taken at different times into a plurality of ranges according to the density of the image on the basis of image data representing the image, and an extracting means which extracts a difference range where a difference exists between the two images by comparing the ranges of the mammary gland distribution map for one of the images with those of the mammary gland distribution map for the other image.

11. An abnormal shadow detecting system as defined in claim 10 which the mammary gland distribution map generating means divides an object range of a side view of each breast into a pectoral muscle range, a mammary gland range and a fat range.

12. An abnormal shadow detecting system as defined in claim 10 which the mammary gland distribution map generating means divides an object range of a front view of each breast into a mammary gland range and a fat range.

13. An abnormal shadow detecting system as defined in claim 10 further comprising an abnormal shadow detecting means which detects a prospective range of the abnormal shadow from the difference ranges, extracted by the extracting means, on the basis of information on the difference range.

14. An abnormal shadow detecting system as defined in claim 13 in which the information on the difference range may comprise at least one of information on the position in the image of the range, information on the area of the range, information on the variance, and information on the density.