Computer aided detection of masses and clustered microcalcifications with single and multiple input image context classification strategies

• US 6,801,645 B1
• Filed: 06/23/2000
• Issued: 10/05/2004
• Est. Priority Date: 06/23/1999
• Status: Expired due to Term

• Current Assignee: iCAD Incorporated
• Original Assignee: iCAD Incorporated
• Inventors: Mitchell, Richard A., Collins, Michael J., Rogers, Steven K.
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): Akhavannik, Hussein
• Application Number:
  US09/602,762
• Time in Patent Office: 1,565 Days
• Field of Search: 382/130, 382/132, 378/37
• US Class Current: 382/130
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 10/36 Applying a local operator, i.e. means to operate on image points situated in the vicinity of a given point; Non-linear local filtering operations, e.g. median filtering
• CPC Class Codes: G06V 10/443 by matching or filtering
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images

Abstract

A computer aided detection method and system to assist radiologists in the reading of medical images. The method and system has particular application to the area of mammography including detection of clustered microcalcifications and densities. A microcalcification detector is provided wherein individual detections are rank ordered and classified, and one of the features for classification is derived using a multilayer perceptron. A density detector is provided including an iterative, dynamic region growing module with embedded subsystem for rank ordering and classification of a best subset of candidate masks. A post processing stage is provided where detections are analyzed in the context of a set of images for a patient. Three analysis methods are used to distribute a limited number of detections across the image set and further within each image, and additionally to perform a normalcy classification. The normalcy classification is used to remove all detections from an image set when predetermined normalcy conditions are met. The final output of the system is a set of indications overlaid on the input medical images.

Claims

1. In a process for analyzing detections on a digital image set comprising a plurality of images, a method for choosing a subset of detections on the image set including the steps of:

a) determining the number images in the set;

b) performing a calculation using the number of images in the set to determine an upper limit, N, of detections to display;

c) providing a quality score for identified detections on the entire image set;

d) performing a rank ordering of detections across the entire image set based on the quality score assigned to the identified detections; and

e) allowing display of the detections on the image set ranked from 1 to N.

2. The method of claim 1 wherein the quality scores are indicative of the likelihood a detection corresponds to a cancerous region on the image set, and the likelihood of a detection corresponding to a cancerous region decreases proceeding from 1 to N.

3. The method of claim 1 wherein all images in the set comprise images from a single case.

4. The method of claim 1 further comprising

i) individually ranking the microcalcifications and densities on each image;

ii) allowing display of microcalcifications with ranks from 1 to M₁; and

iii) allowing display of densities with ranks from 1 to M₂.

5. The method of claim 4 wherein the limiting number of microcalcifications, M₁, is not equal to the limiting number of densities, M₂.

6. The method of claim 1 wherein the images in the image set comprise a case of images for a patient.

7. The method of claim 1 wherein the upper limit of detections, D, is computed as

8. The method of claim 1 wherein

i) the detections comprise more than one category, i, of detections;

ii) separate upper limits, N_i, are calculated for each category of detections;

iii) the detections in each category of detections are separately rank ordered; and

iv) allowing display of each category of detection with corresponding ranks from 1 to N_i.

9. The method of claim 8 wherein the detections comprise two categories, microcalcifications and densities.

10. The method of claim 9 wherein an upper limit N₁ for microcalcifications is different than an upper limit N₂ for densities.

11. The method of claim 10 wherein only the detections ranked 1 to M are retained on each image for each category of detections, where M is a predetermined per-image upper limit for the number of detections retained on the image.

12. The method of claim 11 wherein the per-image upper limit M is different for each category of detection.

13. The method of claim 12 wherein the per-image upper limit M₁ corresponds to microcalcifications and is different than the per-image upper limit M₂ corresponding to densities.

14. A system for analyzing detections on a digital image set comprising a plurality of images and choosing a subset of detections on the image set, the system comprising:

a) means for determining the number of images in the set;

b) means for performing a calculation using the number of images in the set to determine an upper limit, N, of detections to display;

c) means for providing a quality score for identified detections on the entire image set;

d) means for performing a rank ordering of detections across the entire image set based on the quality score assigned to the identified detections; and

e) means for allowing display of the detections on the image set ranked from 1 to N.

15. The system of claim 14 wherein the quality scores are indicative of the likelihood a detection corresponds to a cancerous region on the image set, and the likelihood of a detection corresponding to a cancerous region decreases proceeding from 1 to N.

16. The system of claim 14 wherein all images in the set comprise images from a single case.

17. The system of claim 14 further comprising

i) individually ranking the microcalcifications and densities on each image;

ii) allowing display of microcalcifications with ranks from 1 to M₁; and

iii) allowing display of densities with ranks from 1 to M₂.

18. The system of claim 17 wherein the number of microcalcifications, M₁, is not equal to the number of density mark densities, M₂.

19. The system of claim 14 wherein the images in the image set comprise a case of images for a patient.

20. The system of claim 14 wherein the upper limit of detections, D, is computed as

21. The system of claim 14 wherein

i) the detections comprise more than one category, i, of detections;

ii) separate upper limits, N_i, are calculated for each category of detections;

iii) the detections in each category of detections are separately rank ordered;

iv) allowing display of each category of detection with corresponding ranks from 1 to N_i.

22. The system of claim 21 wherein the detections comprise two categories, microcalcifications and densities.

23. The system of claim 22 wherein an upper limit N₁ for microcalcifications is different than an upper limit N₂ for densities.

24. The system of claim 23 wherein only the detections ranked 1 to M are retained on each image for each category of detections, where M is a predetermined per-image upper limit for number of detections retained on the image.

25. The system of claim 24 wherein the per-image upper limit M is different for each category of detection.

26. The system of claim 25 wherein the per-image upper limit M₁ corresponds to microcalcifications and is different than the per-image upper limit M₂ corresponding to densities.

27. In a process for analyzing detections on a digital image set comprising a plurality of images, a method for choosing a subset of detections on the image set including the steps of:

a) determining the number images in the set;

b) performing a calculation using the number of images in the set to determine an upper limit, N, of detections to display;

c) providing a quality score for identified detections on the entire image set;

d) performing a rank ordering of detections across the entire image set based on the quality score assigned to the identified detections;

e) retaining the detections on the image set which are ranked 1 to N;

f) subsequently, individually ranking on each image the 1 to N detections of the entire image set within microcalcification and density categories; and

g) allowing display of a maximum of M₁ microcalcifications and a maximum of M₂ densities in each image.

28. In a process for analyzing detections on a digital image set comprising a plurality of images, a method for choosing a subset of detections on the image set including the steps of:

a) providing a quality score for identified detections on the entire image set;

b) performing a rank ordering of detections across the entire image set based on the quality score assigned to the identified detections;

c) allowing display of the detections on the image set which are ranked 1 to N; and

d) ranking the detections in each individual image, separately from the rank ordering performed across the entire image set, to retain a predetermined maximum number of microcalcifications and a predetermined maximum number of densities in each image.