Process, system and computer readable medium for pulmonary nodule detection using multiple-templates matching

• US 6,683,973 B2
• Filed: 08/30/2002
• Issued: 01/27/2004
• Est. Priority Date: 11/21/2000
• Status: Active Grant

• Current Assignee: ARCH Development Corporation (University of Chicago)
• Original Assignee: ARCH Development Corporation (University of Chicago)
• Inventors: Li, Qiang, Doi, Kunio, Katsuragawa, Shigehiko
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US10/231,064
• Publication Number:
  US 20030002723A1
• Time in Patent Office: 515 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132, 382/209, 382/216, 382/217, 382/218, 382/219, 382/278, 128/922
• US Class Current: 382/132
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method to determine whether a candidate abnormality in a medical digital image is an actual abnormality, a system which implements the method, and a computer readable medium which stores program steps to implement the method, wherein the method includes obtaining a medical digital image including a candidate abnormality; obtaining plural first templates and plural second templates respectively corresponding to predetermined abnormalities and predetermined non-abnormalities; comparing the candidate abnormality with the obtained first and second templates to derive cross-correlation values between the candidate abnormality and each of the obtained first and second templates; determining the largest cross-correlation value derived in the comparing step and whether the largest cross-correlation value is produced by comparing the candidate abnormality with the first templates or with the second templates; and determining the candidate abnormality to be an actual abnormality when the largest cross-correlation value is produced by comparing the candidate abnormality with the first templates and determining the candidate abnormality to be a non-abnormality when the largest cross-correlation value is produced by comparing the candidate abnormality with the second templates. An actual abnormality is similarly classified as malignant or benign based on further cross-correlation values obtained by comparisons with additional templates corresponding to malignant and benign abnormalities.

Claims

1. In a method to determine whether a candidate abnormality in a medical digital image is an actual abnormality, the improvement comprising:

obtaining at least one first template and at least one second template respectively corresponding to at least one predetermined abnormality and at least one predetermined non-abnormality;

comparing the candidate abnormality in the medical digital image with the obtained first and second templates to determine a degree of matching between the candidate abnormality and the first and second templates; and

determining the candidate abnormality to be an actual abnormality or a non-abnormality based on whether a best match is obtained by comparing the candidate abnormality with the at least one first template or the at least one second template.

2. The method of claim 1, wherein said step of obtaining comprises:

obtaining at least one first template and at least one second template each of which is smaller in size than the candidate abnormality.

3. The method of claim 2, wherein said comparing step comprises:

comparing the candidate abnormality with the obtained first and second templates to derive cross-correlation values between the candidate abnormality and each of the obtained first and second templates; and

shifting the first and second templates in relation to said candidate abnormality to derive cross-correlation values between the candidate abnormality and each of the shifted first and second templates.

4. The method of claim 3, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known non-abnormality structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known non-abnormality structure; and

comparing the candidate second templates to at least one known abnormality structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known abnormality structure.

5. The method of claim 3, further comprising:

obtaining at least one third template and at least one fourth template respectively corresponding to predetermined malignant and predetermined benign abnormalities;

comparing an actual abnormality with the obtained third and fourth templates to determined a degree of matching between the actual abnormality and each of the obtained third and fourth templates; and

classifying the actual abnormality as a malignant abnormality or a non-malignant abnormality based on whether a best match is obtained by comparing the actual abnormality with the at least one third template or the at least one fourth template.

6. The method of claim 2, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known non-abnormality structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known non-abnormality structure; and

comparing the candidate second templates to at least one known abnormality structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known abnormality structure.

7. The method of claim 2, further comprising:

obtaining at least one third template and at least one fourth template respectively corresponding to predetermined malignant and predetermined benign abnormalities;

comparing an actual abnormality with the obtained third and fourth templates to determine a degree of matching between the actual abnormality and each of the obtained third and fourth templates; and

classifying the actual abnormality as a malignant abnormality or a non-malignant abnormality based on whether a best match is obtained by comparing the actual abnormality with the at least one third template or the at least one fourth template.

8. The method of claim 2, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known benign structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known benign structure; and

comparing the candidate second templates to at least one known malignant structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known malignant structure.

9. The method of claim 1, wherein said step of obtaining comprises:

producing additional first and second templates which are mirror images of the obtained first and second templates.

10. The method of claim 1, wherein said step of obtaining comprises:

obtaining at least one additional first template which is a scaled version of the obtained at least one first template.

11. The method of claim 1, wherein said step of obtaining comprises:

obtaining additional first and second templates which are rotated versions of the obtained first and second templates.

12. The method of claim 1, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known non-abnormality structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known non-abnormality structure; and

comparing the candidate second templates to at least one known abnormality structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known abnormality structure.

13. The method of claim 12, further comprising:

obtaining at least one third template and at least one fourth template respectively corresponding to predetermined malignant and predetermined benign abnormalities;

comparing the actual abnormality with the obtained third and fourth templates to determine a degree of matching between the actual abnormality and each of the obtained third and fourth templates; and

classifying the actual abnormality as a malignant abnormality or a non-malignant abnormality based on whether a best match is obtained by comparing the actual abnormality with the at least one third template or the at least one fourth template.

14. The method of claim 1, further comprising:

obtaining at least one third template and at least one fourth template respectively corresponding to predetermined malignant and predetermined benign abnormalities;

comparing an actual abnormality with the obtained third and fourth templates to determine a degree of matching between the actual abnormality and each of the obtained third and fourth templates; and

classifying the actual abnormality as a malignant abnormality or a non-malignant abnormality based on whether a best match is obtained by comparing the actual abnormality with the at least one third template or the at least one fourth template.

23. A system for implementing the method of any one of claims 1-18.

24. A computer readable medium storing a program which when executed by a computer causes the computer to perform the steps recited in any one of claims 1-18.

15. In a method of classifying an abnormality in a medical digital image, the improvement comprising:

obtaining at least one first template and at least one second template respectively corresponding to predetermined malignant and predetermined benign abnormalities;

comparing the candidate abnormality in the medical digital image with the obtained first and second templates to determine a degree of matching between the candidate abnormality and the first and second templates; and

classifying the candidate abnormality to be a malignant abnormality or a benign abnormality based on whether a best match is obtained by comparing the candidate abnormality with the at least one first template or the at least one second template.

16. The method of claim 15, wherein said step of obtaining comprises:

obtaining at least one first template and at least one second template which are smaller in size than said abnormality in said medical digital image.

17. The method of claim 16, wherein said comparing step comprises:

comparing the candidate abnormality with the obtained first and second templates to derive cross-correlation values between the candidate abnormality and each of the obtained first and second templates; and

shifting the first and second templates in relation to the abnormality to derive cross-correlation values between the abnormality and each of the shifted first and second templates.

18. The method of claim 17, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known benign structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known benign structure; and

comparing the candidate second templates to at least one known malignant structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known malignant structure.

19. The method of claim 15, wherein said step of obtaining comprises:

producing additional first and second templates which are mirror images of the obtained at least one first template and at least one second template.

20. The method of claim 15, wherein said step of obtaining comprises:

obtaining at least one additional first template which is a scaled version of the obtained at least one first template.

21. The method of claim 15, wherein said step of obtaining comprises:

obtaining additional first and second templates which are rotated versions of the obtained first and second templates.

22. The method of claim 15, wherein said step of obtaining comprises:

obtaining candidate first templates and candidate second templates;

comparing the candidate first templates to at least one known benign structure and retaining as first templates only those candidate first templates which exhibit a predetermined degree of non-similarity to the at least one known benign structure; and

comparing the candidate second templates to at least one known malignant structure and retaining as second templates only those candidate second templates which exhibit a predetermined degree of non-similarity to the at least one known malignant structure.