Computer-assisted diagnosis method using correspondence checking and change detection of salient features in digital images

• US 6,453,058 B1
• Filed: 06/07/1999
• Issued: 09/17/2002
• Est. Priority Date: 06/07/1999
• Status: Expired due to Term

• Current Assignee: Siemens AG
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG)
• Inventors: Murthy, Sreerama K., Novak, Carol L., Wu, Zhenyu, Qian, Jianzhong
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): NAKHJAVAN, SHERVIN K
• Application Number:
  US09/326,884
• Time in Patent Office: 1,198 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132, 382/133, 382/155, 382/159, 382/160, 382/181, 382/190, 382/191, 382/204, 382/206, 382/217, 382/218, 382/219, 382/220, 382/224, 382/225, 382/226, 382/276, 382/286, 378/4, 378/5, 378/8, 378/21, 378/37, 378/51, 378/53, 378/54, 378/58, 378/62, 378/63, 378/70, 378/74, 378/87, 378/165, 378/166, 128/915, 600/309, 600/921, 706/424
• US Class Current: 382/128
• CPC Class Codes: A61B 6/502 for diagnosis of breast, i.e. mammography
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 10/757 Matching configurations of points or features

Abstract

A computer assisted diagnosis method using change detection of salient features in a first and a second digital image includes the step of converting the first and the second digital image into a first and a second relational attribute graph, respectively. Each graph comprises nodes and arcs. Each node corresponds to an identified salient feature and associated with information comprising a type and characteristics of the corresponding identified salient feature. The arcs correspond to a topological arrangement of the identified salient features. An optimal inexact structural match is determined between the nodes of the first graph and the nodes of the second graph so as to form matched sets of nodes comprising one node from each graph. A node lacking a determined match is matched with a null node. The characteristics of each of the nodes in a matched set are compared to one another to identify variances among the compared characteristics. A corresponding score is generated for each of the nodes in the matched set based on each of the identified variances. The nodes are identified whose appearances vary more than pre-specified thresholds based on the scores corresponding thereto, as well as the nodes matched with the null nodes. Moreover, the nodes corresponding to salient features which have not changed may be identified.

Claims

1. A computer assisted diagnosis method using change detection of salient features in a first and a second digital image, comprising the steps of:

converting the first and the second digital image into a first and a second relational attribute graph, respectively, wherein each graph comprises nodes and arcs, each node corresponding to an identified salient feature and associated with information comprising a type and characteristics of the corresponding identified salient feature, and the arcs corresponding to a topological arrangement of the identified salient features;

determining an optimal inexact structural match between the nodes of the first graph and the nodes of the second graph so as to form matched sets of nodes comprising one node from each graph, wherein a node lacking a determined match is matched with a null node;

comparing the characteristics of each of the nodes in a matched set to one another to identify variances among the compared characteristics and generating a corresponding score for each of the nodes in the matched set based on each of the identified variances; and

reporting as having changed the nodes whose appearances vary more than pre-specified thresholds based on the scores corresponding thereto and the nodes matched with the null nodes.

2. The method according to claim 1, further comprising the step of identifying the salient features in the first and the second digital image.

3. The method according to claim 1, wherein the characteristics corresponding to masses comprise at least one of size and shape.

4. The method according to claim 1, wherein the characteristics corresponding to microcalcifications comprise at least one of a number of calcifications and cluster shapes.

5. The method according to claim 1, wherein said determining step comprises the step of matching subgraphs of the first graph with subgraphs of the second graph.

6. The method according to claim 1, wherein said determining step employs the arcs corresponding to the topological arrangement of the identified salient features in determining the optimal inexact structural match between the nodes of the first graph and the nodes of the second graph.

7. The method according to claim 1, wherein said comparing step is performed with respect to pre-sized regions of the relational attribute graphs.

8. The method according to claim 1, wherein said comparing step compares the characteristics using a distance measure.

9. The method according to claim 1, wherein the characteristics of a particular node correspond to quantitative measurements of at least one physical attribute of the feature corresponding to the particular node, and said comparing step comprises the step of calculating distances between corresponding quantitative measurements of the nodes in the matched sets.

10. The method according to claim 1, wherein the nodes matched with the null nodes correspond to one of newly appearing and disappearing salient features.

11. The method according to claim 1, wherein said identifying step further comprising the step of identifying the nodes corresponding to salient features which have not changed.

12. The method according to claim 1, wherein said identifying step comprises providing at least one of an audio indication and a visual indication of the identified nodes.

13. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform steps for computer-assisted diagnosis using change detection of salient features in a first and a second digital image, said method steps comprising:

converting the first and the second digital image into a first and a second relational attribute graph, respectively, wherein each graph comprises nodes and arcs, each node corresponding to an identified salient feature and associated with information comprising a type and characteristics of the corresponding identified salient feature, and the arcs corresponding to a topological arrangement of the identified salient features;

determining an optimal inexact structural match between the nodes of the first graph and the nodes of the second graph so as to form matched sets of nodes comprising one node from each graph;

comparing the characteristics of each of the nodes in a matched set to one another to identify variances among the compared characteristics and generating a corresponding score for each of the nodes in the matched set based on each of the identified variances; and

reporting as having changed the nodes whose appearances vary more than pre-specified thresholds based on the scores corresponding thereto and the nodes matched with the null nodes.

14. The program storage device according to claim 13, further comprising the step of identifying the salient features in the first and the second digital image.

15. The program storage device according to claim 13, wherein the characteristics corresponding to masses comprise at least one of size and shape.

16. The program storage device according to claim 13, wherein the characteristics corresponding to microcalcifications comprise at least one of a number of calcifications and cluster shapes.

17. The program storage device according to claim 13, wherein said determining step comprises the step of matching subgraphs of the first graph with subgraphs of the second graph.

18. The program storage device according to claim 13, wherein said determining step employs the arcs corresponding to the topological arrangement of the identified salient features in determining the optimal inexact structural match between the nodes of the first graph and the nodes of the second graph.

19. The program storage device according to claim 13, wherein said comparing step is performed with respect to pre-sized regions of the relational attribute graphs.

20. The program storage device according to claim 13, wherein said comparing step compares the characteristics using a distance measure.

21. The program storage device according to claim 13, wherein the characteristics of a particular node correspond to quantitative measurements of at least one physical attribute of the feature corresponding to the particular node, and said comparing step comprises the step of calculating distances between corresponding quantitative measurements of the nodes in the matched sets.

22. The program storage device according to claim 13, wherein the nodes matched with the null nodes correspond to one of newly appearing and disappearing salient features.

23. The program storage device according to claim 13, wherein said identifying step further comprising the step of identifying the nodes corresponding to salient features which have not changed.

24. The program storage device according to claim 13, wherein said identifying step comprises providing at least one of an audio indication and a visual indication of the identified nodes.

25. A computer assisted diagnosis method using change detection of salient features in a first and a second digital image, comprising the steps of:

generating a first and a second relational attribute graph from the first and the second digital image, respectively, wherein each graph comprises nodes and arcs, each node corresponding to an identified salient feature and associated with information comprising a type and characteristics of the corresponding identified salient feature, and the arcs corresponding to a topological arrangement of the identified salient features;

matching the first graph to the second graph so as to form matched sets of nodes comprising one node from each graph;

comparing the characteristics of each of the nodes in a matched set to one another to identify variances among the compared characteristics;

generating a score for each of the nodes in the matched set based on each of the identified variances; and

reporting as having changed the nodes whose appearances vary more than pre-specified thresholds based on the scores corresponding thereto.