Granted Patent B2

US 12,446,842 · App. 18/590,033 · Granted Oct 21, 2025

Inventors: Haili Chui (Fremont, CA); Liyang Wei (San Jose, CA); Jun Ge (Cupertino, CA); Xiangwei Zhang (Fremont, CA); Nikolaos Gkanatsios (Newtown, CT)

Assignee: Hologic, Inc.

A61B6/502A61B6/025G06F18/254G06T7/0012G06T17/10G06V10/806G06V10/809G06V30/2504G06T2207/30068

Abstract

Claims (49)

1. A method for processing breast tissue image data, comprising:

providing a set of image slices that collectively depict at least a portion of a patient's breast tissue, wherein the set of image slices are generated by one of a tomosynthesis acquisition system and a combination tomosynthesis/mammography system;

processing the set of image slices using a first-level feature module to detect at least one first assigned feature of a high-dimensional object present in the patient's breast tissue;

generating a first-level feature map based on the at least one first assigned feature;

processing the set of image slices using a second-level feature module to detect at least one second assigned feature of the high-dimensional object present in the patient's breast tissue,

wherein the at least one first assigned feature is a different level feature than the at least one second assigned feature;

generating a second-level feature map based on the at least one second assigned feature;

and

combining the first-level feature map and the second-level feature map into an object map that indicates a probability region of the high-dimensional object in each image slice.

2. The method of claim 1 , wherein the first-level feature module is configured to recognize the at least one first assigned feature by applying one or more first-level recognition models, and wherein the second-level feature module is configured to recognize the at least one second assigned feature by applying one or more second-level recognition models.

3. The method of claim 1 , wherein the first-level feature module is configured to recognize the at least one first assigned feature by applying one or more first-level recognition templates, and wherein the second-level feature module is configured to recognize the at least one second assigned feature by applying one or more second-level recognition templates.

4. The method of claim 1 , wherein the first-level feature module is configured to recognize the at least one first assigned feature by applying one or more first-level recognition filters, and wherein the second-level feature module is configured to recognize the at least one second assigned feature by applying one or more second-level recognition filters.

5. The method of claim 1 , wherein the first-level feature module is configured to recognize the at least one first assigned feature by applying one or more first-level recognition filters, and wherein the second-level feature module is configured to recognize the at least one second assigned feature by applying one or more second-level recognition models.

6. The method of claim 1 , further comprising creating a two-dimensional synthesized image of the patient's breast tissue based at least in part on the object map.

7. The method of claim 1 , further comprising identifying the high-dimensional object based at least in part on the object maps generated for each of the image slices.

8. The method of claim 1 , wherein combining the first-level feature map and the second-level feature map into the object map comprises combining, by a learning library-based combiner, the first-level feature map and the second-level feature map into the object map.

9. The method of claim 8 , further comprising assigning a first weight to the first-level feature map and a second weight to the second-level feature map.

10. The method of claim 9 , further comprising adjusting at least one of the first weight and the second weight.

11. The method of claim 1 , wherein the at least one first assigned feature is a high-level feature or a mid-level feature.

12. The method of claim 1 , wherein the at least one second assigned feature is a mid-level feature or a low-level feature.

13. The method of claim 1 , further comprising:

processing the set of image slices using a third-level feature module to detect at least one third assigned feature of the high-dimensional object present in the patient's breast tissue, wherein the at least one third assigned features is a different level feature than the at least one first and second assigned features;

generating a third-level feature map based on the at least one third assigned feature; and

combining the third-level feature map into the object map.

14. The method of claim 13 , wherein the at least one third assigned feature is a mid-level feature.

15. The method of claim 1 , wherein the probability region indicates one or more of a location, a size, and a scope of the high-dimensional object.

16. The method of claim 1 , wherein the probability region comprises a probability gradient.

17. The method of claim 1 , wherein the high-dimensional object comprises a three-dimensional image object.

18. The method of claim 1 , further comprising generating a three-dimensional volumetric object map based upon respective first-, second- and third-level object maps.

19. A system comprising:

a non-transitory computer-readable memory storing executable instructions; and

one or more processors in communication with the computer-readable memory, wherein,

when the one or more processors execute the executable instructions, the one or more processors perform:

providing a set of image slices that collectively depict at least a portion of a patient's breast tissue, wherein the set of image slices are generated by one of a tomosynthesis acquisition system and a combination tomosynthesis/mammography system;

processing the set of image slices using a first-level feature module to detect at least one first assigned feature of a high-dimensional object present in the patient's breast tissue;

generating a first-level feature map based on the at least one first assigned feature;

processing the set of image slices using a second-level feature module to detect at least one second assigned feature of the high-dimensional object present in the patient's breast tissue, wherein the at least one first assigned feature is a different level feature than the at least one second assigned feature;

generating a second-level feature map based on the at least one second assigned feature;

and

combining the first-level feature map and the second-level feature map into an object map that indicates a probability region of the high-dimensional object in each image slice.

20. A non-transitory computer readable medium having stored thereon one or more sequences of instructions for causing one or more processors to perform:

providing a set of image slices that collectively depict at least a portion of a patient's breast tissue, wherein the set of image slices are generated by one of a tomosynthesis acquisition system and a combination tomosynthesis/mammography system;

processing the set of image slices using a first-level feature module to detect at least one first assigned feature of a high-dimensional object present in the patient's breast tissue;

generating a first-level feature map based on the at least one first assigned feature;

processing the set of image slices using a second-level feature module to detect at least one second assigned feature of the high-dimensional object present in the patient's breast tissue,

wherein the at least one first assigned feature is a different level feature than the at least one second assigned feature;

generating a second-level feature map based on the at least one second assigned feature;

and

combining the first-level feature map and the second-level feature map into an object map that indicates a probability region of the high-dimensional object in each image slice.

Assignments (4)

Continuity (4)

References Cited (400)