Three-dimensional analysis of lesions represented by image data

Patent metadata

Field	Value
Publication number	US-9089307-B2
Application number	US-201013502475-A
Country	US
Kind code	B2
Filing date	Oct 21, 2010
Priority date	Oct 30, 2009
Publication date	Jul 28, 2015
Grant date	Jul 28, 2015

Abstract

A system for three-dimensional analysis of lesions in image data is disclosed. It comprises a lesion detection subsystem ( 1 ) for detecting individual lesions and three-dimensional positions of the individual lesions, based on e.g. breast image data ( 301 ). It comprises a cluster detection subsystem ( 2 ) for detecting a cluster of lesions ( 302 ), based on three-dimensional position information of lesions, and associating at least some of the individual lesions with the cluster of lesions ( 302 ), based on the three-dimensional positions of the individual lesions. The cluster detection subsystem ( 2 ) is arranged for detecting the cluster of lesions ( 302 ), based on the three-dimensional positions of the individual lesions. It comprises a cluster analysis subsystem ( 3 ) for analyzing the cluster of lesions ( 302 ).

First claim

The invention claimed is:

1. A system for three-dimensional analysis of lesions represented by image data, comprising:
a lesion detection subsystem for detecting individual lesions and three-dimensional positions of the individual lesions, based on the image data; and
a cluster detection subsystem for detecting a cluster of lesions, based on three-dimensional position information of lesions, and associating at least some of the individual lesions with the cluster of lesions based on the three-dimensional positions of the individual lesions,
wherein the cluster comprises a plurality of lesions grouped on the basis of a physical proximity of the lesions to each other.

2. The system according to claim 1 , wherein the cluster detection subsystem is arranged for detecting the cluster of lesions, based on the three-dimensional positions of the individual lesions.

3. The system according to claim 1 , wherein the lesion detection subsystem is arranged for detecting the lesions, based on at least one projection and at least one three-dimensional image of the lesions.

4. The system according to claim 1 , further comprising a cluster analysis subsystem for analyzing the cluster of lesions.

5. The system according to claim 4 , wherein the cluster analysis subsystem comprises a surface generator for generating a circumscribing surface containing the cluster of lesions.

6. The system according to claim 5 , further comprising a visualization subsystem for visualizing the circumscribing surface.

7. The system according to claim 4 , wherein the cluster analysis subsystem comprises a shape model subsystem for adapting a shape model to at least part of the cluster of lesions.

8. The system according to claim 7 , wherein the shape model is associated with a region or structure of a breast.

9. The system according to claim 4 , wherein the cluster analysis subsystem comprises at least one of:
a distance computing subsystem for computing a distance between a pair of three-dimensional positions of lesions within the cluster of lesions;
a volume computing subsystem for computing a volume of a lesion within the cluster of lesions;
an absorption coefficient computing subsystem for computing an absorption coefficient of a lesion within the cluster of lesions;
a roughness coefficient computing subsystem for computing a roughness coefficient of a lesion within the cluster of lesions; or
a shape computing subsystem for computing a shape of a lesion of the cluster of lesions.

10. The system according to claim 1 , wherein the cluster detecting subsystem is arranged for processing distances between pairs of three-dimensional positions of lesions.

11. The system according to claim 1 , further comprising a clinical decision support system for evaluating a characteristic of the cluster of lesions.

12. A mammographic tomosynthesis image-forming apparatus comprising the system according to claim 1 .

13. A medical imaging workstation comprising an input for receiving image data and the system according to claim 1 .

14. A method of three-dimensional analysis of lesions represented by image data, comprising:
detecting individual lesions and three-dimensional positions of the individual lesions, based on the image data; and
detecting a cluster of lesions, based on three-dimensional position information of lesions, and associating at least some of the individual lesions with the cluster of lesions based on the three-dimensional positions of the individual lesions,
wherein the cluster comprises a plurality of lesions grouped on the basis of a physical proximity of the lesions to each other.

15. A computer program product embodied on a non-transitory computer-readable medium comprising instructions for causing a processor system to perform the steps of the method according to claim 14 .

Assignees

• Grass Michael
• Hansis Eberhard Sebastian
• Buelow Thomas
• Erhard Klaus
• Koninkl Philips Nv

Inventors

• Grass Michael
• Hansis Eberhard Sebastian
• Buelow Thomas
• Erhard Klaus

Classifications

• G06F18/2321

  using statistics or function optimisation, e.g. modelling of probability density functions · CPC title

• A61B6/466

  adapted to display 3D data · CPC title

• G06T2207/30068

  Mammography; Breast · CPC title

• A61B6/025

  Tomosynthesis · CPC title

• G06K9/6221

  Physics · mapped topic