SEGMENTING MEDICAL IMAGE DATA SETS

• US 20070076932A1
• Filed: 09/25/2006
• Published: 04/05/2007
• Est. Priority Date: 09/23/2005
• Status: Active Grant

• Current Assignee: Brainlab AG
• Original Assignee: Brainlab AG
• Inventors: BLUMHOFER, Andreas
• Granted Patent: US 7,822,244 B2
• Time in Patent Office: Days
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/20128 Atlas-based segmentation
• CPC Class Codes: G06T 2207/30012 Spine; Backbone
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method for segmenting a body structure image data set produced from a medical imaging method includes: using a reference data set to define delineated body structures in the body structure image data set; and ascertaining an overall mapping function that substantially maps a reference data set onto the body structure image data set, wherein the overall mapping function comprises a portion that includes a dissection of the body structure image data set into structural parts and their individual rearrangement, and a portion which includes a global deformation and/or shift of the body structures.

Claims

1. A method for segmenting a body structure image data set produced from a medical imaging method, comprising:

using a reference data set to define delineated body structures in the body structure image data set; and

ascertaining an overall mapping function that substantially maps a reference data set onto the body structure image data set, wherein the overall mapping function comprises a portion that includes a dissection of the body structure image data set into structural parts and their individual rearrangement, and a portion which includes a global deformation and/or shift of the body structures.

2. The segmenting method according to claim 1, wherein the individual rearrangement includes individually rotating structural parts relative to each other.

3. The segmenting method according to claim 1, wherein the individual rearrangement includes individually shifting structural parts relative to each other.

4. The segmenting method according to claim 1, wherein using includes ascertaining the structural parts and their individual rearrangement before ascertaining the global deformation and/or shift of the body structures.

5. The segmenting method according to claim 1, wherein using includes ascertaining the structural parts and their individual rearrangement together with ascertaining the global deformation and/or shift of the body structures

6. The segmenting method according to claim 1, wherein using includes taking into account ancillary conditions which are predetermined by the anatomy of the body structure and its relationship to surrounding body structures.

7. The segmenting method according to claim 6, wherein the ancillary conditions include anatomically possible axes of rotation, joint axes, changes in the distance of body structures, and/or groups and/or parts of the same.

8. The segmenting method according to claim 1, wherein the body structure image data set includes at least parts of bone structures and/or bone groups.

9. A computer program embodied on a computer readable medium for segmenting a body structure image data set produced from a medical imaging method, comprising:

code that uses a reference data set to define delineated body structures in the body structure image data set; and

code that ascertains an overall mapping function that substantially maps a reference data set onto the body structure image data set, wherein the overall mapping function comprises a portion that includes a dissection of the body structure image data set into structural parts and their individual rearrangement, and a portion which includes a global deformation and/or shift of the body structures.

10. A device for segmenting a body structure image data set produced using a medical imaging method, comprising:

a processor;

a data input device for inputting the body structure image data set and a reference data set assignable to the body structure image data set;

a memory for storing the data sets;

segmentation logic stored in memory and executable by the processor, the segmentation logic comprising

logic that uses a reference data set to define delineated body structures in the body structure image data set; and

logic that ascertains an overall mapping function that substantially maps a reference data set onto the body structure image data set, wherein the overall mapping function comprises a portion that includes a dissection of the body structure image data set into structural parts and their individual rearrangement, and a portion which includes a global deformation and/or shift of the body structures.

11. A method for segmenting a body structure image data set produced from a medical imaging method, comprising using discontinuous deformation to derive at least a portion of the body structure image data set from a reference data set.

12. A method for mapping a reference data set onto a patient image data set to enhance delineations and/or physical features in the patient data set, wherein the patient data set and/or reference data set are separated into regions such that at least one the regions can be individually rearranged relative to other regions in the data set, whereby any twisting and/or shifting of such regions of the patient'"'"'s body relative to the corresponding regions in the patient data set, can be accommodated.

13. The method of claim 12, wherein the reference data set includes an atlas data set.