Slice selection for interpolation-based 3D manual segmentation

• US 10,235,752 B2
• Filed: 11/21/2016
• Issued: 03/19/2019
• Est. Priority Date: 11/21/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Wang, Hongzhi
• Primary Examiner(s): Johns, Andrew W
• Application Number:
  US15/357,667
• Publication Number:
  US 20180144463A1
• Time in Patent Office: 848 Days
• CPC Class Codes: A61B 2576/00 Medical imaging apparatus involving image processing or analysis A61B1/00009, A61B6/52 and A61B8/52 take precedence; image processing per se G06T; details of algorithms for analysing medical images G06T7/0012
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/466 adapted to display 3D data
• CPC Class Codes: A61B 6/469 for selecting a region of interest [ROI]
• CPC Class Codes: A61B 6/503 for diagnosis of the heart
• CPC Class Codes: A61B 6/5211 involving processing of medical diagnostic data
• CPC Class Codes: A61B 6/563 involving image data transmission via a network
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/20128 Atlas-based segmentation
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Slice selection for interpolation-based 3D manual segmentation is provided such that propagation error is minimized during 3D reconstruction. In various embodiments, a plurality of 2D images is read. Each of the plurality of 2D images represents a slice of a 3D volume. Deformable registration is performed between each adjacent pair of the plurality of 2D images. From the deformable registration, propagation error is estimated between each pair of the plurality of 2D images. The plurality of 2D images is clustered into a predetermined number of clusters. A slice is selected for annotation from each of the predetermined number of clusters.

Claims

1. A method comprising:

reading a plurality of 2D images, each of the plurality of 2D images representing a slice of a 3D volume;

performing deformable registration between each adjacent pair of the plurality of 2D images;

from the deformable registration, estimating propagation error between each pair of the plurality of 2D images;

clustering the plurality of 2D images into a predetermined number of clusters;

selecting a slice for annotation from each of the predetermined number of clusters.

2. The method of claim 1, wherein the slice selected from each of the predetermined number of slices is at the center of its cluster.

3. The method of claim 1, wherein clustering the plurality of 2D images comprises applying k-means clustering.

4. The method of claim 1, further comprising:

determining a region of interest mask for each of the plurality of 2D images, and wherein propagation error is estimated within the region of interest.

5. The method of claim 1, wherein estimating propagation error comprises measuring image intensity differences.

6. The method of claim 1, further comprising:

providing the selected slices to a user for annotation.

7. The method of claim 1, wherein clustering comprises repeated clustering to determine a minimum overall propagation error.

8. A computer program product for slice selection for interpolation-based 3D manual segmentation, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:

reading a plurality of 2D images, each of the plurality of 2D images representing a slice of a 3D volume;

performing deformable registration between each adjacent pair of the plurality of 2D images;

from the deformable registration, estimating propagation error between each pair of the plurality of 2D images;

clustering the plurality of 2D images into a predetermined number of clusters;

selecting a slice for annotation from each of the predetermined number of clusters.

9. The computer program product of claim 8, wherein the slice selected from each of the predetermined number of slices is at the center of its cluster.

10. The computer program product of claim 8, wherein clustering the plurality of 2D images comprises applying k-means clustering.

11. The computer program product of claim 8, the method further comprising:

determining a region of interest mask for each of the plurality of 2D images, and wherein propagation error is estimated within the region of interest.

12. The computer program product of claim 8, wherein estimating propagation error comprises measuring image intensity differences.

13. The computer program product of claim 8, the method further comprising:

providing the selected slices to a user for annotation.

14. The computer program product of claim 8, wherein clustering comprises repeated clustering to determine a minimum overall propagation error.

15. A system comprising:

a first datastore comprising a plurality of 2D images, each of the plurality of 2D images representing a slice of a 3D volume;

a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising;

reading the plurality of 2D images from the first datastore;

performing deformable registration between each adjacent pair of the plurality of 2D images;

from the deformable registration, estimating propagation error between each pair of the plurality of 2D images;

clustering the plurality of 2D images into a predetermined number of clusters;

selecting a slice for annotation from each of the predetermined number of clusters.

16. The system of claim 15, wherein the slice selected from each of the predetermined number of slices is at the center of its cluster.

17. The system of claim 15, wherein clustering the plurality of 2D images comprises applying k-means clustering.

18. The system of claim 15, the method further comprising:

determining a region of interest mask for each of the plurality of 2D images, and wherein propagation error is estimated within the region of interest.

19. The system of claim 15, wherein estimating propagation error comprises measuring image intensity differences.

20. The system of claim 15, the method further comprising:

providing the selected slices to a user for annotation.