IMAGE SEGMENTATION WITH ACTIVE CONTOUR

• US 20190333225A1
• Filed: 04/25/2018
• Published: 10/31/2019
• Est. Priority Date: 04/25/2018
• Status: Active Grant

• Current Assignee: MIM Software Incorporated
• Original Assignee: MIM Software Incorporated
• Inventors: WAN, HANLIN, JACOBS, PAUL, BROCKWAY, JERIMY, PIPER, JONATHAN
• Granted Patent: US 11,335,006 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/10104 Positron emission tomography [PET]
• CPC Class Codes: G06T 2207/20096 Interactive definition of curve of interest
• CPC Class Codes: G06T 2207/20101 Interactive definition of point of interest, landmark or seed
• CPC Class Codes: G06T 2207/20112 Image segmentation details
• CPC Class Codes: G06T 2207/20116 Active contour; Active surface; Snakes
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/10 Segmentation; Edge detection motion-based segmentation G06T7/215
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images
• CPC Class Codes: G06V 40/162 using pixel segmentation or colour matching

Abstract

An image segmentation system is discloses that provides one or more possible contours of a feature of an image, in parallel, that respectively correspond to different interpretations of the image. First, a relatively large set of possible contours are generated in accordance with an image segmentation algorithm. Subsequently, this set of possible contours is reduced to a few candidates reflecting representative solutions corresponding to one or more desired applications.

Claims

1. A system, comprising:

a processor coupled to memory storing computer-executable instructions that, when executed by the processor, configure the processor to;

generate a set of contours of a feature in an image in accordance with an image segmentation technique;

select a set of candidate solutions from the set of contours based on secondary information; and

output at least one contour from the set of candidate solutions as a final segmentation of the feature in the image.

2. The system of claim 1, wherein the processor is further configured to:

obtain initialization information; and

generate the set of contours in accordance with the image segmentation technique based on the initialization information.

3. The system of claim 2, wherein the initialization information includes a user selected point of the image within the feature.

4. The system of claim 2, wherein the initialization information includes an initial set of parameters for the image segmentation technique.

5. The system of claim 1, wherein the processor is further configured to receive a user selection of the final segmentation of the feature from the set of candidate solutions.

6. The system of claim 1, wherein the processor is further configured to:

for each candidate solution from the set of candidate solutions, determine a length of the feature, as contoured in the candidate solution, along a predetermined axis; and

select the final segmentation based on respective determined lengths for each candidate solution and a predetermined measured length of the feature.

7. The system of claim 6, wherein the processor is further configured to discard candidate solutions having respective determined lengths which exceed a threshold difference as compared to the predetermined measured length.

8. The system of claim 6, wherein the processor is further configured to select a candidate solution having a determined distance closest to the predetermined measured length as the final segmentation.

9. The system of claim 1, wherein the processor is further configured to refine each contour from the set of candidate solutions.

10. The system of claim 1, wherein the processor is further configured to generate the set of contours in accordance with an active contour segmentation technique.

11. The system of claim 1, wherein the set of contours includes respective results following one or more iterations according to the image segmentation technique.

12. The system of claim 1, wherein the secondary information includes image characteristics.

13. The system of claim 12, wherein the processor is further configured to determine, for each contour, an average distance between each point on the contour and a nearest point of a potential boundary of the feature according to the image characteristics.

14. The system of claim 12, wherein the processor is further configured to determine, for each contour of the set of contours, a mean distance between each point on the contour and a nearest image point corresponding to a local maximum of an image gradient.

15. The system of claim 14, wherein the processor is further configured to select, as candidate solutions, contours that correspond to local extrema of the mean distance when plotted for all contours of the set of contours.

16. The system of claim 15, wherein the processor is further configured to select, as a default candidate solution, a contour having a mean distance which is closest to zero.

17. The system of claim 1, wherein the processor is further configured to generate the set of contours through application of a plurality of image segmentation techniques.

18. The system of claim 1, wherein the processor is further configured to generate the set of contours through application of the image segmentation technique with different sets of parameters.

19. A method for segmenting a medical image, comprising:

generating a set of contours for a feature of an image through application of an image segmentation technique;

applying a secondary criterion to the set of contours to select a set of candidate solutions; and

displaying the contours of the set of candidate solutions to a user for selection of a final segmentation.

20. The method of claim 19, further comprising receiving, from the user, an initialization point of the image for the image segmentation technique, wherein the initialization point is located within the feature of the image.

21. The method of claim 19, further comprising refining a boundary of each contour of the set of candidate solutions based on image properties.

22. The method of claim 19, wherein generating the set of contours comprises:

iterating, according to image segmentation technique, until convergence; and

adding a respective result from each iteration of the image segmentation technique to the set of contours.

23. The method of claim 19, wherein applying the secondary criterion comprises:

determining, for each contour of the set of a contours, a metric specifying a fitness of the contour; and

selecting contours, for inclusion in the set of candidate solutions, having associated metrics satisfying a predetermined condition.

24. The method of claim 23, wherein the metric is a mean distance of each point of the contour to a nearest local maximum of an image gradient associated with the image and the predetermined condition for selection is a mean distance closest to zero within a given subset of similar contours from the set of contours.

25. The method of claim 24, wherein the predetermined condition for selection is a local maximum of the mean distance when plotted versus iteration indices.

26. The method of claim 19, wherein the image segmentation technique is an active contour technique.

27. A computer-readable storage medium having stored thereon computer-executable instructions that configure a processor to:

obtain an initialization point on a medical image as user input, the initialization point being located in a feature of the medical image to be segmented;

generate a set of contours with an active contour technique for the feature of the medical image, wherein each contour in the set corresponds to respective states of an evolving contour of the feature following each iteration of the active contour technique until convergence;

reduce the set of contours to a set of candidate segmentations based on a secondary metric determined for each contour of the set of contours; and

display contours from the set of candidate segmentation to a user for selection of a final segmentation of the feature.

28. The computer-readable storage medium of claim 27, further storing instructions that configure the processor to determine, for each contour of the set of contours, a mean distance between each point of the contour and a nearest local maximum of an image gradient for the medical image.

29. The computer-readable storage medium of claim 28, further storing instructions that configure the processor to identify local maxima on a plot of respective mean distances versus iteration index and to add contours corresponding to the local maxima identified to the set of candidate segmentations.