Method of unsupervised cell nuclei segmentation

• US 6,681,035 B1
• Filed: 09/27/2000
• Issued: 01/20/2004
• Est. Priority Date: 04/03/1998
• Status: Expired due to Fees

• Current Assignee: Cooperative Research Centre for Sensor Signal and Information Processing, The University of Queensland
• Original Assignee: Cssip
• Inventors: Lovell, Brian, Bamford, Pascal
• Primary Examiner(s): Mariam, Daniel G.
• Assistant Examiner(s): GORADIA, SHEFALI DINESH
• Application Number:
  US09/647,269
• Time in Patent Office: 1,210 Days
• Field of Search: 382/133, 382/173, 382/305, 382/288, 382/218, 382/199, 707/1, 377/10, 358/403
• US Class Current: 382/133
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30012 Spine; Backbone
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06V 10/755 Deformable models or variational models, e.g. snakes or active contours
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

A method of segmentation of cell nuclei is described that uses an active contours approach based on a Viterbi search algorithm. Also disclosed is an initialisation technique that maximises the probability of a correct segmentation. The invention also includes a method to classify cell nuclei according to the difficulty of correct segmentation. Nuclei that are difficult to segment can be rejected to minimise the probability of false positives. The probability of false positives can be driven close to 0%.

Claims

1. An improved method of cell nucleus segmentation from an image containing a cell nucleus including the steps of:

(i) constructing a search space in the image;

(ii) searching the search space with a Viterbi algorithm using a first predetermined value of λ

to find a boundary of the nucleus;

(iii) determining a centroid within the boundary;

(iv) constructing an updated search space in the image centred on the centroid;

(v) searching the updated search space with the Viterbi algorithm using the first predetermined value of λ

to find an updated boundary of the nucleus;

(vi) comparing the update boundary with the boundary;

(vii) repeating steps (iii) to (vi) until the boundary and the update boundary are substantially the same;

(viii) storing an image of the cell nucleus and the found boundary when the update boundary is the same as the boundary;

(ix) searching the search space with a Viterbi algorithm using a second predetermined value of λ

to find an update boundary of the nucleus; and

(x) rejecting the nucleus if the update boundary differs from the boundary by greater than a predetermined amount.

2. The method of claim 1 wherein the search space is constructed from an inner and outer curve, said inner curve being substantially within the nucleus and said outer curve being substantially outside the nucleus.

3. The method of claim 1 wherein the first λ

value is greater than 0.5 but less than 0.9.

4. The method of claim 1 wherein the first λ

value is in the range 0.7 to 0.8.

5. The method of claim 1 wherein the second λ

value is greater than 0.0 but less than 0.5.

6. The method of claim 1 wherein second first λ

value is in the range 0.0 to 0.1.

7. A method of cell nuclei segmentation including the steps of:

(i) locating cell nuclei in a sample;

(ii) initialising the segmentation method by determining a contour for a selected cell nucleus;

(iii) conducting a first segmentation step employing a Viterbi algorithm with a first predetermined value of λ

to obtain a first nucleus boundary;

(iv) conducting a second segmentation step employing a Viterbi algorithm with a second predetermined value of λ

to obtain a second nucleus boundary;

(v) comparing the first and second nucleus boundary; and

(vi) storing an image of the cell nucleus if the first boundary and the second boundary are substantially the same within a predetermined limit.

8. The method of claim 7 wherein the initialisation step further includes the steps of:

utilising the Viterbi algorithm at a predetermined value of λ

to obtain a first result for the nucleus contour and reevaluating the centroid of the nucleus;

recalculating the Viterbi algorithm with the same value of λ

to obtain a further centroid of the nucleus;

comparing the first and further centroids; and

storing a high resolution image of the nucleus if the centroids are substantially the same within predetermined limits.

9. An improved method of cell nucleus segmentation from an image containing a cell nucleus including the steps of:

(i) constructing a search space in the image;

(ii) searching the search space with a Viterbi algorithm using a first predetermined value of λ

to find a boundary of the nucleus;

(iii) determining a centroid within the boundary;

(iv) constructing an updated search space in the image centred on the centroid;

(v) searching the updated search space with the Viterbi algorithm using the first predetermined value of λ

to find an updated boundary of the nucleus;

(vi) comparing the updated boundary with the boundary;

(vii) repeating steps (iii) to (vi) until the boundary and the update boundary are substantially the same; and

(viii) storing an image of the cell nucleus and the found boundary when the update boundary is the same as the boundary.

10. The method of claim 9 wherein the step of constructing a search space includes the further steps of:

locating a region of at least one pixel within the nucleus;

constructing the search space between an inner and an outer curve centred on the region.

11. The method of claim 10 wherein the step of locating a region employs a method of converging squares.

12. The method of claim 10 wherein the region is approximately the minimum size of a nucleus in the image.

13. The method of claim 9 wherein the curves are ellipses.

14. The method of claim 9 including the further steps of:

searching the search space with a Viterbi algorithm using a second predetermined value of λ

to find an update boundary of the nucleus; and

rejecting the nucleus if the update boundary differs from the boundary by greater than a predetermined amount.

15. The method of claim 14 wherein the second predetermined value of λ

is less than the first predetermined value of λ

.