Robust segmentation of retinal pigment epithelium layer

• US 10,460,166 B2
• Filed: 11/04/2015
• Issued: 10/29/2019
• Est. Priority Date: 11/06/2014
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Docherty, Andrew, Pan, Ruimin
• Primary Examiner(s): Bali, Vikkram
• Application Number:
  US14/932,807
• Publication Number:
  US 20160133013A1
• Time in Patent Office: 1,455 Days
• Field of Search: None
• CPC Class Codes: A61B 3/0025 characterised by electronic signal processing, e.g. eye models
• CPC Class Codes: A61B 3/102 for optical coherence tomography [OCT]
• CPC Class Codes: G06T 2207/10101 Optical tomography; Optical coherence tomography [OCT]
• CPC Class Codes: G06T 2207/30041 Eye; Retina; Ophthalmic
• 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: G06V 10/755 Deformable models or variational models, e.g. snakes or active contours
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/197 Matching; Classification

Abstract

A method determining a retinal pigment epithelium identifies a plurality of regions using captured image data of the eye and fits a curve into at least some of the regions. A curve score is determined associated with the fitted curve using at least a distance between the fitted curve and at least some of the regions in which a contribution of the regions to the curve score is biased towards (asymmetric) regions below the fitted curve. These steps are repeated whereupon one of the fitted curves is selected, using the corresponding associated curve score, for classifying some of the regions as forming at least a part of a retinal pigment epithelium.

Claims

1. A computer implemented method for determining a retinal pigment epithelium, the method comprising:

identifying a plurality of regions by calculating a degree of polarization uniformity value using captured image data of an eye, the captured image data being polarization-sensitive OCT data of the eye;

fitting, by a processor, a plurality of curves to at least some of the identified regions, each fitted curve defining at least one first depolarising region at a predetermined distance above the fitted curve and at least one second depolarising region at the predetermined distance below the fitted curve;

for each of the fitted curves, determining, by the processor, a curve score, wherein contribution of a corresponding at least one second depolarizing region is weighted higher than contribution of a corresponding at least one first depolarizing region;

comparing, by the processor, the determined curve scores;

selecting, by the processor, one of the fitted curves using a result of the comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

generating an en face RPE thickness map using the determined position of the healthy retinal pigment epithelium; and

evaluating at least one of size and the number of at least one of diseased regions in the en face RPE thickness map.

2. A method according to claim 1, wherein the captured image data comprises at least one B-scan slice of retina, the B-scan slice comprises intensity image data and polarisation data of an area of an eye.

3. A method according to claim 2, further comprising using the intensity image data and polarisation data to produce degree of polarisation uniformity (DOPU) data.

4. A method according to claim 3, wherein the identifying a plurality of regions comprises determining a plurality of connected components of the DOPU data, so that each depolarising region in the DOPU data is described as at least one connected component.

5. A method according to claim 1, wherein depolarising regions below a curve from the plurality of fitted curves are weighted higher than depolarising regions above the curve in determining a curve score.

6. The method according to claim 1, further comprising:

selecting depolarising regions within a further predetermined distance to the selected fitted curve, wherein the further predetermined distance relates to an expected thickness of the retinal pigment epithelium;

fitting a smooth RPE curve to the selected depolarising regions;

classifying depolarising regions of the captured image data as forming at least one of RPE tissues, choroid tissues, and other depolarising particles using the fitted smooth RPE curve.

7. A method according to claim 6, wherein a depolarising region is classified as being part of the RPE if the depolarising region substantially falls within the predetermined distance to the fitted RPE curve.

8. A method according to claim 7, wherein a depolarising region is classified as choroid if the depolarising region is below the fitted RPE curve at a distance exceeding the predetermined distance.

9. A method according to claim 6, wherein a depolarising region is classified as belonging to other depolarising particles if the depolarising region is above the selected fitted curve at a distance exceeding the further predetermined distance.

10. A method according to claim 6, wherein the depolarising regions are classified using corresponding shape descriptors.

11. A method according to claim 1, wherein the captured image data comprises at least one B-scan slice of retina, the B-scan slice comprises intensity image data and polarisation data of an area of an eye, the method further comprising using the intensity image data and polarisation data to produce degree of polarisation uniformity (DOPU) data, wherein the identifying a plurality of regions comprises determining a plurality of connected components of the DOPU data, so that each depolarising region in the DOPU data is described as at least one connected component.

12. The method according to claim 1, wherein the one of the fitted curves is selected by weighting contribution of the at least one second depolarising region higher than contribution of the at least one first depolarising region.

13. A method according to claim 1, wherein the curve score is determined by asymmetrically biasing toward the second plurality of depolarizing regions.

14. A method according to claim 1,

wherein the curve score associated with the each fitted curve is determined using at least a distance between the each fitted curve and at least some of the identified regions, and

wherein a contribution of the identified regions to the curve score is determined by asymmetrically biasing toward the second plurality of depolarizing regions.

15. The method according to claim 1, further comprising:

selecting a candidate PRE curve using the result of the comparing;

selecting depolarizing regions within a further predetermined distance to the candidate RPE curve, wherein the further predetermined distance relates to an expected thickness of the RPE curve of the eye; and

fitting a smooth curve to the selected depolarizing regions to determine the position of the healthy RPE of the eye.

16. The computer implemented method according to claim 1, further comprising:

classifying depolarizing regions of the captured image data as forming at least one of PRE tissue, choroid tissues, and other depolarizing particles using the position of the healthy retinal pigment epithelium.

17. The computer implemented method according to claim 1, further comprising:

using the en face RPE thickness map to diagnose advanced stage of age-related macular degeneration (AMD) by evaluating the size and the number of geographic atrophy (GA) regions as the diseased regions in the en face RPE thickness map.

18. A non-transitory computer readable storage medium having a program recorded thereon, the program being executable by computerised apparatus to determine a retinal pigment epithelium, the program comprising:

code for identifying a plurality of regions by calculating a degree of polarization uniformity value using captured image data of an eye, the captured image data being polarization-sensitive OCT data of the eye;

code for fitting a plurality of curves to at least some of the identified regions, each fitted curve defining at least one first depolarising region at a predetermined distance above the fitted curve and at least one second depolarising region at the predetermined distance below the fitted curve;

for each of the fitted curves, code for determining a curve score, wherein contribution of a corresponding at least one second depolarizing region is weighted higher than contribution of a corresponding at least one first depolarizing region;

code for comparing the determined curve scores;

code for selecting one of fitted curves using a result of comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

code for generating an en face RPE thickness map using the determined position of the healthy retinal pigment epithelium; and

code for evaluating at least one of size and the number of at least one of diseased regions in the en face RPE thickness map.

19. The non-transitory computer readable storage medium according to claim 18, the program further comprising:

code for classifying depolarizing regions of the captured image data as forming at least one of RPE tissues, choroid tissues, and other depolarizing particles using the position of the healthy retinal pigment epithelium.

20. The non-transitory computer readable storage medium according to claim 18, the program further comprising:

code for using the en face RPE thickness map to diagnose advanced stage of age-related macular degeneration (AMD) by evaluating the size and the number of geographic atrophy (GA) regions as the diseased regions in the en face RPE thickness map.

22. The system according to claim 19, wherein:

the steps (A) (A)(i) fitting, by the processor, a curve into at least some of the identified regions;

(A)(ii) determining, by the processor, a curve score associated with the fitted curve, wherein contribution of regions below the fitted curve is weighted higher than contribution of corresponding regions above the fitted curve;

(A)(iii) repeating steps (A)(i) and (A)(ii) at least once;

(A)(iv) comparing, by the processor, the determined curve scores; and

(A)(v) selecting, by the processor, one of the fitted curves, using a result of the comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

further comprise;

(A)(vi) classifying depolarizing regions of the captured image data as forming at least one of RPE tissues, choroid tissues, and other depolarizing particles using the position of the healthy retinal pigment epithelium (RPE); and

the steps (B) (B)(i) fitting, by the processor, a plurality of curves to at least some of the identified regions, each fitted curve defining at least one first depolarising region at a predetermined distance above the fitted curve and at least one second depolarising region at the predetermined distance below the fitted curve;

(B)(ii) for each of the fitted curves, determining, by the processor, a curve score, wherein the contribution of a corresponding at least one second depolarizing region is weighted higher than contribution of a corresponding at least one first depolarizing region;

(B)(iii) comparing, by the processor, the determined curve scores; and

(B)(iv) selecting, by the processor, one of the fitted curves using a result of the comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

further comprise;

(B)(v) classifying depolarizing regions of the captured image data as forming at least one of RPE tissues, choroid tissues, and other depolarizing particles using the position of the healthy retinal pigment epithelium.

23. The system according to claim 19, the program further comprising:

code for using the en face RPE thickness map to diagnose advanced stage of age-related macular degeneration (AMD) by evaluating the size and the number of geographic atrophy (GA) regions as the diseased regions in the en face RPE thickness map.

21. A system for determining retinal pigment epithelium (RPE), the system comprising:

a source of optical coherence tomography (OCT) images representing captured image data of an eye;

a computer system having a processor coupled to a memory, the memory storing a program for execution by the processor to determine the RPE, the program comprising;

code for identifying a plurality of regions by calculating a degree of polarization uniformity value using the captured image data of the eye, the captured image data being polarization-sensitive OCTdata of the eye;

code for one of;

(A) (A)(i) fitting, by the processor, a curve into at least some of the identified regions;

(A)(ii) determining, by the processor, a curve score associated with the fitted curve, wherein a contribution of regions below the fitted curve is weighted higher than contribution of corresponding regions above the fitted curve;

(A)(iii) repeating steps (A)(i) and (A)(ii) at least once;

(A)(iv) comparing, by the processor, the determined scores;

(A)(v) selecting, by the processor, one of the fitted curves, using a result of the comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

(B) (B)(i) fitting, by the processor, a plurality of curves to at least some of the identified regions, each fitted curve defining at least one first depolarising region at a predetermined distance above the fitted curve and at least one second depolarising region at the predetermined distance below the fitted curve;

(B) (ii) for each of the fitted curves, determining, by the processor, a curve score, wherein the contribution of a corresponding at least one second depolarizing region is weighted higher than contribution of a corresponding at least one first depolarizing region;

(B)(iii) comparing, by the processor, the determined score curve; and

(B)(iv) selecting, by the processor, one of the fitted curves using a result of the comparing to determine a position of a healthy retinal pigment epithelium (RPE) of the eye;

code for generating an en face RPE thickness map using the determined position of the healthy retinal pigment epithelium; and

code for evaluating at least one of size and the number of at least one of diseased regions in the en face RPE thickness map.