OSA's Digital Library

Optics Express

Optics Express

  • Editor: C. Martijn de Sterke
  • Vol. 18, Iss. 20 — Sep. 27, 2010
  • pp: 21293–21307
« Show journal navigation

Automated layer segmentation of macular OCT images using dual-scale gradient information

Qi Yang, Charles A. Reisman, Zhenguo Wang, Yasufumi Fukuma, Masanori Hangai, Nagahisa Yoshimura, Atsuo Tomidokoro, Makoto Araie, Ali S. Raza, Donald C. Hood, and Kinpui Chan  »View Author Affiliations


Optics Express, Vol. 18, Issue 20, pp. 21293-21307 (2010)
http://dx.doi.org/10.1364/OE.18.021293


View Full Text Article

Acrobat PDF (1396 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

A novel automated boundary segmentation algorithm is proposed for fast and reliable quantification of nine intra-retinal boundaries in optical coherence tomography (OCT) images. The algorithm employs a two-step segmentation schema based on gradient information in dual scales, utilizing local and complementary global gradient information simultaneously. A shortest path search is applied to optimize the edge selection. The segmentation algorithm was validated with independent manual segmentation and a reproducibility study. It demonstrates high accuracy and reproducibility in segmenting normal 3D OCT volumes. The execution time is about 16 seconds per volume (480x512x128 voxels). The algorithm shows potential for quantifying images from diseased retinas as well.

© 2010 OSA

1. Introduction

The recent advancement of optical coherence tomography (OCT) from time domain OCT to spectral domain OCT (SD-OCT) has greatly improved imaging speed and detection sensitivity and has thus allowed acquisitions of larger volumes of data [1

1. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]

5

5. W. Drexler and J. G. Fujimoto, “State-of-the-art retinal optical coherence tomography,” Prog. Retin. Eye Res. 27(1), 45–88 (2008). [CrossRef]

]. Retinal 3D OCT volumes are now commonly acquired for clinical diagnosis or investigation. But without accurate and fast quantification of these large volumes of data, it is difficult for clinicians or scientists to quickly diagnose or investigate retinal diseases. Therefore, automatically extracting useful information from OCT images, such as calculating total retinal thickness, has become increasingly significant. Segmentation is a critical step towards reliable quantification. Because of segmentation, total retinal thickness and nerve fiber layer (NFL) thickness have become useful indices to indicate the progress of glaucoma. Recent improvements in OCT technology with respect to axial resolution have made the delineation of more intra-retinal layers possible, which will greatly benefit clinical studies in areas such as glaucoma and diabetic retinopathy [6

6. M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]

9

9. D. Cabrera DeBuc and G. M. Somfai, “Early detection of retinal thickness changes in diabetes using Optical Coherence Tomography,” Med. Sci. Monit. 16(3), MT15–MT21 (2010). [PubMed]

].

2. Methods

2.1 Two-step segmentation algorithm

The Canny edge detector has been recognized as a well-optimized edge detector [24

24. J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Mach. Intell. 8(6), 679–698 (1986). [CrossRef] [PubMed]

]. Here, the Canny edge detector is customized using three thresholds: a low-valued threshold to remove the false edges, a high-valued threshold to highlight the significant edges, and a middle-valued threshold to preserve other potential edges as well. The customized Canny edge output is not binary, but rather consists of three discrete values as determined by the thresholds, i.e. the higher the value, the stronger the edge. The parameters of the Canny edge detector, such as kernel size and thresholds, are catered to the different boundary detections, but are common and set automatically for all subjects. All the parameters were experimentally optimized for general use in segmenting Topcon OCT images. The thresholds are set to constant values that vary by boundary. For high contrast boundaries, such as the ILM and IS/OS, the high-valued threshold is set to 0.85, and the middle-valued and low-valued thresholds are set to 0. For low contrast boundaries, such as the ELM, all three thresholds are set to 0. Other intra-retinal boundaries have the high-valued threshold set to 0.8, the middle-valued threshold set to 0.55, and the low-valued threshold set to 0.1. The kernel size of each Canny edge detector varies and is proportionately adjusted by the scan resolution. In most cases, the Canny edge map can correctly show the discontinuous layer edges being detected. Therefore, the Canny edge map can serve as the primary guide within the shortest path search.

A graph based only on node cost assignments is constructed. The node costs are mainly based on a linear combination of Canny edge detector output, axial intensity gradient values, and other information, such as intensity, for specific boundaries, and are given by
C(i,j)=w1·Canny(i,j)+w2·Axial(i,j)+w3·Others(i,j)
(1)
where C(i, j) is the node cost function, i is the X direction index with maximum number n, j is the Y direction index with maximum number m, and w1, w2, and w3 are weights.

2.2 Nine boundary detection

2.3 Experiment and quantitative analysis

Segmentation accuracy was independently validated by comparing our algorithm results to the results [28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

] from a manual segmentation procedure [6

6. M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]

,27

27. D. C. Hood, C. E. Lin, M. A. Lazow, K. G. Locke, X. Zhang, and D. G. Birch, “Thickness of receptor and post-receptor retinal layers in patients with retinitis pigmentosa measured with frequency-domain optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 50(5), 2328–2336 (2009). [CrossRef]

,28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

]. The data sets included 38 scans obtained from 38 individuals, 19 glaucoma patients and 19 controls. The thirty-eight two-dimensional macular scans of the horizontal meridian were acquired using Topcon 3D OCT-1000 equipment, and consisted of 16 overlapping, averaged B-scans (1024 A-scans and 6 mm in length). The image depth was 1.68 mm and axial resolution was 3.5 μm/pixel. The manual segmentation results were provided by four experienced and well-trained segmenters with the use of a computer-aided manual segmentation procedure [6

6. M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]

,27

27. D. C. Hood, C. E. Lin, M. A. Lazow, K. G. Locke, X. Zhang, and D. G. Birch, “Thickness of receptor and post-receptor retinal layers in patients with retinitis pigmentosa measured with frequency-domain optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 50(5), 2328–2336 (2009). [CrossRef]

]. The manual segmentation procedure by trained individuals has been shown to have low within- and between-segmenter variations [28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

]. The segmenters only segmented 5 of the 9 boundaries. The automated segmentation algorithm was applied to the same data sets without any prior information regarding the manual segmentation results. The unsigned and signed local border position differences of the ILM, NFL/GCL, IPL/INL, INL/OPL and BM/Choroid boundaries were compared as follows [22

22. M. K. Garvin, M. D. Abràmoff, X. Wu, S. R. Russell, T. L. Burns, and M. Sonka, “Automated 3-D intraretinal layer segmentation of macular spectral-domain optical coherence tomography images,” IEEE Trans. Med. Imaging 28(9), 1436–1447 (2009). [CrossRef] [PubMed]

]: algorithm versus the average of the four segmenters; algorithm versus each of the four segmenters; and a pair-wise comparison between the four segmenters. The differences were not calculated in image locations where at least one segmenter reported that a boundary was not visible. At each A-scan location within each image, the local border position difference was calculated as the difference between the border position for the algorithm and the manual segmentation. The signed and unsigned local border differences were then averaged across A-scans and images. The average pair-wise difference between segmenters was calculated as the average of the mean local border position differences of all the possible segmenter pairings among the four segmenters [28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

].

To test repeatability of the segmentation algorithm, vertical macular 3D scans of 43 eyes (each with 3 repetitions) from Topcon’s Normative Database were segmented. The 3D scans were acquired using Topcon 3D OCT-1000 machines. Each 3D scan has 128 frames and each frame covers a 6mm by 6mm macular area (B-scan image size 480x512). The intra-class correlation (ICC), mean of coefficients of variation (CV) and mean of standard deviation (SD) were calculated to test the reproducibility of each detected boundary. For these statistical reproducibility calculations, the data within each group of four adjacent A-lines were averaged down to yield a reduced data set of size 128 by 128. An automated foveal centering algorithm was then applied to the first repetition for each subject. Next, the data for the remaining two repetitions were registered to the first using a proprietary registration algorithm that allowed for optimization of translation and rotation based on total retinal thickness information (thickness from ILM to BM/Choroid). The data sets were then cropped to the center 5mm by 5mm square region, resulting in a final data set size of 107 by 107. The ICC, CV, and SD calculations were then performed across the three repetitions for each subject, and the 43 individual results were averaged together yielding collective means of ICC, CV, and SD.

3. Results

3.1 Segmentation experiment results

Figure 4
Fig. 4 Results of a normal 3D vertical scan volume. NFL, GCC (from ILM to IPL/INL) and total retinal thickness maps (from ILM to OS/RPE) covered 5x5mm2 area, in which three B-scans (the 32nd, 64th and 96th images) of this volume were shown. T: temporal; I: inferior; N: nasal; S: superior.
illustrates a typical segmentation result for a three-dimensional volume of a normal eye and shows the corresponding two-dimensional thickness maps for the NFL, ganglion cell complex (GCC: from ILM to IPL/INL boundary) and total retina. The segmentation algorithm was implemented in Matlab and C. All the data sets were processed by a personal computer (CPU: Core 2 Duo 3 GHz, RAM: 4 GB). It took about 45 seconds in C for the full nine layer detection for each 3D volume in normal segmentation processing mode and 16 seconds in fast segmentation mode. The algorithm was not specifically optimized for multiple threads.

The unsigned and signed border position differences for normal segmentation mode are summarized in Tables 1

Table 1. Unsigned border position differences (mean ± SD in um) of 38 scans using normal segmentation mode

table-icon
View This Table
| View All Tables
and 2

Table 2. Signed border position differences (mean ± SD in um) of 38 scans using normal segmentation mode

table-icon
View This Table
| View All Tables
and the results for fast segmentation mode are listed in Tables 3

Table 3. Unsigned border position differences (mean ± SD in um) of 38 scans using fast segmentation mode

table-icon
View This Table
| View All Tables
and 4

Table 4. Signed border position differences (mean ± SD in um) of 38 scans using fast segmentation mode

table-icon
View This Table
| View All Tables
. Looking at the individual boundaries, the ILM, INL/OPL and BM/Choroid border position differences for the algorithm (Auto) versus manual segmentation were very close to the between-segmenter differences, which were within axial resolution (3.5 µm) range. The NFL/GCL and IPL/INL differences for algorithm versus manual segmentation were larger than the differences between segmenters. One major source of the larger discrepancy for the IPL/INL was in the foveal region. Because all of the tested images spanned the horizontal meridian, this factor affected all scans in our algorithm versus manual segmentation comparison. In particular, the auto-segmented IPL/INL around the fovea tended to be below the manually segmented boundary as shown in Fig. 5
Fig. 5 Comparison between manual segmentation (average of four segmenters’ results, yellow lines) and algorithm segmentation (blue lines). The blue arrows illustrate the commonly seen NFL/GCL discrepancy between human segmenters and the automated algorithm. The red arrows indicate the commonly seen IPL/INL difference between human segmenters and the automated algorithm around the fovea.
. When the fovea area (the central 250 of the 1024 A-scans) was excluded, the unsigned border position difference of the algorithm versus average of the four segmenters for IPL/INL boundary was improved to 3.72 ± 1.03µm from 4.31 ± 1.04µm. The discrepancy in the NFL/GCL measurements arose mainly from temporal side measurements where the NFL is very thin. This thin NFL could be detected by our algorithm, but the detected thicknesses were thicker than what were measured by the human segmenters. When only the nasal side NFL was counted in the accuracy validation, the unsigned border position difference of the algorithm versus the average of the four segmenters was improved to 2.97 ± 0.57µm, which was smaller than the between-segmenter difference. It is notable that the human experts also have difficulty performing consistently within and between segmenters in segmenting thin temporal NFL outer boundaries as well as the intra-retinal layer boundaries in the central fovea area [28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

]. The overall unsigned border position difference (3.39 ± 0.96µm) for the algorithm versus averaged manual segmentation was close to the axial resolution of 3.5µm and slightly larger than the averaged between-segmenter difference (3.28 ± 1.03µm). The results for the fast segmentation mode followed the same trend with slightly larger numbers in unsigned border position differences.

The grand average ICC, CV and SD reproducibility results for both normal and fast segmentation modes are listed in Table 5

Table 5. Reproducibility of 43 normal eyes (3 repetitions)

table-icon
View This Table
| View All Tables
. Overall the ICC of each boundary was above 0.94, the mean coefficient of variation was less than 7.4%, and the mean standard deviation was less than 2.8µm. There was virtually no difference between normal mode reproducibility results and fast mode results. The high reproducibility, including NFL/GCL and IPL/INL boundary detection [29

29. A. Bruce, I. E. Pacey, P. Dharni, A. J. Scally, and B. T. Barrett, “Repeatability and reproducibility of macular thickness measurements using fourier domain optical coherence tomography,” Open Ophthalmol J 3(1), 10–14 (2009). [CrossRef] [PubMed]

32

32. A. Garas, P. Vargha, and G. Holló, “Reproducibility of retinal nerve fiber layer and macular thickness measurement with the RTVue-100 optical coherence tomograph,” Ophthalmology 117(4), 738–746 (2010). [CrossRef] [PubMed]

], indicates the potential for monitoring disease progress in a clinical application.

Figure 6b demonstrates the segmentation robustness in the presence of blood vessel artifacts. Discontinuities caused by blood vessels are commonly seen in macular OCT images, and this represents a significant challenge for automated segmentation and thickness map generation [33

33. D. C. Hood, B. Fortune, S. N. Arthur, D. Xing, J. A. Salant, R. Ritch, and J. M. Liebmann, “Blood vessel contributions to retinal nerve fiber layer thickness profiles measured with optical coherence tomography,” J. Glaucoma 17(7), 519–528 (2008). [CrossRef] [PubMed]

]. Our algorithm incorporates neighboring gradient information to overcome the blood vessel discontinuity issue as illustrated in Fig. 2.

3.2 Segmentation results on diseased macula

Figure 7
Fig. 7 Results of a 3D horizontal scan volume with drusen. The total thickness map (from ILM to OS/RPE) and the RPE complex thickness map (from OS/RPE to BM/Choroid) cover 5x5mm2 area, in which two B-scans (the 20th and 40th images) of this volume are shown. T: temporal; I: inferior; N: nasal.
displays the segmentation result from a horizontal 3D scan volume with drusen. The results show the segmented boundaries (OS/RPE, IS/OS) accurately track the elevation of retinal pigment epithelium, and the drusen area/volume become easy to quantify via the segmented OS/RPE and BM/Choroid boundaries. The thickness maps from the OS/RPE to BM/Choroid and the ILM to OS/RPE clearly show the RPE elevation area. Figure 8
Fig. 8 Results of a 3D vertical scan volume from a glaucoma patient. NFL and total retinal thickness maps (from ILM to OS/RPE) cover 5x5mm2 area, in which two sample B-scans (the 15th and 47th images) from the volume are shown. T: temporal; I: inferior; N: nasal; S: superior.
illustrates the results based on a vertical 3D scan volume for a patient with glaucoma. The thinning of the NFL in the nasal and inferior hemispheres is clearly visible from the NFL thickness map as well as the total thickness map.

4. Discussion and conclusions

In summary, a novel automated segmentation algorithm has been developed and evaluated. The segmentation algorithm is mainly based on dual-scale gradient information that captures both global and local features. As indicated by the reported high accuracy, reproducibility, and fast execution speed, it holds high promise toward practical applications.

Among the nine boundaries segmented in our paper, the GCL/IPL and ELM are not commonly found in literature. Several attempts at quantifying the combined GCL and IPL thickness have shown its potential application in the detection of diseases such as glaucoma and diabetic retinopathy [6

6. M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]

9

9. D. Cabrera DeBuc and G. M. Somfai, “Early detection of retinal thickness changes in diabetes using Optical Coherence Tomography,” Med. Sci. Monit. 16(3), MT15–MT21 (2010). [PubMed]

]. Our results technically demonstrate the feasibility of automated GCL/IPL and ELM segmentation and/or quantification, which may be beneficial in future clinical study. With current OCT single image quality, manual segmentation of the GCL/IPL and ELM boundaries is not yet available. The IS/OS and OS/RPE were not tested for accuracy due to the availability of manual segmentation data [28

28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

]. While visually these two untested boundaries appear to be performing well, they too should be validated against manual segmentation in future work.

The algorithm aims to achieve high speed without degrading accuracy. Ideally, to achieve the highest degree of sensitivity and accuracy, there would be customized Canny edges for each individual boundary detection. However, to reduce processing time in our implementation, the ILM and IS/OS share the information from the same Canny edge detector and axial gradient map, as do the OS/RPE and BM/Choroid, and the NFL/GCL and IPL/INL. The GCL/IPL and INL/OPL can be calculated separately in relatively little time, since the prior information of neighboring boundaries narrows down the detection area, which is an advantage of sequential detection. In this way, about 45 seconds are needed to complete the segmentation of a full-size 3D volume. To further gain speed, the A-scan reduction technique has been implemented, thereby successfully reducing the execution time to 16 seconds per volume without notably degrading the accuracy or reproducibility.

Although the algorithm is primarily based on gradient information, it still has the flexibility to add any other useful information to the final optimization because its post-processing is based on a convenient graph framework. Currently, only node cost is assigned. But link cost, usually assigned as smoothing terms, can also be used. If link cost were to be assigned, the additional smoothing procedure in the current algorithm may no longer be needed. Additional flexibility rests in the possibility for different kernel sizes, Canny thresholds for the various boundaries, and weight factors of different terms in the cost function. Our settings for these parameters were designed experimentally and intuitively based on various OCT images including normal and diseased eyes with different scan types. In other words, the algorithm has been optimized for general use rather than only for the data sets in this work.

In the final processing step for 3D volumes, the layer height information from neighboring B-scans is utilized to additionally smooth the 2D segmentation results. In our approach, the 3D information is integrated after 2D segmentation. Our current algorithm has the advantage of utilizing 3D information without building a complex 3D segmentation model, but it loses the ability to fully utilize the 3D information. In the future, efficiently integrating 3D information prior to 2D segmentation may be beneficial.

So far, this algorithm has been extensively evaluated for macular OCT images of relatively normal scans. More work is needed to evaluate our approach on retinas affected with outer and inner retinal diseases. Our approach can also be extended to other scan modes as well, such as circumpapillary and optic nerve head images.

References and links

1.

D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]

2.

A. F. Fecher, C. K. Hitzenberger, G. Kamp, and S. Y. Elzaizt, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117(1-2), 43–48 (1995). [CrossRef]

3.

M. Wojtkowski, T. Bajraszewski, P. Targowski, and A. Kowalczyk, “Real-time in vivo imaging by high-speed spectral optical coherence tomography,” Opt. Lett. 28(19), 1745–1747 (2003). [CrossRef] [PubMed]

4.

B. Cense, N. Nassif, T. Chen, M. Pierce, S. H. Yun, B. Park, B. Bouma, G. Tearney, and J. de Boer, “Ultrahigh-resolution high-speed retinal imaging using spectral-domain optical coherence tomography,” Opt. Express 12(11), 2435–2447 (2004). [CrossRef] [PubMed]

5.

W. Drexler and J. G. Fujimoto, “State-of-the-art retinal optical coherence tomography,” Prog. Retin. Eye Res. 27(1), 45–88 (2008). [CrossRef]

6.

M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]

7.

O. Tan, V. Chopra, A. T. Lu, J. S. Schuman, H. Ishikawa, G. Wollstein, R. Varma, and D. Huang, “Detection of macular ganglion cell loss in glaucoma by Fourier-domain optical coherence tomography,” Ophthalmology 116(12), 2305.e1–2314.e2, (2009). [CrossRef]

8.

H. W. van Dijk, P. H. Kok, M. Garvin, M. Sonka, J. H. Devries, R. P. Michels, M. E. van Velthoven, R. O. Schlingemann, F. D. Verbraak, and M. D. Abràmoff, “Selective loss of inner retinal layer thickness in type 1 diabetic patients with minimal diabetic retinopathy,” Invest. Ophthalmol. Vis. Sci. 50(7), 3404–3409 (2009). [CrossRef] [PubMed]

9.

D. Cabrera DeBuc and G. M. Somfai, “Early detection of retinal thickness changes in diabetes using Optical Coherence Tomography,” Med. Sci. Monit. 16(3), MT15–MT21 (2010). [PubMed]

10.

H. Ishikawa, D. M. Stein, G. Wollstein, S. Beaton, J. G. Fujimoto, and J. S. Schuman, “Macular segmentation with optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 46(6), 2012–2017 (2005). [CrossRef] [PubMed]

11.

M. Shahidi, Z. Wang, and R. Zelkha, “Quantitative thickness measurement of retinal layers imaged by optical coherence tomography,” Am. J. Ophthalmol. 139(6), 1056–1061 (2005). [CrossRef] [PubMed]

12.

O. Tan, G. Li, A. T. Lu, R. Varma, and D. HuangO. TanG. LiA. T. LuR. VarmaD. HuangAdvanced Imaging for Glaucoma Study Group, “Mapping of macular substructures with optical coherence tomography for glaucoma diagnosis,” Ophthalmology 115(6), 949–956 (2008). [CrossRef]

13.

D. Koozekanani, K. Boyer, and C. Roberts, “Retinal thickness measurements from optical coherence tomography using a Markov boundary model,” IEEE Trans. Med. Imaging 20(9), 900–916 (2001). [CrossRef] [PubMed]

14.

D. Cabrera Fernández, H. M. Salinas, and C. A. Puliafito, “Automated detection of retinal layer structures on optical coherence tomography images,” Opt. Express 13(25), 10200–10216 (2005). [CrossRef] [PubMed]

15.

T. Fabritius, S. Makita, M. Miura, R. Myllylä, and Y. Yasuno, “Automated segmentation of the macula by optical coherence tomography,” Opt. Express 17(18), 15659–15669 (2009). [CrossRef] [PubMed]

16.

M. Mujat, R. Chan, B. Cense, B. Park, C. Joo, T. Akkin, T. Chen, and J. de Boer, “Retinal nerve fiber layer thickness map determined from optical coherence tomography images,” Opt. Express 13(23), 9480–9491 (2005). [CrossRef] [PubMed]

17.

A. Mishra, A. Wong, K. Bizheva, and D. A. Clausi, “Intra-retinal layer segmentation in optical coherence tomography images,” Opt. Express 17(26), 23719–23728 (2009). [CrossRef]

18.

M. K. Garvin, M. D. Abramoff, R. Kardon, S. R. Russell, X. Wu, and M. Sonka, “Intraretinal layer segmentation of macular optical coherence tomography images using optimal 3-D graph search,” IEEE Trans. Med. Imaging 27(10), 1495–1505 (2008). [CrossRef] [PubMed]

19.

V. Kajić, B. Považay, B. Hermann, B. Hofer, D. Marshall, P. L. Rosin, and W. Drexler, “Robust segmentation of intraretinal layers in the normal human fovea using a novel statistical model based on texture and shape analysis,” Opt. Express 18(14), 14730–14744 (2010). [CrossRef] [PubMed]

20.

S. Lu, C. Cheung, J. Liu, J. Lim, C. Leung, and T. Wong, “Automated Layer Segmentation of Optical Coherence Tomography Images,” IEEE Trans. Biomed. Eng. in press. [PubMed]

21.

E. Götzinger, M. Pircher, W. Geitzenauer, C. Ahlers, B. Baumann, S. Michels, U. Schmidt-Erfurth, and C. K. Hitzenberger, “Retinal pigment epithelium segmentation by polarization sensitive optical coherence tomography,” Opt. Express 16(21), 16410–16422 (2008). [CrossRef] [PubMed]

22.

M. K. Garvin, M. D. Abràmoff, X. Wu, S. R. Russell, T. L. Burns, and M. Sonka, “Automated 3-D intraretinal layer segmentation of macular spectral-domain optical coherence tomography images,” IEEE Trans. Med. Imaging 28(9), 1436–1447 (2009). [CrossRef] [PubMed]

23.

G. Quellec, K. Lee, M. Dolejsi, M. K. Garvin, M. D. Abràmoff, and M. Sonka, “Three-dimensional analysis of retinal layer texture: identification of fluid-filled regions in SD-OCT of the macula,” IEEE Trans. Med. Imaging 29(6), 1321–1330 (2010). [CrossRef] [PubMed]

24.

J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Mach. Intell. 8(6), 679–698 (1986). [CrossRef] [PubMed]

25.

M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis, and Machine Vision (Thomson, 2008).

26.

M. Baroni, P. Fortunato, and A. La Torre, “Towards quantitative analysis of retinal features in optical coherence tomography,” Med. Eng. Phys. 29(4), 432–441 (2007). [CrossRef]

27.

D. C. Hood, C. E. Lin, M. A. Lazow, K. G. Locke, X. Zhang, and D. G. Birch, “Thickness of receptor and post-receptor retinal layers in patients with retinitis pigmentosa measured with frequency-domain optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 50(5), 2328–2336 (2009). [CrossRef]

28.

D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]

29.

A. Bruce, I. E. Pacey, P. Dharni, A. J. Scally, and B. T. Barrett, “Repeatability and reproducibility of macular thickness measurements using fourier domain optical coherence tomography,” Open Ophthalmol J 3(1), 10–14 (2009). [CrossRef] [PubMed]

30.

A. Polito, M. Del Borrello, M. Isola, N. Zemella, and F. Bandello, “Repeatability and reproducibility of fast macular thickness mapping with stratus optical coherence tomography,” Arch. Ophthalmol. 123(10), 1330–1337 (2005). [CrossRef] [PubMed]

31.

A. O. González-García, G. Vizzeri, C. Bowd, F. A. Medeiros, L. M. Zangwill, and R. N. Weinreb, “Reproducibility of RTVue retinal nerve fiber layer thickness and optic disc measurements and agreement with Stratus optical coherence tomography measurements,” Am. J. Ophthalmol. 147(6), 1067–1074.1 (2009). [CrossRef] [PubMed]

32.

A. Garas, P. Vargha, and G. Holló, “Reproducibility of retinal nerve fiber layer and macular thickness measurement with the RTVue-100 optical coherence tomograph,” Ophthalmology 117(4), 738–746 (2010). [CrossRef] [PubMed]

33.

D. C. Hood, B. Fortune, S. N. Arthur, D. Xing, J. A. Salant, R. Ritch, and J. M. Liebmann, “Blood vessel contributions to retinal nerve fiber layer thickness profiles measured with optical coherence tomography,” J. Glaucoma 17(7), 519–528 (2008). [CrossRef] [PubMed]

OCIS Codes
(100.0100) Image processing : Image processing
(170.4470) Medical optics and biotechnology : Ophthalmology
(170.4500) Medical optics and biotechnology : Optical coherence tomography

ToC Category:
Medical Optics and Biotechnology

History
Original Manuscript: August 12, 2010
Revised Manuscript: September 9, 2010
Manuscript Accepted: September 13, 2010
Published: September 22, 2010

Virtual Issues
Vol. 5, Iss. 14 Virtual Journal for Biomedical Optics

Citation
Qi Yang, Charles A. Reisman, Zhenguo Wang, Yasufumi Fukuma, Masanori Hangai, Nagahisa Yoshimura, Atsuo Tomidokoro, Makoto Araie, Ali S. Raza, Donald C. Hood, and Kinpui Chan, "Automated layer segmentation of macular OCT images using dual-scale gradient information," Opt. Express 18, 21293-21307 (2010)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-18-20-21293


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]
  2. A. F. Fecher, C. K. Hitzenberger, G. Kamp, and S. Y. Elzaizt, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117(1-2), 43–48 (1995). [CrossRef]
  3. M. Wojtkowski, T. Bajraszewski, P. Targowski, and A. Kowalczyk, “Real-time in vivo imaging by high-speed spectral optical coherence tomography,” Opt. Lett. 28(19), 1745–1747 (2003). [CrossRef] [PubMed]
  4. B. Cense, N. Nassif, T. Chen, M. Pierce, S. H. Yun, B. Park, B. Bouma, G. Tearney, and J. de Boer, “Ultrahigh-resolution high-speed retinal imaging using spectral-domain optical coherence tomography,” Opt. Express 12(11), 2435–2447 (2004). [CrossRef] [PubMed]
  5. W. Drexler and J. G. Fujimoto, “State-of-the-art retinal optical coherence tomography,” Prog. Retin. Eye Res. 27(1), 45–88 (2008). [CrossRef]
  6. M. Wang, D. C. Hood, J. S. Cho, Q. Ghadiali, G. V. De Moraes, X. Zhang, R. Ritch, and J. M. Liebmann, “Measurement of local retinal ganglion cell layer thickness in patients with glaucoma using frequency-domain optical coherence tomography,” Arch. Ophthalmol. 127(7), 875–881 (2009). [CrossRef] [PubMed]
  7. O. Tan, V. Chopra, A. T. Lu, J. S. Schuman, H. Ishikawa, G. Wollstein, R. Varma, and D. Huang, “Detection of macular ganglion cell loss in glaucoma by Fourier-domain optical coherence tomography,” Ophthalmology 116(12), 2305.e1–2314.e2, (2009). [CrossRef]
  8. H. W. van Dijk, P. H. Kok, M. Garvin, M. Sonka, J. H. Devries, R. P. Michels, M. E. van Velthoven, R. O. Schlingemann, F. D. Verbraak, and M. D. Abràmoff, “Selective loss of inner retinal layer thickness in type 1 diabetic patients with minimal diabetic retinopathy,” Invest. Ophthalmol. Vis. Sci. 50(7), 3404–3409 (2009). [CrossRef] [PubMed]
  9. D. Cabrera DeBuc and G. M. Somfai, “Early detection of retinal thickness changes in diabetes using Optical Coherence Tomography,” Med. Sci. Monit. 16(3), MT15–MT21 (2010). [PubMed]
  10. H. Ishikawa, D. M. Stein, G. Wollstein, S. Beaton, J. G. Fujimoto, and J. S. Schuman, “Macular segmentation with optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 46(6), 2012–2017 (2005). [CrossRef] [PubMed]
  11. M. Shahidi, Z. Wang, and R. Zelkha, “Quantitative thickness measurement of retinal layers imaged by optical coherence tomography,” Am. J. Ophthalmol. 139(6), 1056–1061 (2005). [CrossRef] [PubMed]
  12. O. Tan, G. Li, A. T. Lu, R. Varma, D. Huang, and Advanced Imaging for Glaucoma Study Group, “Mapping of macular substructures with optical coherence tomography for glaucoma diagnosis,” Ophthalmology 115(6), 949–956 (2008). [CrossRef]
  13. D. Koozekanani, K. Boyer, and C. Roberts, “Retinal thickness measurements from optical coherence tomography using a Markov boundary model,” IEEE Trans. Med. Imaging 20(9), 900–916 (2001). [CrossRef] [PubMed]
  14. D. Cabrera Fernández, H. M. Salinas, and C. A. Puliafito, “Automated detection of retinal layer structures on optical coherence tomography images,” Opt. Express 13(25), 10200–10216 (2005). [CrossRef] [PubMed]
  15. T. Fabritius, S. Makita, M. Miura, R. Myllylä, and Y. Yasuno, “Automated segmentation of the macula by optical coherence tomography,” Opt. Express 17(18), 15659–15669 (2009). [CrossRef] [PubMed]
  16. M. Mujat, R. Chan, B. Cense, B. Park, C. Joo, T. Akkin, T. Chen, and J. de Boer, “Retinal nerve fiber layer thickness map determined from optical coherence tomography images,” Opt. Express 13(23), 9480–9491 (2005). [CrossRef] [PubMed]
  17. A. Mishra, A. Wong, K. Bizheva, and D. A. Clausi, “Intra-retinal layer segmentation in optical coherence tomography images,” Opt. Express 17(26), 23719–23728 (2009). [CrossRef]
  18. M. K. Garvin, M. D. Abramoff, R. Kardon, S. R. Russell, X. Wu, and M. Sonka, “Intraretinal layer segmentation of macular optical coherence tomography images using optimal 3-D graph search,” IEEE Trans. Med. Imaging 27(10), 1495–1505 (2008). [CrossRef] [PubMed]
  19. V. Kajić, B. Považay, B. Hermann, B. Hofer, D. Marshall, P. L. Rosin, and W. Drexler, “Robust segmentation of intraretinal layers in the normal human fovea using a novel statistical model based on texture and shape analysis,” Opt. Express 18(14), 14730–14744 (2010). [CrossRef] [PubMed]
  20. S. Lu, C. Cheung, J. Liu, J. Lim, C. Leung, and T. Wong, “Automated Layer Segmentation of Optical Coherence Tomography Images,” IEEE Trans. Biomed. Eng. in press. [PubMed]
  21. E. Götzinger, M. Pircher, W. Geitzenauer, C. Ahlers, B. Baumann, S. Michels, U. Schmidt-Erfurth, and C. K. Hitzenberger, “Retinal pigment epithelium segmentation by polarization sensitive optical coherence tomography,” Opt. Express 16(21), 16410–16422 (2008). [CrossRef] [PubMed]
  22. M. K. Garvin, M. D. Abràmoff, X. Wu, S. R. Russell, T. L. Burns, and M. Sonka, “Automated 3-D intraretinal layer segmentation of macular spectral-domain optical coherence tomography images,” IEEE Trans. Med. Imaging 28(9), 1436–1447 (2009). [CrossRef] [PubMed]
  23. G. Quellec, K. Lee, M. Dolejsi, M. K. Garvin, M. D. Abràmoff, and M. Sonka, “Three-dimensional analysis of retinal layer texture: identification of fluid-filled regions in SD-OCT of the macula,” IEEE Trans. Med. Imaging 29(6), 1321–1330 (2010). [CrossRef] [PubMed]
  24. J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Mach. Intell. 8(6), 679–698 (1986). [CrossRef] [PubMed]
  25. M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis, and Machine Vision (Thomson, 2008).
  26. M. Baroni, P. Fortunato, and A. La Torre, “Towards quantitative analysis of retinal features in optical coherence tomography,” Med. Eng. Phys. 29(4), 432–441 (2007). [CrossRef]
  27. D. C. Hood, C. E. Lin, M. A. Lazow, K. G. Locke, X. Zhang, and D. G. Birch, “Thickness of receptor and post-receptor retinal layers in patients with retinitis pigmentosa measured with frequency-domain optical coherence tomography,” Invest. Ophthalmol. Vis. Sci. 50(5), 2328–2336 (2009). [CrossRef]
  28. D. C. Hood, J. Cho, A. S. Raza, B. A. Dale, and W. Min, “Reliability of a computer-aided, manual procedure for segmenting OCT scans,” Optom. Vis. Sci. (to be published). [PubMed]
  29. A. Bruce, I. E. Pacey, P. Dharni, A. J. Scally, and B. T. Barrett, “Repeatability and reproducibility of macular thickness measurements using fourier domain optical coherence tomography,” Open Ophthalmol J 3(1), 10–14 (2009). [CrossRef] [PubMed]
  30. A. Polito, M. Del Borrello, M. Isola, N. Zemella, and F. Bandello, “Repeatability and reproducibility of fast macular thickness mapping with stratus optical coherence tomography,” Arch. Ophthalmol. 123(10), 1330–1337 (2005). [CrossRef] [PubMed]
  31. A. O. González-García, G. Vizzeri, C. Bowd, F. A. Medeiros, L. M. Zangwill, and R. N. Weinreb, “Reproducibility of RTVue retinal nerve fiber layer thickness and optic disc measurements and agreement with Stratus optical coherence tomography measurements,” Am. J. Ophthalmol. 147(6), 1067–1074.1 (2009). [CrossRef] [PubMed]
  32. A. Garas, P. Vargha, and G. Holló, “Reproducibility of retinal nerve fiber layer and macular thickness measurement with the RTVue-100 optical coherence tomograph,” Ophthalmology 117(4), 738–746 (2010). [CrossRef] [PubMed]
  33. D. C. Hood, B. Fortune, S. N. Arthur, D. Xing, J. A. Salant, R. Ritch, and J. M. Liebmann, “Blood vessel contributions to retinal nerve fiber layer thickness profiles measured with optical coherence tomography,” J. Glaucoma 17(7), 519–528 (2008). [CrossRef] [PubMed]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited