Transcription

COVER STORYThe Evolution ofSpectral-domainOptical CoherenceTomographyBY ALAN C. SULL, BA; LAUREL N. VUONG, BS; VIVEK J. SRINIVASAN, PHD; ANDRE J. WITKIN, MD;MACIEJ WOJTKOWSKI, PHD; JAMES G. FUJIMOTO, PHD; AND JAY S. DUKER, MDCommercial Development of Ophthalmic OCT1991 Demonstration of OCT in vitro1993 First in vivo images of the retina1994 Technology transfer to Humphrey Instruments1995 Clinical studies (glaucoma, AMD, diabeticretinopathy)1996 Commercial ophthalmic OCT instrumentintroduced2000 2nd-generation ophthalmic OCT2 instrumentintroduced2002 3rd-generation Stratus OCT introduced2004 OCT becoming a standard of care2006 6,000 Stratus OCT systems sold2006 Multiple companies introduce OCT instrumentsReprinted from Progr Retin Eye Res, 27, Drexler W, Fujimoto JG, State-of-the-artretinal optical coherence tomography, 44, 2008, with permission from Elsevier.Optical coherence tomography (OCT) is adiagnostic imaging technology that provides cross-sectional images of biologicaltissues. In ophthalmology, OCT can perform “optical biopsy” noninvasively, imaging the retinawith a resolution higher than any other imaging modality other than histology. Now, nearly two decades sinceits introduction, OCT has become indispensable forresearch, screening, diagnosing, and monitoring diseasesof the macula and optic nerve head.Optical coherence tomography was developed byresearchers at the Massachusetts Institute of Technologyand collaborators and was first reported in Science in1991.1 Because ocular media are transparent, the retinaprovided an ideal tissue for OCT imaging. The first in vivostudies of the human retina were published in 1993,2,3 andthese were soon followed by clinical studies performed atNew England Eye Center4 and other medical centers.Although OCT was originally commercialized through astartup company, Advanced Ophthalmic Devices, in 1994the technology was transferred to HumphreyInstruments, a subsidiary of Carl Zeiss (Jena, Germany).5In 1996, the first commercial OCT instrument, the ZeissOCT, was introduced. As imaging speed increased,ergonomics improved, and standardized clinical data andclinical experience became more available, the first-generation device was succeeded by OCT2 and Stratus OCT.Stratus OCT is now accepted as a standard-of-care instru-Figure 1. Summary of major milestones in the commercialdevelopment of ophthalmic OCT.ment in ophthalmology, with more than 6,000 units soldby 2006.6 A summary of major milestones in the development of ophthalmic OCT is depicted in Figure 1.PR INCIPLE S OF O CTOCT imaging is analogous to B-scan ultrasonography,except that OCT measures light rather than acousticMAY/JUNE 2008 I RETINA TODAY I 39

COVER STORYFigure 2. Comparison of schematics for OCT with (A) timedomain detection and (B) SD/FD detection.waves. OCT is performed by measuring the echo delayand intensity of backscattered light from the internaltissue microstructure. Because the echo time delays oflight are too fast to measure directly, an optical correlation technique, known as Michelson low coherenceinterferometry, is used. Low-coherence light from asuperluminescent diode (SLD) is directed through abeam splitter and divided into a sample beam that isfocused onto the patient’s retina and a reference beamthat travels a calibrated delay path (Figure 2A). Lightbackscattered by retinal structures interferes with lightfrom the reference beam, and the interference of theechoes is detected to measure the backscattering signalversus delay or depth.Conventional OCT systems use time-domain (TD)detection, in which the reference mirror is mechanicallyscanned to produce axial scans (A-scans) of the lightechoes vs depth. The optical beam is scanned in thetransverse direction to obtain 2-D cross-sectional images(B-scans) of microstructure. Stratus OCT, the thirdgeneration instrument, uses TD detection and acquires400 A-scans per second.The axial image resolution depends on the bandwidthof the light source. Stratus OCT utilizes a near-infraredsuperluminescent diode light source centered at 820 nm with a 25 nm bandwidth to attain an axial resolution of 8 to 10 µm in the eye. Conversely, transverseresolution is determined by size of the focused optical40 I RETINA TODAY I MAY/JUNE 2008Figure 3. Comparison of normal retina scanned by (A) StratusOCT (512 A-scans taken in 1.3 sec) and (B) prototype SD/FDultrahigh-resolution OCT (8,192 A-scans taken in 0.36 sec).SD/FD OCT shows sharper delineation of intraretinal layers,less motion artifact, and finer speckle. NFL nerve fiber layer,GCL ganglion cell layer, IPL inner plexiform layer, INL inner nuclear layer, OPL outer plexiform layer, ONL outernuclear layer, ELM external limiting membrane, IS/OS inner-outer segment junction, RPE retinal pigment epithelium, CC choriocapillaris/choroids.beam. Due to the tradeoff between spot size and depthof focus, commercial OCT systems typically use 20 to25 µm transverse resolution to achieve adequate depthof focus for retinal imaging.Application of OCT in the ophthalmology clinic hasenabled imaging of the retina with previously unprecedented detail. The restricted speed of OCT with TDdetection, however, limits image quality and retinal coverage. Eye movements introduce motion artifacts thatrequire digital processing, which may obscure important pathologic features. The number of A-scans thatcan be acquired is limited, restricting retinal coverageand increasing sampling errors for detecting focalpathologies.ULTR AHI GH-RE SOLUTI ON O CTIn 1999, ultrahigh-resolution OCT, with an axial resolution of 2 to 3 µm was first demonstrated.8 This resulted from improvements in light-source technology,replacing the SLD with a broad-bandwidth femtosecondtitanium:sapphire laser and later with multiplexedSLDs.9 This advance enabled improved delineation ofintraretinal layers10 (Figure 3), enabling “optical biopsy”of the retina. In vitro studies correlating ultrahigh-resolution OCT images with histology from animal retinas11

Invest Ophthal Vis Sci. Online by Gloesmann, M. Copyright 2003 by InvestigativeOphthalmology & Visual Science. Reproduced with permission of InvestigativeOphthalmology & Visual Science in the format Brochure via Copyright Clearance Center.COVER STORYsition speeds also enable greater retinalcoverage and more precise registration ofOCT images with fundus features. WhileStratus OCT mapped the macula basedon interpolated data from six radial linescans,21 SD/FD OCT enables dense rasterscans composed of closely spaced B-scansacquired as 3-D–OCT data sets. Clinically,this enables physicians to identify focallesions near the fovea in poorly fixatingFigure 4. Correlation of pig retinal layers from (A) ultrahigh-resolution OCTpatients, eccentric macular holes, drusen,with (B) histology.or pockets of fluid in AMD.After acquisition of the 3-D–OCT data(Figure 4) confirmed these findings, and qualitativeset, an OCT fundus image showing landmark features,studies demonstrated new information regarding specif- such as blood vessels and optic disc, may be constructic layer involvement and photoreceptor integrity ined by summing the data in the axial direction, withpathologies such as age-related macular degenerationeach OCT image or B-scan registered to the fundus(AMD), central serous chorioretinopathy, macular hole, image.22 This feature allows physicians to pinpointpathology on the fundus image and simultaneouslyand choroidal neovascularization (CNV).12Despite these advances in resolution, however, sever- examine the lesion in greater detail in the corresponal limitations remained. Although high-performanceding cross-section. In addition, by importing and alignSLD light sources became available, which coulding images from subsequent visits or from other modalachieve ultrahigh-resolution OCT without the need for ities such as fundus photography, indocyanine greenexpensive femtosecond lasers, the limited detectionangiography (ICGA), or fluorescein angiography (FA), itsensitivity of OCT with TD detection limited the image will be possible to longitudinally track changes in retinalacquisition speeds.pathology and enable a more complete understandingof disease mechanisms. Figure 5 depicts longitudinalHI GH-SPEED O CTtracking of neovascular AMD after treatment byAdvances in detection techniques, known as Fourier- ranibizumab (Lucentis, Genentech) injection.domain (FD) or spectral-domain (SD) detection,Three-dimensional OCT also improves the quantificahelped overcome many of the speed limitations of TDtion of thickness and volume. By interpolating less thandetection. SD/FD detection uses an interferometer with TD OCT, greater accuracy in thickness measurementsa high-speed spectrometer (Figure 2B) and measuresmay be attained with SD/FD OCT. 3-D–OCT data maylight echoes from all time delays simultaneously, rather also be used to quantify intraretinal lesions, such asthan sequentially as in TD detection. This enablesdrusen, fluid, retinal detachment, or tumors, to objecincreases of more than 50 times in axial scanningtively assess disease progression. Figure 6 demonstratesspeeds. A similar technique, swept source/FD detecvolumetric segmentation in CNV.tion, also known as optical frequency-domain imaging,Three-dimensional OCT is ideal for application in aniuses a frequency-swept light source and a photodetec- mal models. Ultrahigh resolution is required for imagingtor instead of spectrometer.13-15 Although the principle rodent retinas in order to visualize the major intraretiof SD/FD detection was introduced in 1995,16 it wasnal layers. Although OCT was used in mouse models innot applied for retinal imaging until 2002,17 and its2001,23 limitations in resolution and difficulties in regisspeed and sensitivity advantages were not realized until tration limited its potential. 3-D OCT with SD/FD2003,18-20 after charged-coupled device (CCD) cameradetection allows quantitative measurements at highertechnology matured.resolutions to be performed with precise registration toThe faster acquisition speeds of OCT with SD/FDfundus features,24 enabling longitudinal tracking inrodent models.25 The technology avoids histologicaldetection improve image quality by reducing eyeprocessing artifacts and minimizes the number of sacrimotion artifacts, resulting in a more accurate portrayalficed animals. To date, 3-D–OCT images have been corof the true retinal contour. Furthermore, the improvedrelated with histology in several rodent knockout modtransverse sampling and increased axial resolutionels of retinal degeneration,26 and OCT promises to be aenable more detailed images to be produced withoutpowerful tool for drug discovery and development.having to increase power-exposure levels. Faster acquiMAY/JUNE 2008 I RETINA TODAY I 41

COVER STORYCOMMERCIALSD/FD O CTWith the advent ofbroad-bandwidth, SLDlight sources and recognition of performancegains offered by SD/FDdetection, at least sevenmanufacturers haveintroduced OCT instruments since 2006. Theseinstruments offer similar specifications, withaxial image resolution inthe 4 to 7 µm rangeand acquisition speed inthe 20,000 to 40,000A-scans per secondrange. All models offerthe benefits ofFigure 5. Longitudinal tracking of neovascular AMD. Patient was imaged on Cirrus HD-OCT (A)improved coverage andbefore intravitreal ranibizumab injection and (B) 1-month post-injection. Reduction of perifovealprecise registration overpigmented epithelial detachment and subretinal fluid accumulation can be seen. By registeringOCT with TD detection.scans, cross-sections of the same regions across visits can be compared for changes in focalSome units arepathologies.designed as standaloneOCT (Bioptigen SDOCT [Research Triangle Park, NC], Optopol/ ReichertCopernicus, [Zawierce, Poland], Optovue RTVue-100[Fremont, CA], Carl Zeiss Meditec Cirrus HD-OCT[Dublin, CA]), while others combine OCT withmicroperimetry (Opko/OTI Spectral OCT/SLO[Toronto, Canada]), color fundus photography(Topcon 3D OCT-1000 [Paramus, NJ]), FA, ICGA, autofluorescence, or red-free imaging (Heidelberg SpectralisHRA OCT [Heidelberg, Germany]). Most models offerimportation of images from other instruments fordirect comparison with OCT images.Different products have different ergonomic featuresas well as differences in software. Scan protocols candiffer in type and density, depending on the clinician’spreference for sampling resolution vs the patient’s ability to fixate reliably for a long time. Motion artifactsFigure 6. Volumetric analysis of 3-D–OCT data measuredand registration are dealt with through software,in an eye with choroidal neovascularization. (A) Centralalthough one model (Spectralis HRA OCT) employshorizontal cross-sectional OCT image of a patient with aeye tracking. Two models (Copernicus, RTVue-100)choroidal neovascular membrane secondary to highcurrently offer normative databases for macular thickmyopia. (B) Segmentation of the neovascular lesion acrossness to aid in analysis. Segmentation algorithms usedthe full image set and displayed in 3-D (orange) in relationto define layer boundaries for macular thickness measto the central cross-sectional OCT image. (C) Volumetricurements vary as well. Retinal thickness measurementsrendering of the retina with segmented neovascularon the Stratus OCT were previously defined from thelesion. (D) Volume calculation of the lesion.internal limiting membrane to the inner segment/outersegment junction. In cases of photoreceptor loss,42 I RETINA TODAY I MAY/JUNE 2008

COVER STORYhowever, this approach is likelypostprocessing options.inadequate. 3-D OCT systemsImproved ability to quantitativeare able to resolve photoreceply analyze data, with refinementtor outer segment structure, soof segmentation algorithms, isthe standard for thickness calcuanticipated as clinicians learn tolations may change as compadeal with the massive increase innies adopt either the inner ordata sets. As cost decreases forouter border of retinal pigmentCCD cameras and broad-bandepithelium as the conventionwidth multiplexed superluminesfor the posterior retinal boundcent light sources, ultrahigh-resary. Currently, however, there isolution OCT may become comno consensus of which boundmercially available.aries to use, adding to the disDevelopments in severalparity in thickness measurefields of OCT research promisements between instruments.27exciting advances in performDespite the performance gainsance and clinical applications.in speed and sensitivity, SD/FDFirst, imaging at a 1050-nmOCT still has notable limitations.wavelength enables deeperFirst, SD/FD OCT cannot distinchoroidal penetration andguish between positive and negimprovements in signal levelsative echo delays, which resultsover imaging at 850 nm inin mirror artifacts (Figure 7A).cataract patients.28 Second,improvements in light sourceSecond, the detection sensitivity,technology, such as Fourierdynamic range, and axial imagedomain mode-locking (FDML)resolution vary as a function oflasers, enable swept source/FDdepth. This occurs because ofOCT acquisition at greater thanresolution limitations in the200,000 A-scans per second,spectrometers used in SD/FD10 times faster than SD/FD OCTinstruments and means that theand 500 times faster than TDretina will have different intensiFigure 7. Demonstration of limitations of SD/FDdetection,29 although increasesties if the distance from the eyedetection. Dense raster scans were acquired onin line scan camera detectionto the instrument is changedrates may eventually yield simi(Figure 7C). Finally, because OCT Cirrus HD-OCT in succession in a normal subject.(A) “Mirror”artifact caused by inability to distinlar results. Higher sampling denrelies on an estimate of boundguish between positive and negative echo delays. sities enabled by these acquisiaries of the retina to measure(B) Normal B-scan. (C) Loss of sensitivity as a function speeds should allow newthickness, media opacities, suchscan protocols and even betteras cataracts, may degrade image tion of depth.visualization of intraretinal layquality and make detection ofers. Third, adaptive optics, in which a deformable eleboundaries less reproducible. Different estimates for thement corrects ocular aberrations, improves transverserefractive index of tissue, different segmentation algorithms, and different scan protocols among manufactur- resolution by decreasing spot size of the sample beamon the retina.30,31 Coupled with OCT, isotropic cellularers may also contribute to the systematic variability inthickness measurements. These limitations may be mini- resolution imaging may be achieved,32 which may beuseful for studying photoreceptor morphology inmized through skilled operation and continued refinegreater detail.ment in image data processing.FUTURE DEVELOPMENTSCurrent commercial OCT instruments offer clear performance gains compared with previous OCT technology, although clinical gains are still being assessed.Manufacturers continue to update their software features, such as the addition of normative database andCONCLUSI ONIn the past decade, OCT has established itself as a clinically useful diagnostic modality and a standard for theevaluation of macular and optic nerve diseases. Withfourth-generation SD/FD OCT, clinicians now haveunprecedented advantages in retinal coverage, preciseMAY/JUNE 2008 I RETINA TODAY I 43

COVER STORYregistration of OCT images to fundus features, and 3-Dimaging. More accurate diagnosis, longitudinal monitoring of disease progression and response to treatment, aswell as enhanced understanding of disease pathogenesisand drug development, can be achieved using thenewest OCT technology. These advances promise to further solidify the role of OCT as an integral part of boththe clinical and research settings. This work was supported in part by a Research toPrevent Blindness Challenge grant to the New EnglandEye Center/Department of Ophthalmology, TuftsUniversity School of Medicine, NIH contract RO1EY11289-22, Air Force Office of Scientific ResearchFA9550-07-1-0101 and FA9550-040-1-0011, and NationalScience Foundation BES-0522845.Alan C. Sull, BA, and Laurel N. Vuong, BS, areresearch interns at the New England Eye Center,Tufts Medical Center in Boston. Mr. Sull is amedical student at University of Arkansas forMedical Sciences and Ms. Vuong is a medicalstudent at Tufts University School of Medicine.Vivek J. Srinivasan, PhD, is a graduate student in electrical engineering and computer science at the ResearchLaboratory of Electronics, Massachusetts Institute ofTechnology, Cambridge.Andre Witkin, MD, is a resident physician in ophthalmology at New England Eye Center, Tufts Medical Center.Maciej Wojtkowski, PhD, is Assistant Professor at theInstitute of Physics, Nicolaus Copernicus University,Poland.James G. Fujimoto, PhD, is Professor of ElectricalEngineering and Computer Science in the Department ofElectrical Engineering and Computer Science, ResearchLaboratory of Electronics, Massachusetts Institute ofTechnology. Dr. Fujimoto receives royalties from intellectual property owned by MIT and licensed to Carl ZeissMeditec, Inc., and has stock options in Optovue, Inc.Jay S. Duker, MD, is the Director of the NewEngland Eye Center and Chairman of theDepartment of Ophthalmology, TuftsUniversity School of Medicine in Boston. Dr.Duker is a Retina Today Editorial Board member. He receives research support from Carl Zeiss Meditec,Inc., Optovue, Inc., and Topcon Medical Systems, Inc. Dr.Duker can be contacted by phone: 1 617 636 4677; fax: 1 617 636 4866; or: [email protected] Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science. 1991; 254(5035):1178–1181.2. Swanson EA, Izatt JA, Hee MR, et al. In-vivo retinal imaging by optical coherence tomography. Opt Lett. 1993;18 (21):1864–1866.44 I RETINA TODAY I MAY/JUNE 20083. Fercher AF Hitzenberger CK, Drexler W, Kamp G, Sattmann H. In-Vivo optical coherencetomography. Am J Ophthalmol. 1993;116(1):113–115.4. Puliafito CA, Hee MR, Lin CP, et al. Imaging of macular diseases with optical coherencetomography. Ophthalmology. 1995;102(2):217–229.5. Drexler W, Fujimoto JG. State-of-the-art retinal optical coherence tomography. Prog RetinEye Res. 2008;27:45–88.6. Carl Zeiss Meditec, Inc. 6000th Stratus OCT system installed. Visions. Winter/Spring2007. Available mbedTitelIntern/USStratusOCT/ File/Stratus OCT.pdf; Accessed May 12, 2008.7. Fujimot, JG. Optical coherence tomography for ultrahigh resolution in vivo imaging. NatBiotechnol. 2003;21(11):1361–1367.8. Drexler W, Morgner U, Ghanta RK, KÑrtner FX, Schuman JS, Fujimoto JG. Ultrahigh-resolution ophthalmic optical coherence tomography. Nat Med. 2001;7:50–507.9. Ko TH, Adler DC, Fujimoto JG, et al. Ultrahigh resolution optical coherence tomographyimaging with a broadband superluminescent diode light source. Opt Express.2004;12(10):2112–2119.10. Ko TH, Fujimoto JG, Schuman JS, et al. Comparison of ultrahigh and standard resolutionoptical coherence tomography for imaging macular pathology. Ophthalmology.2005;112:1922–1935.11. Gloesmann M, Hermann B, Schubert C, Sattmann H, Ahnelt PK, Drexler W. Histologiccorrelation of pig retina radial stratification with ultrahigh-resolution optical coherencetomography. Invest Ophthalmol Vis Sci. 2003; 44:1696–1703.12. Drexler W, Sattmann H, Hermann B, et al. Enhanced visualization of macular pathologywith the use of ultrahigh-resolution optical coherence tomography. Arch Ophthalmol.2003;121:695–706.13. Chinn SR, Swanson, EA, Fujimoto JG. Optical coherence tomography using a frequencytunable optical source. Opt Lett. 1997;22(5):340–342.14. Yun SH, Tearney GJ, de Boer JF, Iftimia N, Bourma BE. High-speed optical frequencydomain imaging. Opt Express. 2003;11(22):2953-2963.15. Lee ECW, de Boer JF, Mujat M, Lim H, Yun SH. In vivo optical frequency domain imagingof human retina and choroid. Opt Express. 2006;14(10):4403–4411.16. Fercher AF, Hitzenberger CK, Kamp, G, El-Zaiat SY. Measurement of intraocular distancesby backscattering spectral interfereometry. Opt Commun. 1995;117:43–48.17. Wojtkowski M, Leitgeb R, Kowalcyzk A, Bajraszewski T, Fercher AF. In vivo human retinalimaging by Fourier domain optical coherence tomography. J Biomed Opt.2002;7(3):457–463.18. de Boer JF, Cense B, Park BH, Pierce MC, Tearney GJ, Bouma BE. Improved signal-tonoise ratio in spectral-domain compared with time-domain optical coherence tomography.Opt Lett. 2003;28(21):2067–2069.19. Leitgeb R. Hitzenberger CK, Fercher AF. Performance of fourier domain vs. time domainoptical coherence tomography. Opt Express. 2003;11(8)889–894.20. Choma MA, Sarunic MV, Yang CH, Izatt JA. Sensitivity advantage of swept source andFourier domain optical coherence tomography. Opt Express. 2003;11(18):2183–2189.21. Podoleanu A, Charalambous I, Plesea L, Dogariu A, Rosen R. Correction of distortions inoptical coherence tomography imaging of the eye. Physics of Medicine and Biology.2004;49(7):1277–1294.22. Wojtkowski M, Srinivasan VJ, Fujimoto JG, et al. Three dimensional retinal imaging withhigh-speed, ultrahigh resolution, optical coherence tomography. Ophthalmology.2005;112(10):1734–1746.23. Li Q, Timmers AM, Hunter K, et al. Noninvasive imaging by optical coherence tomography to monitor retinal degeneration in the mouse. Invest Ophthal Vis Sci.2001;42(12):2931–2989.24. Srinivasan VJ, Ko TH, Wojtkowski M, et al. Noninvasive volumetric imaging and morphometry of the rodent retina with high-speed, ultrahigh-resolution optical coherence tomography. Invest Ophthalmol Vis Sci. 2006;47(12):5522–5528.25. Ruggeri M, Wehbe H, Jiao S, et al. In vivo three-dimensional high-resolution imaging ofrodent retina with spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci.2007;48(4):1808–1814.26. Kim KH, Puoris’haag M, Maguluri GN, et al. Monitoring mouse retinal degeneration withhigh-resolution spectral-domain optical coherence tomography. J Vis. 2008;8(1):1–11.27. Leung CK, Cheung CY, Weinreb, RN, et al. Comparison of macular thickness measurements between time domain and spectral domain optical coherence tomography. InvestOphthalmol Vis Sci. 2008; [Epub ahead of print]28 Pova ay, B, Hermann B, Unterhuber A, et al. Three-dimensional optical coherence tomography at 1050 nm versus 800 nm in retinal pathologies: enhanced performance and choroidalpenetration in cataract patients. J Biomed Opt. 2007;12(4):041211.29. Huber R, Adler DC, Srinivasan VJ, Fujimoto JG. Fourier domain mode locking at 1050nm for ultra-high-speed optical coherence tomography of the human retina at 236,000 axialscans per second. Opt Lett. 2007;32(14):2049–2051.30. Fernandez EJ, Povozay B, Hermann B, et al. Three-dimensional adaptive optics ultrahighresolution optical coherence tomography using a liquid crystal spatial light modulator.Vision Res. 2005;45:3432–3444.31. Zawadski R, Jones S, Olivier S, et al. Adaptive-optics optical coherence tomography forhigh-resolution and high-speed 3D retinal in vivo imaging. Opt Express.2005;13:8532–8546.32. Zhang Y, Cense B, Rha J, Jonnal RS, et al. High-speed volumetric imaging of cone photoreceptors with adaptive optics spectral-domain optical coherence tomography. Opt Express.2006;14(10):4380–4394.

(Topcon 3D OCT-1000 [Paramus, NJ]), FA, ICGA, auto-fluorescence, or red-free imaging (Heidelberg Spectralis HRA OCT [Heidelberg, Germany]). Most models offer importation of images from other instruments for direct comparison with OCT images. Different products have different ergonomic featur