502 NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001

NEW TECHNOLOGY

Here we present new technology for optical coherence to-mography (OCT) that enables ultrahigh-resolution, non-in-vasive in vivo ophthalmologic imaging of retinal and cornealmorphology with an axial resolution of 2–3 µm. This resolu-tion represents a significant advance in performance over the10–15-µm resolution currently available in ophthalmic OCTsystems and, to our knowledge, is the highest resolution forin vivo ophthalmologic imaging achieved to date. This reso-lution enables in vivo visualization of intraretinal and intra-corneal architectural morphology that had previously onlybeen possible with histopathology. We demonstrate imageprocessing and segmentation techniques for automaticidentification and quantification of retinal morphology.Ultrahigh-resolution OCT promises to enhance early diagno-sis and objective measurement for tracking progression of oc-ular diseases, as well as monitoring the efficacy of therapy.

Current clinical practice emphasizes the development oftechniques to diagnose disease in its early stages, whentreatment is most effective and irreversible damage can beprevented or delayed. In ophthalmology, the precise visual-ization of pathology is especially critical for the diagnosisand staging of ocular diseases. Therefore, new imaging tech-niques have been developed to augment conventional fun-doscopy and slit-lamp biomicroscopy. Ultrasonography isroutinely used in ophthalmology, but requires physical con-tact with the eye and has axial resolutions of approximately200 µm (ref. 1). High-frequency ultrasound enables approxi-mately 20 µm axial resolutions, but due to limited penetra-tion, only anterior eye structures can be imaged2. Confocalmicroscopy has been used to image the cornea with sub-mi-crometer transverse resolution3. Scanning laser ophthal-moscopy enables en face fundus imaging with micron-scaletransverse and approximately 300-µm axial resolution4,5.None of these techniques, however, permits high-resolu-tion, cross-sectional imaging of the retina in vivo.

Recently, optical coherence tomography (OCT) hasemerged as a promising new technique for high-resolution,cross-sectional imaging6,7. OCT is attractive for ophthalmicimaging because image resolutions are 1–2 orders of magni-tude higher than conventional ultrasound, imaging can beperformed non-invasively and in real time, and quantitativemorphometric information can be obtained. OCT is some-what analogous to ultrasound imaging except that it useslight instead of sound. High-resolution, cross-sectional im-ages are obtained by measuring the echo time delay of re-flected infrared light using a technique known as lowcoherence interferometry8,9. OCT imaging was first demon-strated in the human retina in vitro6 and in vivo10,11. Recently,it has been extended to a wide range of other non-trans-parent tissues to function as a type of optical biopsy12–15. To

date, however, the most important clinical applications ofOCT have been retinal imaging in ophthalmic diagno-sis7,16–22. Current ophthalmic OCT systems have 10–15-µmaxial resolution and provide more detailed structural infor-mation than any other non-invasive ophthalmic imagingtechnique6,7,10,11. However, the resolution of current clinicalophthalmic OCT technology is significantly below what istheoretically possible. Improving the resolution of OCTophthalmic imaging would enable structural imaging ofretinal pathology at an intraretinal level, as well as improvethe accuracy of morphometric quantification.

The axial resolution of conventional ophthalmic OCT sys-tems is limited to 10–15 µm by the bandwidth of lightsources used for imaging. Short-pulse, solid-state lasers cangenerate ultrabroad bandwidth, low-coherence light13. Abroadband Cr:Forsterite laser operating in the near infrared(1,300 nm) has permitted cellular-level OCT imaging indevelopmental biology specimens with 6-µm axial resolu-tion13. For ophthalmic imaging, light sources operating at800 nm are necessary to avoid absorption in the ocularmedia. Recently, a state-of-the-art broadband Ti:Al2O3 laserhas been developed for ultrahigh (∼ 1 µm) axial resolution,spectroscopic OCT imaging in non-transparent tissue at800-nm center wavelength23,24. We describe here ultrahigh-resolution ophthalmic OCT based on this state-of-the-artoptical technology and demonstrate its potential to provideenhanced structural and quantitative information for oph-thalmologic imaging.

Ultrahigh-resolution OCTUltrahigh-resolution, in vivo ophthalmic imaging with axialresolutions of 2–3 µm was demonstrated in normal humansubjects with a new OCT imaging system using a state-of-the-art Ti:Al2O3 laser25. Using image-processing techniques,we identified and quantitatively measured retinal architec-tural morphology. A comparison of conventional (top) andultrahigh-resolution (middle) OCT images of the retinaalong the papillomacular axis (spanning the fovea and theoptic disc) of a normal human subject is shown in Fig. 1.The axial resolution is approximately 3 µm, as determinedby measuring the full-width half maximum of isolated re-flections from structures in the retina. The axial dimensionof the OCT image is normalized by the index of refraction(1.36) to convert optical delay into geometrical distances.The transverse resolution is governed by the numerical aper-ture of the pupil and is approximately 15 µm in all of theretinal OCT images. Ultrahigh-resolution OCT enables visu-alization of the foveal and optic disc contour, as well as in-ternal architectural morphology of the retina and choroidthat is not resolvable with conventional resolution OCT.

Ultrahigh-resolution ophthalmic optical coherence tomography

WOLFGANG DREXLER1, UWE MORGNER1, RAVI K. GHANTA1, FRANZ X. KÄRTNER1, JOEL S. SCHUMAN2 & JAMES G. FUJIMOTO1

1Department of Electrical Engineering and Computer Science and Research Laboratory of Electronics,Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

2New England Eye Center, Tufts University School of Medicine, Boston, Massachusetts, USACorrespondence should be addressed to J.G.F.; email: jgfuji@mit.edu

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001 503

NEW TECHNOLOGY

The retinal nerve fiber layer (NFL) is clearly differentiated inFig. 1, and the variation of its thickness towards the opticdisc, where the ganglion cell axons penetrate the sclera, canbe observed26. The thickening of the NFL coincides with adecrease in the ganglion cell layer from the fovea to theoptic disc27. The thickness of the NFL, as well as the macularganglion cell layer, may be an important indictor for earlyglaucomatous changes28. Choroidal vasculature in andaround the optic disc can also be resolved.

A comparison of an in vivo cross-sectional, ultrahigh-reso-lution OCT image of a normal human macula, compared toa textbook histological micrograph26, is shown in Fig. 2.These results demonstrate that ultrahigh-resolutionophthalmic OCT enables in vivo visualization of retinal ar-chitectural morphology. The retina is usually divided intoten distinct layers, including four cell layers and two layersof neuronal interconnections26,27,29. Several of these layerscan be resolved by ultrahigh-resolution ophthalmic OCT(Fig. 2). Layers that consist of nerve fibers or plexiform lay-ers are optically backscattering, whereas nuclear layers areweakly backscattering. The internal limiting membrane(ILM) has previously only been visible with electron mi-croscopy and is not resolved with OCT. The obliquely run-ning photoreceptor axons are sometimes considered as aseparate layer in the outer plexiform layer (OPL), known as

Henle’s fiber layer, and is highly backscattering26. Note thatthe junction between the inner and outer segment of thephotoreceptors (IS/OS PR) could be visualized. This structureis not a physical membrane, but an alignment of junctionalcomplexes between Müller cells and photoreceptor cells30.The retinal pigment epithelium (RPE), which containsmelanin, is a very strongly backscattering layer31. Bruch’smembrane, which is only 1–4-µm thick, could not be visual-ized32. The choriocapillaris is vascular and strongly back-scattering.

Ultrahigh-resolution OCT images can be processed usingsegmentation algorithms to identify and quantitativelymeasure intraretinal structure relevant for the early diagno-sis and monitoring of ophthalmologic diseases. Examples ofa segmented OCT image of the fovea (a), the measuredthicknesses of the various retinal layers (b), and a measure-ment of the retinal layers along the papillomacular axis (c)are shown in Fig. 3. Expert segmentation algorithms usingboundary detection17,20,21 were developed to identify retinallayers in the OCT image automatically for rapid measure-ment and interpretation. Errors in the boundary detectionfrom signal dropouts or blood vessels result in sharp discon-tinuities in the boundaries and are corrected using a trans-verse continuity condition. The thickness of the retinalstructures is calculated using the thickness measured from

Fig. 1 Conventional (top) and ultrahigh-resolu-tion (bottom) in vivo OCT images along the papil-lomacular axis. The log of the backscattered lightintensity (from 5 to 45 dB) is represented on afalse-color scale. Axial resolution is 10 µm (top)and 3 µm (bottom). The image sizes are 8.46 ×0.95 mm (transverse × axial, top) and 8.43 × 0.82mm (bottom). The image has been expanded by afactor of two in the axial direction. Conventionaland ultrahigh-resolution OCT images were ac-quired with 100 × 180 (transverse × axial) and 600× 725 pixel sampling, respectively. This corre-sponds to a pixel spacing of 84.6 × 5.2 µm (trans-verse × axial) and 14.1 × 1.1 µm, respectively. Thenerve fiber layer is well differentiated and varies inthickness between the fovea and optic disc.

Fig. 2 Comparison of an in vivo ultrahigh-resolutionOCT image (top) of the normal human macula to ahistologic micrograph of the normal macula (bottom)taken from an ophthalmic textbook26. The OCT imagehas 1:1 aspect ratio to permit comparison to histol-ogy. Several layers can be resolved and have been la-beled in the image: Inner limiting membrane (ILM),nerve fiber layer (NFL), ganglion cell layer (GCL), innerplexiform layer (IPL), inner nuclear layer (INL), junc-tion between the inner and outer segment of the pho-toreceptors (IS/OS PR), outer nuclear layer (ONL),retinal pigment epithelium (RPE). The foveola, foveacentralis, as well as the parafoveal region, are also in-dicated. (Histological micrograph reproduced withpermission from Stereoscopic Atlas of Macular Diseases:Diagnosis and Treatment26.)

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

504 NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001

NEW TECHNOLOGY

the OCT images divided by the group refractive index (1.36)of the retina17,20,21. In the foveal region (Fig. 3a and b), themeasured retinal thickness, that is, the distance from theILM to the RPE, varies from a minimum of 155 µm in the fo-veola to a maximum 275 µm in the nasal perifoveal region.The NFL was segmented only in the nasal portion, but thenumerical values are not plotted as it is extremely thin(mean: 12 µm). Instead the NFL has been included with theganglion cell layer (GCL) and inner plexiform layer (IPL)and the combined thickness of these three layers plotted.The optical contrast between the GCL and IPL layers was in-sufficient to segment them indi-vidually. In the foveola region, theretina consists mainly of the outernuclear and the photoreceptor layer(Fig. 3a and b), in agreement withthe expected retinal morphology26.The GCL and IPL are thickest in theparafoveal region (mean: 100 µm).The mean values of the inner nuclearlayer (INL) and OPL in this area are38 µm and 23 µm, respectively, alsoin agreement with the literature33,34.The mean RPE layer thickness was 9µm (ref. 31). The layers correspond-ing to the choriocapillaris and super-ficial choroidal layers had a meanthickness of 24 µm and 32 µm,respectively. Fig. 3c shows thicknessmeasurements of the retinal NFL,GCL and IPL, as well as retinal thick-ness obtained by segmenting a papil-lomacular OCT image (data notshown). The visualization and mea-

surement of the NFL is especially important for glaucoma di-agnosis. The NFL thickens towards the optic disc, whereasthe GCL and IPL thicknesses decrease. This is consistentwith retinal morphology, which indicates that the GCL istwo or fewer cells in thickness in the extramacular retina,and five or more cells in thickness at the fovea26. Precisequantification of the GCL thickness in the macular regionmight be an indicator for early glaucoma diagnosis28.

In addition to the retina, ultrahigh-resolution OCT canalso image the anterior segment with unprecedented resolu-tion. In this case, the transverse resolution can be improved

Fig. 3 Quantification of intraretinal structures.a, Image-processed, segmented OCT image ofthe fovea. b, Corresponding plot of thickness ofthe retinal layers. c, Plot of thickness of retinallayers along the papillomacular axis. (The seg-mented OCT image of the papillomacular axis isnot shown, but similar to that in Fig. 1.) Thetomogram has been expanded by a factor ofthree in the vertical direction for better visualiza-tion of the retinal substructure. The measuredthicknesses of the segmented layers (except theRPE and choroidal layers) are plotted as a func-tion of transverse position (b and c). Clear visual-ization and quantification of the NFL is especiallyimportant for glaucoma diagnosis (c). NFL thick-ness increases towards the optic disc with a cor-responding reduction in the thickness of the GCLand IPL layer towards the optic disc (see c). Choriocapillaris (CC); choroidal layer (CL); gan-glion cell layer (GCL); inner plexiform layer (IPL);inner nuclear layer (INL); internal limiting mem-brane (ILM); nerve fiber layer (NFL); outer plexi-form layer (OPL); photo receptor layer (PRL);retinal pigment epithelium (RPE); retinal thick-ness (RT).

a

b

c

Fig. 4 In vivo ultrahigh-resolution corneal OCT image of a normal human subject. The imagehas 6 × 2 µm (transverse × axial) resolution. The image size is 1.1 × 0.8 mm (transverse × axial)with 600 × 800 pixels, corresponding to 1.8 × 1 µm pixel spacing. The corneal epithelium,Bowman’s layer, intrastromal morphology and endothelium can be visualized. Boxes show en-largement by a factor of two and correspond to matching shapes in large figure.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001 505

NEW TECHNOLOGY

significantly because the pupil does not limit the numericalaperture for beam focusing. An in vivo ultrahigh-resolutionOCT image of the cornea of a normal human subject withapproximately 2 µm axial and 6 µm transverse resolutions isshown in Fig. 4. The corneal epithelium, Bowman’s layer,intrastromal morphology (for example, corneal lamellae)can clearly be differentiated. Descement’s membrane wasnot resolved, possibly because of insufficient contrast be-tween the endothelium and stroma. The endothelial surfaceof the cornea is also visible, although the endothelial celllayer is approaching the image resolution and is difficult todifferentiate.

DiscussionConventional resolution ophthalmic OCT has enabled thenon-invasive, real-time visualization of retinal pathology,and numerous clinical studies have already been performed.Ultrahigh-resolution ophthalmic OCT achieves a significantimprovement in performance, enabling imaging of internalretinal and corneal structure, which had previously onlybeen possible with histopathology. In combination withimage processing and segmentation techniques, ultrahigh-resolution OCT permits the quantitative measurement of in-ternal retinal architectural morphology, providing a clinicaldiagnostic tool for diseases such as glaucoma, diabeticretinopathy, and macular degeneration—the three leadingcauses of blindness. Retinal and choroidal structures areoften involved early in disease, but are difficult to resolvewith conventional resolution OCT. Ultrahigh-resolutionophthalmologic OCT enables better visualization andquantification of intraretinal structures for more sensitiveand earlier diagnosis of ocular pathologies.

Previous studies have examined the correlation of conven-tional resolution OCT and histology in animals35,36.

Correlation with histology was shown, butwas challenging because of the small retinaldimensions and limited image resolution35.OCT imaging of fixed in vitro human retinaspecimens showed discrepancies with histol-ogy37 that might be explained by changes inoptical properties due to fixation and thehighly directional reflectance of the NFL (ref.38). Because it is not practical to perform one-to-one comparisons of in vivo OCT humanretinal imaging with histology, our study sub-jectively compares OCT images of normalhuman eyes to textbook histology.

Reproducibility of morphometric measure-ments is governed by many factors, for exam-ple, image resolution, optical properties ofretinal layers, variations within the patientpopulation and so forth. However, improvingthe instrumental image resolution is a criticalstep in improving the accuracy and diagnosticusefulness of morphometric imaging. A moresensitive and repeatable quantification of theretinal NFL thickness, as well as the ability toquantify the ganglion cell layer, could enablemore sensitive glaucoma diagnosis, stagingand monitoring. Ultrahigh-resolution OCTshould also increase the sensitivity of stagingdiseases such as macular holes. Early detec-

tion and diagnosis of disease, before the onset of visualloss, is important because early treatment can delay or reduce visual impairment.

Ultrahigh-resolution OCT could be a powerful techniquefor imaging the anterior eye segment. The ability to visualizecorneal microstructure in real time could be useful for mon-itoring and intraoperative control of kerato-refractivesurgery. Ultrahigh-resolution OCT could also be used totrack disease progression in animal models of disease, pro-viding longitudinal data without the need to sacrifice theanimal as in standard biopsy and histopathology.

The ultrahigh-resolution ophthalmologic OCT system re-ported here is based on a prototype instrument. However,the ultimate availability of this ultrahigh-resolution OCTtechnology will depend mainly on the availability of lasersor other sources of ultrabroad-bandwidth light. At present,solid-state lasers are still relatively expensive research instru-ments. With continuing research in laser technology, morecompact and less expensive light sources for OCT imagingcan be expected in the near future.

In summary, the results presented in this study demon-strate ultrahigh-resolution ophthalmic OCT in normalhuman subjects with axial resolutions of 3 µm in the retinaand 2 µm in the cornea. Intraretinal morphology can be vi-sualized and retinal layers can be identified and quantita-tively measured using image-processing algorithms.Ultrahigh-resolution retinal imaging and its ability to meas-ure quantitatively and track changes in morphology shouldcontribute to a better understanding of pathologies and en-able increased sensitivity and specificity for early diagnosisand treatment of ocular diseases. Although further basic andclinical studies are necessary, these results suggest that ultra-high-resolution OCT has the potential to be a powerful re-search and clinical tool in ophthalmology.

Fig. 5 Ultrahigh-resolution ophthalmologic OCT system using a titanium:sapphirelaser light source. The interferometer was dispersion balanced (using H2O water; BK7glass; FS fused silica) and optimized for ultrabroad bandwidths to achieve axial resolu-tions of 2–3 µm. Polarization paddles (PC polarization controllers) were used to controlincident polarization. The system was interfaced to a slit-lamp biomicroscope to permitsimultaneous OCT imaging and en face retinal viewing.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

506 NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001

NEW TECHNOLOGY

AcknowledgmentsWe thank S. Bursell, A. Claremont, E. Ippen, C. Pitris, K. Saunders, E.Swanson, J. MacNutt, V.M. Miller, J. Prien, S. Hel and C. Bang. Supported inpart by the NIH contract RO1-EY11289-13 and EY11006-04, the Air ForceOffice of Scientific Research contract F49620-98-1-0139, the Joint ServicesElectronics Program contract DAAH04-95-1-0038 and the Medical FreeElectron Laser Program N00014-97-1-1066. W.D. was supported by theMax Kade Foundation and the Österreichische Akademie der Wissenschaf-ten. U.M. and F.X.K. were supported by the Deutsche Forschungs-gemeinschaft. R.K.G. was supported from the Howard Hughes MedicalInstitute.

1. Bamber, J.C. & Tristam, M. Diagnostic Ultrasound. in The Physics of MedicalImaging. (ed. Webb, S.) 319–388 (Adam Hilger, Bristol and Philadelphia,1988).

2. Pavlin, C.J., McWhae, J.A., McGowan, H.D. & Foster, F.S. Ultrasound biomi-croscopy of anterior segment tumors. Ophthalmology 99, 1220–1228 (1992).

3. Masters, B.R. & Thaer, A.A. Real-time scanning slit confocal microscopy of thein vivo human cornea. Appl. Opt. 33, 695–701 (1994).

4. Webb, R.H., Hughes, G.W. & Pomerantzeff, O. Flying spot TV ophthalmo-scope. Appl. Opt. 19, 2991–2997 (1980).

5. Bille, J.F., Dreher, A.W. & Zinser, G. Scanning laser tomography of the livinghuman eye. in Noninvasive Diagnostic Techniques in Ophthalmology. (ed.Master, B.R.) 528–547 (Springer Verlag, New York, 1990).

6. Huang, D. et al. Optical coherence tomography. Science 254, 1178–1181(1991).

7. Puliafito, C.A., Hee, M.R., Schuman, J.S. & Fujimoto, J.G. in Optical Coherence

MethodsOptical coherence tomography. OCT performs cross-sectionalimaging of internal tissue microstructure by measuring theecho time delay of backscattered infrared light using aMichelson interferometer and a low coherence lightsource6,12–14. One arm of the interferometer directs the lightonto the sample and a second reference arm has a retroreflect-ing mirror that is precisely, mechanically translated. Interfer-ence between the light from the sample and reference pathsoccurs only when the optical distance traveled by the light inboth paths matches within its coherence length. The interfer-ometer output is detected and demodulated to measure theecho time delay and magnitude of the backscattered light. Byacquiring sequential transverse–axial measurements and re-moving motion artifacts using cross-correlation algorithms11,the resulting data set is a two-dimensional array representingthe optical backscattering in a cross-sectional plane of the tis-sue. This can be digitally processed and displayed as a gray-scale or false-color image. As OCT is based on fiber optics, theophthalmic OCT system (Fig. 5) is easily interfaced to a slit-lamp biomicroscope or fundus camera, which relay images thebeam into the eye. A pair of computer-controlled mirrors scanthe transverse position of the beam on the retina to acquire atomographic image, which is processed and displayed in realtime on a computer. A CCD provides a simultaneous view ofthe fundus and allows the OCT images to be registered withretinal landmarks or pathology.

Axial and transverse image resolution. OCT can achieve ex-tremely high axial resolution. The axial resolution of OCT is deter-mined by the coherence length (inversely proportional to thebandwidth) of the light source used. Superluminescent diodes(SLD) have been used extensively for OCT imaging systems be-cause they are compact and relatively inexpensive. However,axial resolutions are limited to 10–15 µm. Although incandescentlight sources have broad bandwidths and can achieve high reso-lutions, they do not have sufficient intensity in a single spatialmode for high-speed imaging. Ultrashort pulse laser sources areattractive for OCT because they have very broad spectral band-widths and high single-mode power. Ultrahigh-resolution oph-thalmic OCT imaging was performed using a Ti:Al2O3 laser thatgenerates pulses as short as 5.4 femtoseconds, corresponding tobandwidths of up to 350 nm centered at 800 nm (ref. 25). Theseultra-broad bandwidths enable the axial resolution of 2–3 µm, afactor of 4–5 times better than standard, 10–15-µm OCT systemsusing SLDs. In order to achieve ultrahigh-resolution OCT, thefiber-optic interferometer and optical components were de-signed to transmit broad optical bandwidths. Special care mustbe taken to balance optical dispersion in the sample and refer-

ence paths of the interferometer because the axial resolution willbroaden if there is a significant dispersion mismatch between thepaths. Dispersion was balanced by using variable thicknesses offused silica (FS) and BK7 (Fig. 5). The dispersion of ocular mediawas compensated using 25-mm water (Fig. 5). For ophthalmicimaging, chromatic aberration in the eye itself appears to pose afundamental limit to resolution.

The best achievable transverse resolution for OCT is given bythe smallest achievable diffraction-limited spot size on the retina,which is in the best case 10 µm (ref. 39). In our system, the trans-verse spot size is chosen to be approximately 15 µm, to improvethe depth of focus and permit imaging through an undilatedpupil. This spot size is maintained in the retina; however, scatter-ing in deeper structures such as the RPE and choroid can increasethe spot size and degrade transverse resolution40.

Dynamic range and sensitivity. Interferometry has the advan-tage of measuring the field of the reflected optical signal ratherthan the intensity. Thus, extremely high dynamic ranges and sen-sitivities can be achieved. Dual-balanced detection was used tocancel excess amplitude noise that was present in thetitanium:sapphire laser to achieve a nearly quantum-limited sensi-tivity of 95 dB (ref. 11). The sensitivity is defined as the minimumdetectable backscattered power compared to the incident power.

Image acquisition time and laser safety. Image acquisition timedepends upon the number of transverse pixels. Our prototypesystem performs axial scans at a rate of 50 Hz, resulting in imageacquisition times of 4–12 s for a 200–600 transverse pixel image.Higher axial scanning rates of 100–300 Hz should be possiblewith future design modifications to reduce the image acquisitiontime to 1–2 s. As OCT achieves high detection sensitivity, imag-ing can be performed with extremely low light exposures. Inorder to reduce the peak intensity of the femtosecond pulses, thebeam was coupled into a 100-m length of optical fiber, whichdispersively broadened the pulse duration to several hundred pi-coseconds. As the laser pulse repetition rate was 90 MHz, theindividual pulse energies and peak intensities are extremely lowand the integrated output is equivalent to a continuous wave-light exposure41. For a broad bandwidth centered at 800 nm, amaximum permissible exposure of 300 µW with illuminationtime of 27 s for fixed intra-beam viewing is the maximum permis-sible exposure according to the American National StandardsInstitute (ANSI) standard, taking ∼ 700 nm as the shortest wave-length. Assuming a scanning beam with 30 consecutive scans,the maximum permissible exposure is 740 µW. Measurementswere performed on normal, healthy subjects using a 300 µW inci-dent optical power in a scanned beam, well within the ANSI ex-posure limits41.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

NATURE MEDICINE • VOLUME 7 • NUMBER 4 • APRIL 2001 507

NEW TECHNOLOGY

Tomography of Ocular Disease. (Slack, Thorofare, New Jersey, 1995).8. Youngquist, R.C., Carr, S. & Davies, D.E.N. Optical coherence domain reflec-

tometry: A new optical evaluation technique. Opt. Lett. 12, 158–160 (1987). 9. Takada, K., Yokohama, I., Chida, K. & Noda, J. New measurement system for

fault location in optic waveguide devices based on an interferometric tech-nique. App. Opt. 26, 1603–1606 (1987).

10. Fercher, A.F., Hitzenberger, C.K., Drexler, W., Kamp, G. & Sattmann, H. Invivo optical coherence tomography. Am. J. Ophthalmol. 116, 113–114 (1993).

11. Swanson, E.A. et al. In vivo retinal imaging by optical coherence tomography.Opt. Lett. 18, 1864–1866 (1993).

12. Fujimoto, J.G. et al. Optical biopsy and imaging using optical coherence to-mography. Nature Med. 1, 970–972 (1995).

13. Boppart, S.A. et al. In vivo cellular optical coherence tomography imaging.Nature Med. 4, 861–865 (1998).

14. Tearney, G.J. et al. In vivo endoscopic optical biopsy with optical coherencetomography. Science 276, 2037–2039 (1997).

15. Fujimoto, J.G., Pitris, C., Boppart, S. & Brezinski, M. Optical coherence to-mography, an emerging technology for biomedical imaging and opticalbiopsy. Neoplasia 2, 9–25 (2000).

16. Hee, M.R. et al. Optical coherence tomography of the human retina. Arch.Ophthalmol. 113, 325–332 (1995).

17. Hee, M.R. et al. Topography of diabetic macular edema with optical coher-ence tomography. Ophthalmology 105, 360–370 (1998).

18. Hee, M.R. et al. Optical coherence tomography of macular holes.Ophthalmology 102, 748–756 (1995).

19. Hee, M.R. et al. Optical coherence tomography of age-related macular de-generation and choroidal neovascularization. Ophthalmology 103,1260–1270 (1996).

20. Schuman, J.S. et al. Quantification of nerve fiber layer thickness in normal andglaucomatous eyes using optical coherence tomography. Arch. Ophthalmol.113, 586–596 (1995).

21. Schuman, J.S. et al. Reproducibility of nerve fiber layer thickness measure-ments using optical coherence tomography. Ophthalmology 103, 1889–1898(1996).

22. Bowd, C., Weinreb, R.N., Williams, J.M. & Zangwill, L.M. The retinal nervefiber layer thickness in ocular hypertensive, normal and glaucomatous eyeswith optical coherence tomography. Arch. Ophthalmol. 118, 22–26 (2000).

23. Drexler, W. et al. In vivo ultrahigh resolution optical coherence tomography.Opt. Lett. 24, 1221–1223 (1999).

24. Morgner, U. et al. Spectroscopic optical coherence tomography. Opt. Lett.25, 111–113 (2000).

25. Morgner, U. et al. Sub-two-cycle pulses from a Kerr-lens mode-locked Ti:sap-phire laser. Opt. Lett. 24, 411–413 (1999).

26. Gass, J.D.M. Normal macula. in Stereoscopic Atlas of Macular Diseases:Diagnosis and Treatment. vol. 1, 1–49 (Mosby, St. Louis, Missouri, 1997).

27. Hogan, H., Alvarado, J.A. & Wedell, J.E. Histology of the Human Eye: An Atlasand Textbook. 393–522 (W.B. Saunders, Philadelphia, 1971).

28. Zeimer, R., Asrani. S., Zou. S., Quigley. H. & Jampel, H. Quantitative detec-tion of glaucomatous damage at the posterior pole by retinal thickness map-ping. Ophthalmology 105, 224–231 (1998).

29. Krebs, W. & Krebs, I. in Primate Retina and Choroid—Atlas of Fine Structure inMan and Monkey. 4–8 (Springer, New York, 1991).

30. Spalton, D.J., Hitchings, R.A. & Hunter, P.A. Anatomy of the retina. in Atlas ofClinical Ophthalmology 2nd edn., 13.3–13.9 (Mosby, St. Louis, Missouri,1994).

31. Oyster, C.W. in The Human Eye — Structure and Function. (ed. Oyster, C.W.)660–661 (Sinauer Associates, Sunderland, Massachusetts, 1999).

32. Ramrattan, R.S. et al. Morphometric analysis of Bruch’s membrane, the chori-ocappillaris, and the choroid in aging. Invest. Ophthalmol. Vis. Sci. 35,2857–2864 (1994).

33. Rohen, J.W. Morphologie und embryologie des sehorgans. in Augenheilkundein Klinik und Praxis (ed. Francois, J. & Hollwich, F.) 1.17–1.19 (Georg Thieme,Stuttgart, 1977).

34. Marshall, J. & Heckenlively, J.R. Pathologic findings and putative mechanismsin retinitis pigmentosa. in Retinitis Pigmentosa. (ed. Heckenlively, J.R.) 37–67(J.B. Lippincott, Philadelphia, 1988).

35. Toth, C.A. et al. A comparision of retinal morphology viewed by optical co-herence tomography and light microscopy. Arch. Ophthalmol. 115,1425–1428 (1997).

36. Huang, Y. et al. Relation of optical coherence tomography to microanatomyin normal and rd chickens. Invest. Ophthalmol. Vis. Sci. 39, 2405–2416(1998).

37. Chauhan, D.S. & Marshall, J. The interpretation of optical coherence tomog-raphy images of the retina. Invest. Ophthalmol. Vis. Sci. 40, 2332–2341(1999).

38. Knighton, R.W. & Huang, X.R. Optical coherence tomography of retinal nervefiber layer (e-letter). Invest. Ophthalmol. Vis. Sci. 41 (2000).

39. Howland, H.C. & Howland, B. A subjective method for the measurement ofmonochromatic aberrations of the eye. J. Opt. Soc. Am. 67, 1508–1518(1977).

40. Thrane, L., Yura, H.T. & Andersen, P.E. Analysis of optical coherence tomog-raphy systems based on the extended Huygens-Fresnel principle. J. Opt. Soc.Am. A. Opt. Image Sci. Vis. 17, 484–490 (2000).

41. Safe Use of Lasers, ANSI Z 136.1. (American National Standards Institute,New York, 1993).

ON THE MARKET

A new ELISA for the quantitative de-tection of human cytochrome C incell culture lysates, human whole bloodor serum is now available from BenderMedSystems. The ELISA provides re-searchers with a tool for detecting theactivation of apoptotic programs incells. Recent research on serum frompatients with hematological malignan-cies indicates that serum cytochrome C

monitoring might also serve as a clini-cal marker that indicates the onset ofapoptosis and cell turnover in vivo.Tel. (+43) 1-796-40-40-0Fax (+43) 1-796-40-40-400www.bendermedsystems.com

Chemicon has launched a brain-derived neurotrophic factor (BDNF)ELISA, which is designed to provide auseful tool for studying the regulation ofBDNF, including the changes that occurduring brain development and in the in-vestigation and treatment of neurologi-cal diseases. It is formulated with aprimary polyclonal capture antibodyand a detecting, enzyme-conjugatedmonoclonal antibody — a format that issaid to have high efficiency, high sensi-tivity and does not generate false-positive reactions. Precoated strips, pre-pared reagents make the assay easy touse and cost effective.Tel. (+1) 909-676-8080Fax (+1) 909-676-9209www.chemicon.com

PBL Biomedical Laboratories offers anew catalog of interferon productsfor 2001. The 16-page catalog containslistings and product descriptions forover 150 products, including a humaninterferon alpha serum sample ELISAkit and a mouse interferon alpha ELISAkit. Detailed product data sheets areavailable online.Tel. (+1) 732-828-8881Fax (+1) 732-828-3736www.interferonsource.com

ELISAs

Standard curve with the humancytochrome C ELISA.

MAIL-ORDER MEANS

2001 catalog ofinterferon products.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com