OCT—for rapid detection of diseases of the retina

Retinal imaging before OCT

To diagnose and monitor health problems, clinicians often peer inside the body with procedures that expose the underlying pathological processes. In ophthalmology, for instance, surveying details of the retina can help uncover and decipher conditions that threaten sight.

Until the early 1990s, ophthalmologists used clinical ophthalmoscopic examination to diagnose eye disorders, including leading causes of blindness such as age-related macular degeneration (AMD), diabetic retinopathy, and glaucoma, but the instrumentation lacked the fine depth resolution needed to provide valuable structural information. Furthermore, aspects of eye examinations were often subjective, especially for early-stage disease. The intravenous dye injections that were typically used to spot the pernicious blood vessels that leak fluid into the retina in severe AMD and diabetic retinopathy can trigger nausea and vomiting—and, on rare occasions, they even spur a dangerous allergic reaction. Eye pressure, which forms a mainstay of glaucoma detection, fluctuates, so diagnoses could be missed, and visual field testing, another conventional tool for glaucoma assessment, can be unreliable.

Constructive interference

A major inspiration for surmounting these limitations came from a 1971 study of ultrahigh-speed photography by Michael A. Duguay (then at Bell Laboratories, now at Laval University in Québec). He proposed that related methods might eventually facilitate “seeing through the human skin.” In this scenario, light shot into the body would bounce back from structures that it encountered. The farther the target, the longer the time for the echoes to return. In this way, analogous to ultrasound, light could pinpoint object depth within the body. Unlike ultrasound, however, there would be no contact with the instrumentation.

The development of femtosecond optics moved Duguay’s vision forward. In 1986, Fujimoto, an expert in ultrashort lasers, reported that he could use them to distinguish layers within skin and to determine the corneal thickness of anesthetized rabbits. Although this scheme achieved the time resolution necessary to inspect some tissues, it lacked the sensitivity required to detect surfaces that reflect poorly, such as the retina. Furthermore, measurement times were slow and the lasers were prohibitively expensive.

Fujimoto began considering a different tactic, based on a phenomenon called interference that had originally been described in the 1600s. When certain types of light beams meet, they interact to create patterns. This phenomenon results from waves whose amplitudes combine when they are in phase and cancel when they are out of phase. By exploiting this behavior, scientists can measure weak reflections and distances from unknown objects. This approach offered powerful advantages relative to ultrafast lasers in terms of sensitivity, speed, and cost.

Although interferometric principles had been known for hundreds of years, their methodology was not implemented in biological tissues until the 20^th century. In the 1980s, Fujimoto wanted to investigate whether that avenue could be opened further by combining recent technical advances with short coherence-length interferometry. Unlike most lasers, short coherence-length light spans a wide optical spectrum. This property enables resolution of interactions between light waves that have been reflected from the sample of interest and the reference mirror, a key operational element. Interferometry can amplify the feeble signals that ricochet from anatomical substructures, and this attribute facilitates extraction of the desired spatial information. Even electronic clocks are not fast enough to measure the miniscule time delays, but interferometry does the trick.

So Fujimoto assigned MD-PhD student Huang a mission: Replicate the results obtained from the femtosecond lasers using instead interferometry with short coherence-length light. Huang succeeded in measuring corneal thickness in a cow eye, but the process was too slow and inefficient to use on people.

From skies to eyes

Over at MIT’s Lincoln Laboratory, Swanson was working on intersatellite laser communications and high-speed fiber optic networks. Fujimoto recognized that Swanson’s cutting-edge skills and these state-of-the-art technologies might benefit the applications that he and Huang were exploring. In particular, the systems that Swanson was developing harnessed interferometry strategies to track light from distant satellites with precise spatial resolution, which required reliable detection of weak incoming signals. Fujimoto invited Swanson to join his biomedical engineering venture, and Swanson welcomed the opportunity to learn something new and collaborate with talented people.

With Swanson’s resources and expertise, the researchers borrowed concepts and components from optical communications. Crucially, they moved to fiber optics, which helped the light waves to interfere efficiently and bestowed additional advantages. This innovation, married with many others, eventually increased speeds by orders of magnitude and made the equipment more compact and practical.

In the meantime, Huang had realized that the setup could do more than measure tissue thickness. Beyond the retina’s sharp front edge, dozens of peaks emerged. Huang understood that he was differentiating physiologically meaningful layers and, within short order, figured out how to transform the one-dimensional locational information into an image. By panning the beam side-to-side and up-and-down, he could collect many scans that, together, built a cross-sectional, two-dimensional picture of the retinal substrata.

OCT lights the way

In 1991, Fujimoto, Huang, Swanson, and colleagues published a paper about their technique, which they called optical coherence tomography (OCT). Their depictions of a cadaver retina and a coronary artery—with its conspicuous atherosclerotic plaque—achieved resolutions of 12-17 micrometers, about the size of a cell. OCT’s ability to reveal internal details of largely transparent tissue such as the retina was especially impressive.

Crucial elements of the OCT data matched known anatomic hallmarks as well as features obtained by standard staining of the same specimen. Of particular interest, the retinal nerve fiber layer was visible. Structural changes in this region underlie glaucoma, so the results hinted that OCT could provide a way to quantitatively assess progression of that and other ocular diseases.

The researchers took the apparatus through multiple mechanical, electrical, and optical iterations—and they tailored it for human subjects. In 1993, the MIT group, in collaboration with Carmen Puliafito and Joel Schuman at the New England Eye Center (Tufts Medical Center), used OCT to collect retinal images from healthy eyes of live people. In the following years, they scanned thousands of patients and published studies in which they demonstrated the power of OCT and correlated its findings with those of routine methods for uncovering ocular pathologies.

Seeing is believing

Puliafito, Swanson, and Fujimoto established a company that aimed to propel OCT into widespread use. They sold it to Humphrey Zeiss in 1994 and continued to advise on improvements as the instruments gained usability and moved toward mass production. In 1996, after substantial investment and development from Zeiss, the first commercially available OCT device for ophthalmic diagnoses entered the market.

Physicians slowly learned how to wield OCT from papers that retinal specialists published and a 1996 atlas from Fujimoto and his clinical colleagues. In the early 2000s, a new therapy for an advanced form of age-related macular degeneration catalyzed extensive adoption of OCT. This agent, an anti-VEGF antibody (2010 Lasker~Debakey Clinical Medical Research Award), blocks the inappropriate blood-vessel formation that characterizes the condition and can cause blindness. OCT became indispensable for deploying the remedy because it offered an objective way to predict who would benefit. Furthermore, it showed who was responding and who was not—and indicated when follow-up administration was necessary. A 2018 study estimated that individualized, OCT-directed anti-VEGF regimens have saved billions of dollars by avoiding unnecessary treatment.

In the last 30 years, the field has boomed and scientists have continually upgraded the technology (FIGURE 1). Of the 71,000 research articles published in Science since 1900, the original 1991 OCT report is the 13th most referenced, with 10,775 citations in papers from top journals. Increasingly sophisticated light sources and novel techniques have boosted imaging speeds spectacularly, and quality has jumped in concert. Furthermore, two-to-three second scans are now generating images that are 3D rather than 2D.

As suggested in the original OCT study, measuring nerve fiber thickness in the retina affords a way to document progressive thinning and has provided a noninvasive precise, sensitive, and objective way to monitor glaucoma. By catching these changes and others, doctors can now diagnose and treat multiple eye disorders at early stages, thus minimizing vision loss. OCT is a standard ingredient of care; about 30 million OCT ophthalmic procedures are performed annually, approximately one every second.

A bright future

OCT’s use in the eye far surpasses other applications, but specialists and laboratory investigators across the globe are exploring its potential in numerous medical arenas, including cardiology, surgical guidance, gastroenterology, and dermatology, to illuminate our understanding of disease and inform clinical decisions (FIGURE 2). Engineers have combined OCT with catheters and other probes, and cardiologists are exploiting these tools to inspect arteries. The capacity to sightsee within the body presents the possibility of performing optical biopsies rather than removing tissue.

By stitching together optics, telecommunications engineering, and medicine, Fujimoto, Huang, and Swanson translated fundamental research discoveries into a technology that has revolutionized ophthalmological practice and benefitted multitudes of people. Its reach and impact have only begun to grow.

by Evelyn Strauss

Selected Publications – James G. Fujimoto, David Huang, and Eric A. Swanson

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

Swanson E.A., Izatt, J.A., Hee, M.R., Huang, D., Lin, C.P., Shuman, J.S., Puliafito, C.A., and Fujimoto, J.G. (1993). In vivo retinal imaging by optical coherence tomography.  Optics Lett. 18, 1864-1866.

Hee, M.R., Izatt, J.A., Swanson, E.A., Huang, D., Schuman, J.S., Lin, C.P., Puliafito, C.A., and Fujimoto, J.G. (1995). Optical coherence tomography of the human retina. Arch. Ophthalmol. 113, 325-332.

Tearney, G.J., Brezinski, M.E., Bouma, B.E., Boppart, S.A., Pitris, C., Southern, J.F., and Fujimoto, J.G. (1997). In vivo endoscopic optical biopsy with optical coherence tomography. Science. 276 2037-2039.

Jia, Y., Tan, O., Tokayer, J., Potsaid, B.M., Wang, Y., Liu, J.J., Kraus, M.F., Subhash, H., Fujimoto, J.G., Hornegger, J., and Huang, D. (2012). Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt. Express. 20, 4710-4725.

Jia, Y., Bailey, S.T., Hwang, T.S., McClintic, S.M., Gao, S.S., Pennesi, M.E., Flaxel, C.J., Lauer, A.K., Wilson, D.J., Hornegger, J., Fujimoto, J.G., and Huang, D. (2015). Quantitative optical coherence tomography angiography of vascular abnormalities in the living human eye. Proc. Natl. Acad. Sci. USA. 112, E2395-E2402.

Windsor, M.A., Sun, S.J.J., Frick, K.D., Swanson, E.A., Rosenfeld, P.J., and Huang, D. (2018). Estimating public and patient savings from basic research – a study of optical coherence tomography in managing antiangiogenic therapy. Am. J. Ophthalmol. 185, 115-122.