Boston—Optical coherence tomography (OCT) is finally coming into its own. The technology has advanced to the point that the resolution is down to micrometers and three-dimensional images of retinal structures can be obtained. These advances are expected to make possible diagnosis and management of retinal diseases before they become clinically apparent.
Based on the principle of optical reflectometry, OCT is akin to ultrasound but uses light rather than sound. The technology was used to obtain the first in vivo retinal images at the Massachusetts Institute of Technology (MIT) in 1993, and in 1994 the first prototype OCT was developed for use in the clinic at the New England Eye Center, both in Boston.
First-generation OCT became commercially available in 1996. Since then the technology has progressed in leaps and bounds, according to Jay S. Duker, MD, who described the advantages and looked at the future of OCT.
Work on ultra high-resolution OCT (UHR-OCT), with a resolution of between 2 and 3 µm, was in progress at MIT and yielded the first images in 2001.
"The major difference between Stratus OCT 3 and UHR-OCT is in the light source, which uses a mode-locked titanium sapphire laser as a low-coherence light source. The laser generates spectral bandwidths of about 150 nm compared with that of Stratus OCT-3 of about 25 nm," Dr. Duker said. He is professor and chairman, New England Eye Center, Tufts University School of Medicine, Boston.
More than 600 patients with various retinal diseases have undergone imaging at the New England Eye Center using UHR-OCT, and the results were compared with the OCT-3 images. Each OCT-3 image consists of 1,024 axial pixels and 512 transverse pixels and the images are acquired in about 1.3 seconds. The UHR-OCT images are composed of 3,000 axial pixels and 600 transverse pixels, for about three times the axial resolution of its predecessor, with an acquisition time of about 4 seconds, Dr. Duker said. Another major difference is that UHR-OCT is slower than OCT-3.
"The results from the imaging study suggest that UHR-OCT can significantly improve the visualization of retinal architectural morphology, especially fine intraretinal structures, such as the external limiting membrane or the photoreceptor segments. This promises to enable a better understanding of disease pathogenesis as well as earlier diagnosis and possibly more sensitive monitoring of disease progression," Dr. Duker said. "We found that UHR-OCT imaging provides more morphologic information than standard-resolution Stratus OCT in a number of diseases, such as age-related macular degeneration, diabetic retinopathy, epiretinal membrane, macular hole, retinitis pigmentosa, and macular dystrophies."
High-speed technology The latest advance in the technology is ultra high-speed UHR-OCT, which is about 40 times faster in acquiring images compared with the conventional OCT system. This high-speed technology is referred to as spectral domain detection because echo time delays of light are detected by measuring the interference spectrum of the light signal from the tissues.
"This instrument is so fast that it can now collect data and produce a video of OCT with a resolution of 2 to 3 µm. No interferometer is needed, there are no moving parts, and as a result motion-correction algorithms are not required. And because all of the reflected light is measured at once, rather than light that returns at a given echo delay, there is a dramatic increase in detection sensitivity. The increased detection sensitivity and data acquisition rate enable ultra high-speed retinal images to be obtained. The resolution is about 2.1 µm, and 16,000 axial scans per second can be obtained," Dr. Duker reported. Another advantage of this technology is that it may eliminate sampling errors that arise from lack of registration of successive OCT scans.
Ultra high-speed UHR-OCT carries with it a few important advantages over the previous generations of OCT. Using ultra high speed, UHR-OCT acquires images with higher transverse pixel densities in a shorter imaging time and provides substantially more information on the morphology of the intraretinal layer than conventional OCT.