19 Optical Coherence Tomography Diagnosis of Retinal Diseases
Optical coherence tomography (OCT) was developed through a collaborative effort between the New England Eye Center, Tufts University School of Medicine, the Department of Electrical Engineering and Computer Science at Massachusetts Institute of Technology (MIT), and Lincoln Laboratory, MIT. It is a new, noninvasive, noncontact, transpupillary imaging technology for in vivo evaluation of retinal structures and has a resolution of 10 to 17 µm. Similar to B-scan, it produces cross-sectional images of the retina, the difference being that it uses the optical backscattering of light unlike the sound waves, which are used in B-scan. It therefore uses the optical properties, rather than acoustic properties, of the tissues and therefore obtains a much higher axial resolution of approximately 10 µm. The anatomical layers of the retina can be differentiated, and retinal thickness can be measured. It can also be used for anterior-segment imaging to visualize the cornea, iris, lens, and angle. 1 Retinal imaging is performed using infrared light at approximately 800-nm wavelength, whereas for anterior-segment OCT, light of 1300-nm wavelength is used.
The image is displayed using a false color map that corresponds to detected backscattered light levels ranging between 4 × 10–10 to 10–6 of the incident light. Although retinal imaging is its most common application so far, OCT can be applied in a wide variety of fields other than ophthalmology. 2 , 3 , 4 The ability of OCT to perform a nonexcisional optical biopsy in real time while giving detailed qualitative and quantitative information is its main advantage. Compared with ultrasound, its main disadvantage is that light is highly scattered or absorbed by most biological tissues, all the more in case of opaque tissues. Hence, optical imaging is restricted to only the superficial tissues, which are optically accessible. The advantage of OCT is its axial resolution of about 10 µm, which is 10 to 20 times more than standard ultrasound B-mode imaging. 5 Research OCT imaging systems have even higher resolution of up to about 3 µm. 6 The axial resolution of OCT is determined by the physical properties of the light source, whereas transverse resolution is determined by the focused spot size of the optical beam and is generally around 20 to 25 µm. The absolute minimum spot size is limited by the optical aberrations of that particular eye, unlike in other imaging applications. 7 Image resolution also depends on the speed of acquisition, pixels in the image, and the basic resolution of the system.
19.1 Principle of OCT
Sir Isaac Newton first established the technique of low-coherence or white-light interferometry. OCT performs cross-sectional imaging based on low-coherence interferometry by using a continuous beam of low-coherence light. This light is back-reflected from different tissue boundaries, and the machine measures the echo-time delay and intensity of backscattered or back-reflected light from the microstructures inside the tissues. The light is backscattered differently from nonhomogeneous tissues, depending on their optical properties and refractive indices. Serial axial measurements are taken at different transverse positions. These signal intensities are processed by the computer and displayed as grayscale or as a false color-coded image. In grayscale, white corresponds to the strongest backscattered signal and black to the weakest one. Grayscale is not as informative as the false color-coded image because computer monitors have only 8-bit gray resolution, or 256 gray levels. Also, the eye has a limited ability to distinguish between subtle shades of gray. 7 Postprocessing of the image makes it possible to obtain measurements or to reconstruct topography maps. Software programs are available for different scan patterns and different image processing protocols. Table 19.1 describes the parts of the OCT machine.
Light source | Super luminescent diode |
Partially reflecting mirror | Beam splitter splits the incident light beam into two parts and also receives the back-reflected reference optical beam |
Photodetector | Measures the distance the light has traveled to and from the ocular structures. The light pulses will coincide if they are same and produce phenomenon of interference. |
Interferometer | Constructed using fiber-optic coupler, which can precisely measure the echo structure of reflected light and perform high-resolution measurements of the distance and thickness of different structures |
Computer | Signals are processed electronically and displayed on a computer. |
Conventional slit-lamp biomicroscope and a fixed + 78 D condensing lens | Required for retinal examination |
Infrared video camera | Required for viewing and scanning the probe beam on the fundus |
19.2 Color Coding in OCT
A rainbow of colors is used for ophthalmic imaging (Table 19.2). Images are displayed as grayscale/false color scale. Maximum intensity signal (50 dB) is displayed as white in grayscale and red in false color scale. Weakest intensity signal (95 dB) is displayed as black and blue.
Layers of retina | Backscattering | False color in OCT |
Nerve-fiber and plexiform layers | High | Red |
Nuclear layers | Weak | Blue to black |
Ganglion cell layer | Weak | Blue to black |
Outer plexiform layer | High | Red |
Inner plexiform layer | Moderate | Green to yellow |
Boundary between photoreceptor inner segments and outer segments | Thin high | Red |
Retinal pigment epithelium | Strong | Red |
Choriocapillaries | High | Red |
19.3 Interpretation of OCT of the Normal Retina
The light beam of the OCT can be transmitted, absorbed, or scattered, depending on tissue properties. Tissues with high absorption or backscattering, such as hemoglobin and melanin, can cause shadowing of the underlying tissues.
As seen in Fig. 19.1, there is an increase in backscattering at the vitreoretinal interface. The fovea is seen as a thinner area where the inner layers disappear and the photoreceptor layer thickens. Only the outer nuclear layer (ONL) and photoreceptor layer are seen in the fovea. The nerve-fiber layer (NFL) is a highly scattering layer at the inner margin of the retina and is seen as red. It is thickest at the optic disc margins and absent at the fovea. All axonal layers, such as the NFL and the plexiform layers, are more backscattering (inner plexiform layer [IPL]), moderately backscattering (outer plexiform layer [OPL]), or highly backscattering (Henle fiber layer) and hence are seen as red. The nuclear layers (ganglion cell layer [GCL], inner nuclear layer [INL], and outer nuclear layer [ONL]) are poorly backscattering and are seen as blue-black. The GCL increases in thickness in the parafoveal region. The reflection between the inner and outer segments is seen immediately anterior to the retinal pigment epithelium (RPE) as another highly scattering layer as a result of the difference in the refractive index of the inner segment (IS) and the outer segment (OS), which contains rhodopsin. The IS and OS are thicker in the foveal region, which can be seen in the OCT. The external limiting membrane (ELM) may sometimes be seen as a thin backscattering layer behind the ONL. The RPE and choriocapillaris are visualized as the posterior limit of the retina and are highly scattering, seen as red. The RPE, Bruch membrane, and the choriocapillaris cannot be identified separately. The light beam is relatively attenuated on passing through the retinal layers and the choriocapillaris so that structures posterior to this are not seen well because of shadowing.
19.4 Interpretation of OCT of the Optic Nerve Head
The optic disc shows a characteristic contour on OCT (Fig. 19.2). The NFL is thickest near the disc rim, which is composed almost entirely of the NFL. The backscattering decreases as the fibers turn to enter the optic disc because they are no longer perpendicular to the light beam. The photoreceptor layer, RPE, and choriocapillaris terminate at the lamina cribrosa.
19.5 Scanning Protocols
19.5.1 Circumpapillary OCT Scans
Boundary-detection software automatically detects the NFL, and its thickness is measured (Fig. 19.3). The circumpapillary scan is “unwrapped,” and the corresponding quadrants are marked. Normally, the thickest NFL layer is seen in the superotemporal and inferotemporal quadrants. The retinal vessels may be seen as they emerge from the disc as shadowing of the posterior layers.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
