Basic principles of optical coherence tomography

Figure 2.1

Basic principles of time-domain OCT. Interference of reference and reflected light beams from different layers within the tissue occurs as a function of axial transition of the reference mirror.

As of this date, two different OCT technologies are still available, the older time-domain (TD) and the more recent spectral-domain (SD) OCT, also called Fourier-domain (FD) OCT [10]. The latter name refers to Jean-Baptiste Joseph Fourier (1768–1830), a French Professor at the École Polytechnique, who is known for his work on the Fourier Transform [11].

Time-domain (TD) optical coherence tomography

The basic components and setup of a TD-OCT system are detailed in Figure 2.1. As the name of this technology suggests, TD-OCT is based on the difference in the time delay of the sample light echoes reflected from the different retinal layers as a function of their depth within the tissue (here, the retina) and the reference beam echo, a single echo that is varied by changing the position of the mobile reference mirror (i.e., the reference arm-length).

Interference of light beams with low coherence only occurs in cases where the distance traveled by the light in both arms of the interferometer is equivalent to within the coherence length in order to allow interference by co-occurrence on the level of the photodetector. This phenomenon allows accurate measurement of the echo time delay. Light reflected from superficial structures of the retina or uppermost retinal layers has a shorter echo time delay than light reflected from deeper retinal structures (innermost layers) [2, 5]. The reference mirror is axially translocated in order to match echo time delays from various tissue layers. As the path length of the moving reference mirror is known, it is possible to calculate the depth of the tissue from which the fraction of reflected light arises based on the specific time delay. Based on the amplitude of the interference signals that arise from interference of the reflected light from retinal layers of different depths and the reference light from various path lengths, a single axial scan, the so-called A-scan, is deduced. Key to longitudinal scanning is the fact that the reference mirror can be mechanically mobilized, resulting in a shift of the reference beam, extending the reference path. The reference mirror moves with a specific distance- and time-interval constant, leading to multiple adjacent A-scans, which, in sum, generate the cross-sectional 2-D or longitudinal B-scan of the retina (Figure 2.1). In TD-OCT the cross-sectional image representing the different retinal layers is generated as a function of time delay of the reflected sample light beams. In turn, the time delay depends on the composition and depth of the different layers and the position or axial translation of the reference mirror. These basic principles have led to the name of this technology: time-domain OCT.

TD-OCT has an axial resolution of approximately 10 µm or less (Table 2.1). Image acquisition speed is limited in TD-OCT, however, because the reference mirror needs to be moved.

Table 2.1 Technical characteristics of commercially available OCT devices

Device Fundus image OCT image Manufacturer Web address
Live image Size Optional modes Technology Light source

Scan Speed


Transverse resolution** Axial resolution** Scan depth Min. pupil diameter Max. no. of A-scans/B-scan
SPECTRALIS cSLO 30°, 55°, 165° AF, ICGA, FA, MC SD 870 nm 40000 14 µm 7 µm 1.9 mm 2 mm 1536 Heidelberg Engineering
Stratus IR Cam 26° × 20.5° TD 820 nm 400 20 µm 10 µm 2 mm 3.2 mm 768 Zeiss Meditec AG
CIRRUS 4000 SLO 36° × 30° SD 840 nm 27000 15 µm 5 µm 2 mm 2 mm 4096 Zeiss Meditec AG
3D OCT 2000 IR Cam 45° CF Camera,AF*, FA* SD 840 nm 50000 20 µm 6 µm 2.3 mm 2.5 mm 1024 Topcon
DR 1 SLO 43° SwS 1050 nm 100000 20 µm 8 µm 2.5 mm 1024 Topcon
iVue IR Cam 32° × 23° SD 840 nm 25000 15 µm 5 µm 2.3 mm 3 mm 1024 Optovue
RS-3000 SLO 40°x 30° SD 880 nm 53000 20 µm 7 µm 2.1 mm 2.5 mm Nidek
OCT SLO cSLO 29° SD 830 nm 25000 20 µm 10 µm 2 mm 3 mm Optos
Copernicus HR IR Cam 30° SD 850 nm 52000 12 µm 3 µm 2 mm 3 mm 20000 Optopol
Canon OCT HS-100 SLO 44° × 33° SD 855 nm 70000 20 µm 3 µm 2 mm 3 mm Canon

cSLO confocal Scanning Laser Ophthalmoscope

SLO Scanning Laser Ophthalmoscope

AF Auto Fluorescence

ICGA Indocyanin Green Angiographay

FA Fluorescence Angiography

MC Multicolor

* only 3-D OCT FA Plus

** optical resolution

Spectral-domain (SD) optical coherence tomography

The most recent, so-called fourth-generation or spectral-domain (SD) OCT technology, is based on the mathematical Fourier transform equation. This methodology is, therefore, also known as Fourier-domain technology. The Fourier transform eliminates the need for a movement of the reference beam mirror. SD-OCT replaces the photodetector from TD-OCT with a spectrometer capable of analyzing the full spectrum of interference signals at one time point generated when the sample and the reference beam meet along the same path. It allows the analysis of all frequencies simultaneously. As opposed to TD-OCT, in SD-OCT the interference signal is a function of the different wavelengths and not of the different echo time delays. The whole wavelength spectrum is converted into time delay signals by the Fourier transform. In retinal OCT imaging, this allows the analysis of all echoes from the different retinal layers simultaneously. As a consequence, SD-OCT is much more rapid than its counterpart, while at the same time providing excellent resolution. In TD-OCT, high-resolution imaging can only be achieved at the expense of an increase in acquisition times. Scanning speed is 50–100 times faster with SD-OCT than with TD-OCT [10,11].

Rapid scanning allows larger numbers of B-scans per time interval, and high-speed macular scans with an increasing B-scan density have become possible with the latest devices (see Table 2.1). SD-OCT achieves an axial resolution in the range of 5–6 µm, while digital resolution can be even higher (Table 2.1). High-resolution scans together with real-time averaging significantly increase the signal-to-noise ratio (SNR), ensuring superior image resolution and quality in SD-OCT.

Currently, a number of TD- and SD-OCT devices are commercially available. Technical characteristics of different OCT devices are detailed in Table 2.1.

Only gold members can continue reading. Log In or Register to continue

Mar 31, 2017 | Posted by in NEUROLOGY | Comments Off on Basic principles of optical coherence tomography
Premium Wordpress Themes by UFO Themes