In the last 25 years, Optical Coherence Tomography (OCT) has advanced to be one of the most innovative and most successful optical imaging techniques, reaching very high performance required for challenging applications like ophthalmic medicine, dermatology, angiography, and material analysis.
OCT principle and domains
The OCT technique is classically based on the principle of a Michelson interferometer, with a broadband light source as an input. The interferometric signature of this low coherence signal obtained with a Fourier Transform is a wave packet, with a coherence length inversely proportional to the spectral linewidth of the source.
Figure 1. Low Coherence Interferometry: an 850 nm SLED broadband source with a 50 nm FWHM spectral linewidth produces a wave packet (“interferogram”) with a 14 µm coherence length.
Generating successive Optical Path Differences (OPD) of this wave packet, the Time Domain OCT establishes the reflectivity profile of the sample. The Fourier Domain OCT exploits the Fourier transform properties to directly convert the spectral modulations in depth information, which greatly improves the speed and the imaging quality. Direct Spatial OCT information can also be obtained without scanning a reference mirror, with static detection of interference pattern of a standing wave.
Figure 2. Schematic of different OCT modalities: OCT systems can be classified into Time Domain (TD) and Fourier Domain (FD) systems (with grating spectrometer or swept source based systems); we introduce a Direct Spatial Domain OCT without scanning mirror, like the SWIFTS technology. Figure adapted from J. Biomed. Opt. publication  of Professors Wolfgang Drexler and Rainer A. Leitgeb (Medical University of Vienna, Austria).
High performance OCT
The main performance characteristics associated with the OCT technique are:
- the axial resolution of the reflectivity profile; current FD-OCT systems reach micrometric resolutions
- the imaging depth, depending on the spectral resolution of the spectrometer
- the detection sensitivity, defined as the inverse of the minimum reflectivity of a sample detected with a signal to noise equal to 1 ; the best current systems reach more than 100 dB sensitivities
- the detection speed, that defines the number of profiles achievable per second, for a given power; current high-end systems operate over 100 kHz.
High-depth OCT over 1 cm
High-depth spatial information can be very useful for ophthalmology of human and animal fundus eyes. It is also applicable and increasingly needed in industrial applications, such as nondestructive testing (NDT), material thickness measurements (silicon and semiconductors wafers) or surface roughness characterization.
Because of the spectral resolution of the current grating spectrometers, the imaging depth is often limited to depths of the order of 1 to 2 cm maximum.
New direct spatial techniques like the SWIFTS technology can provide deeper information up to 43 mm, thanks to its static detection of high optical path differences . Even if this improved performance is currently obtained with less sensitivity and axial resolution, the solution has good potential for high-depth imaging.
The principle has already been successfully tested on the fast detection of high-OPD signal of typical Michelson setups.
Figure 3. Monitoring of a High Optical Path Differences of a Michelson setup, with a ZOOM Spectra spectrometer (SWIFTS technology) detecting discrete OPD positions up to 43 mm, with 20 µm spatial resolution.
The latest developments will soon lead to micrometric spatial resolution, and even higher depth OCT information up to 60 mm.
 Drexler W, Liu M, Kumar A, Kamali T, Unterhuber A, Leitgeb RA; “Optical coherence tomography today: speed, contrast, and multimodality.” J. Biomed. Opt. 0001;19(7):071412. doi:10.1117/1.JBO.19.7.071412.
 C. Bonneville, F. Thomas, M. de Mengin Poirier, E. Le Coarer, P. Benech, et al., “SWIFTS: a groundbreaking integrated technology for high-performance spectroscopy and optical sensors “, Proc. SPIE 8616, MOEMS and Miniaturized Systems XII, 86160M (March 13, 2013). doi:10.1117/12.2000451