Optical Coherence Tomography (OCT)

Paul Riedmann, winter semester 2021/2022

Optical Coherence Tomography (OCT) is a non-invasive and non-destructive imaging technique using electromagnetic waves that provides accurate cross-sectional views (2D or 3D) of surface profiles or subsurface micro-structures of a specimen in real time ^[1,2].

Principle of the Measurement

General Setup

The OCT measurement bases on interferometric imaging using electromagnetic waves in the range of infrared wavelength ^[1]. It can provide a high lateral resolution imaging on the order of a few micrometers, to depths up to few millimeters in a scattering media. The imaging technique of OCT has no need for a coupling medium or contact with the specimen. It easily measures small or thin parts, delicate samples, and rough surfaces, making it ideal for non-destructive testing ^[2]. ^[3]

OCT detects back-scattered light from different regions of a sample, which were exposed to a broadband light source of short temporal coherence beforehand ^[1], to generate a 3D image. Therefore, the information in the axial direction (along the optical beam, $//<![CDATA[ \begin{array}{l}z\end{array} //]]>$-axis) and the transverse direction (plane perpendicular to the optical beam, $//<![CDATA[ \begin{array}{l}xy\end{array} //]]>$-plane) are obtained. In the axial direction the time delay of light reflected from structures or layers in the sample (similar to ultrasound testing, but with light being used instead of sound) defines the gathered information. The relative movement between the specimen and the OCT system in the $//<![CDATA[ \begin{array}{l}xy\end{array} //]]>$-plane allows for a 2D or 3D image. Given the high speed of light and the resulting challenge in measuring the delay induced by the segments within the specimen, OCT systems indirectly measure the time delay using low-coherence interferometry. ^[3]

There are three general requirements of light sources for OCT imaging according to Schmitt ^[1]:

1. Emission in the near infrared
2. Short coherence length
3. High irradiance

The first requirement is due to the ability of infrared light to penetrate into tissue adequately. The deepest penetration can be achieved by using wavelengths between 1200 nm and 1800 nm. In addition, the optimum wavelength of a source is influenced by other factors like backscatter contrast and optical absorption determining the contrast of OCT images. ^[1]

Short coherence length, the second requirement, sets a relationship between the temporal coherence function of the light source and the width of the axial point-spread function of an OCT scanner. Generally, with broader emission bandwidth of the source, a better resolution and contrast can be achieved. Also, the shape of the spectrum of the source is an important variable affecting the dynamic range of the scanner near strong reflections. ^[1]

The last requirement of high irradiance is needed imaging weakly backscattering structures deep below the tissue surface and is based on wide dynamic range and high detection sensitivity. ^[1]

Testing Methods

Time Domain OCT (TD-OCT)

Figure 1 shows the setup of an OCT system. The configuration is identical to the Michelson interferometer with its roots lying on white-light interferometry ^[1]. The light emitted from a source (1) is split by the beamsplitter (2) into two paths called the reference (3) and sample arms (4) of the interferometer. The light from each arm is reflected back and combines at the detector (5) after passing the beamsplitter again. An interference effect (modulations in intensity) can be detected only if the time traveled by light in the two arms are nearly equal. Thus, the presence of interference serves as a relative measure of distance traveled by the light. ^[3]

The reference arm then is scanned in a controlled manner by varying the distance between the mirror (6) and the beamsplitter. The resulting light intensity is recorded on the detector and a rapid modulation interference pattern occurs when the distance between beamsplitter and mirror is nearly identical to the distance between the beamsplitter and the scattering structures in the sample (7). Although the light beam passes through different structures in the sample, the amount of reflection from each unique structure in the path of the beam is distinguishable. In doing the so, the material scattering, and hence its structure, can be measured as a function of depth. By mapping the sample in the $//<![CDATA[ \begin{array}{l}xy\end{array} //]]>$-plane while collecting information from the back-scattered light, a complete 3D image of the sample can be reconstructed. ^[3]

Frequency Domain OCT (FD-OCT)

Frequency domain OCT, also called Fourier domain OCT, is based on a pretty similar setup compared to TD-OCT, but with a different technique for signal detection. It provides an even more efficient and faster way to collect the information of the sample than scanning the sample by changing the length of the reference arm, as done in TD-OCT. This is due to the fact that all back reflections from the sample along the $//<![CDATA[ \begin{array}{l}z\end{array} //]]>$-axis are being measured simultaneously and so the mirror can be kept in a fixed position. This so-called spectral interference describes the modulations in intensity when recorded as a function of frequency. The rate of variation of the intensity over different frequencies indicates the location of different reflecting layers in the sample. Hence, this increase of speed opened up a new area of applications like real-time visualization for OCT, delivering a video-rate acquisition at up to 30 images per second. ^[3]

FD-OCT systems are known to provide a better sensitivity compared to TD-OCT setups. Thus, FD-OCT is the method of choice for situations with low light such as high-speed imaging. ^[4]

The procedures of spectral domain OCT and swept-source OCT described below are sub-types of FD-OCT.

Spectral Domain OCT (SD-OCT)

In addition to FD-OCT, SD-OCT uses a light source with a broad bandwidth of wavelengths and all are measured simultaneously using a spectrometer as detector. ^[3]

Figure 2 shows the different setup of the SD-OCT system in comparison to the general setup in figure 1. All components but the light detector are kept identical. As described in section 1.2.2 (FD-OCT), there is no need for a length variation of the reference arm and thus the mirror is kept still. The light is not directly detected, but rather spectrally decomposed by its wavelength using a prism or grid (8) and afterwards detected by a line sensor (9).

Swept-Source OCT (SS-OCT)

For SS-OCT the light source is swept through a range of wavelengths and the temporal output of the detector is converted to spectral interference. In contrast to SD-OCT, there is no need for a spectrometer while detecting the signal in SS-OCT, because the wavelength of the signal is time dependent and so can be matched with the corresponding wavelength the sample was irradiated with in the first place. ^[3]

SS-OCT provides the fastest way of scanning a specimen by OCT providing high quality images with scanning rates of many 100 kHz. ^[5]

Compatible Materials for Probing

Wasatch Photonics ^[2] describes the materials named hereinafter compatible for probing with OCT.

• All dielectric materials
• Paints, glasses, foils, coatings
• Polymers, silicone, rubber
• Plastics (to depths of ~ 2 mm)
• Metals (surface features only)

Key Parameters for OCT Systems

Resolution

The resolution of an OCT system in axial and transverse direction are independent from each other. Whereas in transverse direction this depends on the focus of the infrared light beam and the movement of the probe, the axial resolution (depth) is related to the bandwidth of the light source. For a Gaussian spectrum, the axial resolution $//<![CDATA[ \begin{array}{l}\lambda_c\end{array} //]]>$ is given by

where $//<![CDATA[ \begin{array}{l}\lambda\end{array} //]]>$ is the central wavelength and $//<![CDATA[ \begin{array}{l}\Delta \lambda\end{array} //]]>$ is the total bandwidth of the source ^[6,3]. The spectrum measured by the detector is affected by the response of the sample and the detector itself and so differs from the spectrum of the source. ^[3,1]

Typically, the resolution in axial direction lies between 7 µm and 15 µm ^[7]. A sub-micrometer axial resolution OCT can obtain a free-space resolution of ~0.75 µm ^[8].

Penetration

The imaging depth is primarily limited by the depth the light source penetrates the sample and so is highly dependent on the materials' characteristics. Additionally, for FD-OCT the depth is affected by two factors: the finite number of pixels and the optical resolution of the spectrometer. The Nyquist theorem defines the total length or depth after Fourier transform and is limited by the sampling rate of the spectral data. Therefore, the maximum imaging depth $//<![CDATA[ \begin{array}{l}Z_{\text{max}}\end{array} //]]>$ achievable in FD-OCT is

with the central wavelength $//<![CDATA[ \begin{array}{l}\lambda\end{array} //]]>$ and the total bandwidth $//<![CDATA[ \begin{array}{l}\Delta \lambda\end{array} //]]>$ sampled by $//<![CDATA[ \begin{array}{l}N\end{array} //]]>$ pixels ^[6]. ^[3]

Probing depths exceeding 20 mm have been demonstrated in transparent tissues, including the eye and the frog embryo. In the skin and other highly scattering tissues, OCT can visualize small blood vessels and other structures as deep as 1 mm or maximum 2 mm beneath the surface. ^[1,7]

Speed

The speed of an OCT system depends on several factors. The first being the amount of light received on the detector. Speed is directly related to the time the system needs to accumulate enough photons for a good signal to noise ratio. For a SD-OCT system the speed of the camera sensor and electronics are usually the limiting factors. For SS-OCT the speed of the swept source laser is the limit. ^[3]

^{Comparison to Other Non-Destructive Testing Methods}

OCT is a novel yet well-established technology that provides intermediate imaging depth at both high resolution and speed. It retains the flexibility of ultrasound testing in taking the probe to the sample, while eliminating the need for a coupling medium or contact with the specimen. Ultrasound testing as well as OCT, both methods use runtime measurement to analyze the sample ^[7]. OCT can be used as a powerful and efficient complement to ultrasound testing for material inspection and non-destructive testing due to the ability of ultrasound to achieve greater probing depths but with lower resolution ^[3]. An advantage of OCT over high-frequency ultrasonic imaging, a competing technology, is the relative simplicity and lower cost of the hardware on which OCT systems are based ^[1]. While ultrasonic inspection has become the standard in sub-surface imaging, it is limited in its speed, resolution, and ability to probe small or irregular samples. ^[2]

Confocal imaging provides sub-micrometer resolution, but is very expensive and limited to depths of less than 1 mm ^[2]. The superb optical sectioning ability of OCT enables these scanners to image microscopic structures in tissue at depths beyond the reach of conventional bright-field and confocal microscopes ^[1].

Hence, OCT fills the gap between microscopy and ultrasound testing and so complements these analysis methods for example in the field of medical diagnosis.

^Applications

In medicine various types of biological imaging such as ophthalmology, dermatology, and angiography are typical applications for diagnosis and monitoring ^[3]. Due to the ability of OCT having a high depth resolution as well as real-time visualization, OCT is well established as a medical diagnosis method providing in-situ process feedback ^[7]. Furthermore, one technical application is the online monitoring of printed electronics ^[9] and also in the art sector the varnish of paintings is analyzed using OCT ^[10].

References

1. J. M. Schmitt, “Optical coherence tomography (oct): A review”, IEEE Journal on Selected Topics in Quantum Electronics, vol. 5, no. 4, pp. 1205–1215, 1999. http://wwwlabs.uhnresearch.ca/biophotonics/staff_papers/vitkin/OCT_schmitt_overview.pdf.

2. Wasatch Photonics, “OCT Inspection for Non-Destructive Testing”, https://wasatchphotonics.com/product-category/optical-coherence-tomography/oct-inspection-ndt/ (visited on 01/03/2022).

3. Wasatch Photonics. “OCT Tutorial”, https://wasatchphotonics.com/oct-tutorial/ (visited on 01/03/2022).

4. R. Leitgeb, C. K. Hitzenberger, and A. F. Fercher, “Performance of fourier domain vs. time domain optical coherence tomography”, Opt. Express, vol. 11, no. 8, pp. 889–894, Apr. 2003. doi: 10.1364/OE.11.000889. http://www.osapublishing.org/oe/abstract.cfm?URI=oe-11-8-889.

5. W. Wieser, B. R. Biedermann, T. Klein, C. M. Eigenwillig, and R. Huber, “Multi-megahertz oct: High quality 3d imaging at 20 million a-scans and 4.5 gvoxels per second”, Opt. Express, vol. 18, no. 14, pp. 14 685–14 704, Jul. 2010. doi: 10.1364/OE.18.014685. http://www.osapublishing.org/oe/abstract.cfm?URI=oe-18-14-14685.

6. S. Aumann, S. Donner, J. Fischer, and F. Müller, "Optical Coherence Tomography (OCT): Principle and Technical Realization", Springer, 2019, isbn: 978-3-030-16637-3. doi: 10.1007/978-3-030-16638-0_3. https://link.springer.com/chapter/10.1007/978-3-030-16638-0_3.

7. Fakultät für Elektrotechnik und Informationstechnik. “Optische Kohärenztomographie”, https://etit.ruhr-uni-bochum.de/ptt/forschung/forschungsbereiche/optische-kohaerenztomographie/ (visited on 01/03/2022).

8. B. Povazay, K. Bizheva, A. Unterhuber, B. Hermann, H. Sattmann, A. F. Fercher, W. Drexler, A. Apolonski, W. J. Wadsworth, J. C. Knight, P. S. J. Russell, M. Vetterlein, and E. Scherzer, “Submicrometer axial resolution optical coherence tomography”, Optics Letters, vol. 27, no. 20, pp. 1800–1802, 2002.

9. E. Alarousu, A. Alsaggaf, and G. E. Jabbour, “Online monitoring of printed electronics by spectral-domain optical coherence tomography”, Scientific Reports, vol. 3, 2013.

10. T. Arecchi, M. Bellini, C. Corsi, R. Fontana, M. Materazzi, L. Pezzati, and A. Tortora, “A new tool for painting diagnostics: Optical coherence tomography”, Optics and Spectroscopy (English translation of Optika i Spektroskopiya), vol. 101, no. 1, pp. 23–26, 2006.