10.1117/2.1200909.1812
High-speed, full-range optical
coherence tomography
Ruikang K. Wang
A new technique to remove complex-conjugate artifacts inherent
to conventional Fourier-domain interferometry can double accessible
imaging depth.
In the past decade, optical coherence tomography (OCT)1 has
emerged as an important, noninvasive tool for depth-resolved
imaging of biological structures. Current developments are
shifting its focus toward spectral-domain and swept-source
techniques, since these frequency-domain OCT (FDOCT) versions have better acquisition speed and sensitivity than
time-domain imaging.2 However, a major drawback limiting
FDOCT’s practical application is the complex-conjugate ambiguity. The detected, real-valued spectral interferogram is Fourier
transformed to localize the scatter in the sample. The Fourier
transform of a real-valued function is Hermitian (i.e., its complex conjugate is equal to the original function with the variable changed in sign), so the reconstructed image is symmetrical
with respect to the interferometer’s zero-phase delay. This leads
to ambiguity in interpreting the resulting OCT images. Overcoming this requires placing the sample entirely within positive
or negative space, which leaves only half of the imaging depth
available in practice. Resolving the complex-conjugate ambiguity can double the imaging depth. In turn, this provides additional flexibility to explore the most sensible measurement range
near the zero-delay line.3
Researchers proposed several methods for achieving fullrange complex imaging for FDOCT. In spectral-domain OCT
(SDOCT), phase-shifting methods were among the first approaches attempting to achieve full-range imaging.4 However,
they have limited accuracy and constrain the imaging speed
and complex-conjugate rejection ratio, since several phase steps
must occur at the same sampling location. Instead, scientists
proposed an alternative technique. It uses a 3 × 3 optical
coupler as a phase-shifting element and enables simultaneous
detection of two quadrature phase-shifted interferograms to obtain complex, ambiguity-free images.5 This approach requires
two separate detectors for image acquisition, which adds cost
Figure 1. Optical coherence tomography (OCT) images from real-time
videos of (top) the anterior chamber of the human eye and (bottom)
a three-day-old chicken embryo. (a) and (c) Standard Fourier-domain
OCT (FDOCT). (b) and (d) Full-range complex FDOCT. The images’
physical size is 2.7×4.4mm2 (lateral × depth).
and complexity and thus limits its practical application. For
high-speed, full-range complex imaging, we (as well as others)
proposed a second alternative scheme in which a constant modulation frequency is introduced into the spatial interferogram
Continued on next page
10.1117/2.1200909.1812 Page 2/3
during transverse sample scanning.6, 7 Here, the complex data
is obtained through either Fourier or Hilbert transformation of
the B-scan (2D, cross-sectional) interferogram in the transverse
direction. This enables us to achieve full-range complex imaging
at high speed. Yet this method still needs a piezo stage to provide modulation in the interferograms, limiting the system’s full
imaging speed because of the stage’s mechanical movement.
We recently introduced a new FDOCT technique for in vivo
imaging, in which the carrier frequency is introduced by the
galvo scanner8 used for transverse scanning of the sample (i.e.,
in the x direction). To achieve this, we offset the sample beam
from the x scanner’s pivot axis. This causes path-length modulation and thus introduces a modulation frequency into the Bscan interferograms. The main advantage of this new method is
that it does not need any additional phase-shifting elements to
realize frequency modulation, while the modulation frequency
is given by the system itself. An additional, important advantage of using frequency modulation is that it is relatively
insensitive to sample movement, which is critical for in vivo applications. These aspects may allow us to achieve high-speed,
full-range complex SDOCT without additional hardware requirements, and without limiting imaging speed.
For high-speed SDOCT applications, another factor that could
constrain the system’s imaging speed is its data-acquisition
capability, particularly its data-flow management, since the
higher speed requires more computing power to rapidly capture, process, and display imaging information. To improve the
system’s acquisition rate, we developed custom software using
the Visual C++ platform, enabling the acquired raw data to connect with the computer’s random access memory directly, thus
optimizing the data-streaming process.
We have implemented an SDOCT system based on the beamoffset method, in conjunction with an algorithm that exploits
Hilbert transformation6 to obtain full-range complex imaging.
The resulting system can use the camera’s full imaging speed.
We have used a line-scan camera with a maximum speed of
47,000 A scans (depth-resolved scattering profiles) per second
to obtain full-range imaging. Before reconstructing the OCT image, we first converted the captured spectral data from optical
wavelength to optical wavenumber space. The DC autocorrelation terms (where DC refers to the mean value of the waveform)
were subsequently removed from the A scans by subtracting the
reference spectrum.9 We obtained the latter from averaging the
spectrogram over the B scans. Before Hilbert transformation to
construct the interferogram’s analytical function, we applied a
Fourier filtering technique around the zero-frequency region of
the time-variable interferogram to suppress self-correlation artifacts.
Figure 2. OCT images from real-time videos of (top) the palm skin and
(bottom) finger nail near the nail fold region of a human volunteer in
vivo. (a) and (c) Standard FDOCT. (b) and (d) Full-range complex
FDOCT. The images’ physical size is 2.7×4.4mm2 (lateral × depth).
We have used our full-range imaging system for a wide variety of applications. We demonstrated the technique’s feasibility
for high-speed in vivo imaging in human anterior-eye segments,
nail folds, skins, and chicken embryos at a speed of 47,000 A
scans/s (see Figures 1 and 2).
In summary, we have developed a novel technique to achieve
full-range complex FDOCT imaging for in vivo applications,
without the need for additional phase-shifting devices. Since
the phase modulation necessary for complex-signal reconstruction is given by the system itself, we can use the camera’s full
imaging speed without any limitations. Our ongoing studies focus on developing ultrahigh-speed, full-range, complex FDOCT
systems based on complementary metal oxide semiconductor
(CMOS) detectors for ophthalmic applications. CMOS cameras
Continued on next page
10.1117/2.1200909.1812 Page 3/3
are suitable for imaging the anterior and posterior chambers of
the eye, since the high imaging speed can overcome motion artifacts, which are unavoidable in human retinal imaging. Moreover, using CMOS one can select active pixels, thus offering
more freedom to choose imaging parameters and time.
Our research is currently supported by grants from the National Institutes of Health (R01 HL093140, R01 EB009682, and R01 DC010201)
and the American Heart Association (0855733G).
Author Information
Ruikang Wang
Oregon Health and Science University
Portland, OR
Ruikang Wang is a professor of biomedical engineering,
anesthesiology, and peri-operative medicine, and head of the
Biophotonics and Imaging Laboratory. His main research interests include high-resolution functional optical imaging using
coherence gating and confocal gating techniques applied to
healthcare, optical biopsy and functional imaging in tissue engineering, photoacoustic imaging, and light propagation in biological tissue.
References
1. D. Huang, E. Swanson, C. Lin, J. Schuman, W. Stinson, W. Chang, M. Hee,
T. Flotte, K. Gregory, C. Puliafito, and J. Fujimoto, Optical coherence tomography,
Science 254, pp. 1178–1181, 1991.
2. M. A. Choma, M. V. Sarunic, C. Yang, and J. Izatt, Sensitivity advantage of swept
source and Fourier domain optical coherence tomography, Opt. Express 11, pp. 2183–
2189, 2003.
3. G. Häusler and M. W. Lindner, Coherence radar and spectral radar: new tools for
dermatological diagnosis, J. Biomed. Opt. 3, pp. 21–31, 1998.
4. M. Wojtkowski, A. Kowalczyk, R. Leitgeb, and A. Fercher, Full range complex
spectral optical coherence tomography technique in eye imaging, Opt. Lett. 27, pp. 1415–
1417, 2002.
5. M. Sarunic, M. A. Choma, C. Yang, and J. A. Izatt, Instantaneous complex conjugate
resolved spectral domain and swept-source OCT using 3 × 3 fiber couplers, Opt. Express
13, pp. 957–967, 2005.
6. R. K. Wang, In vivo full range complex Fourier domain optical coherence tomography,
Appl. Phys. Lett. 90, p. 054103, 2007.
7. Y. Yasuno, S. Makita, T. Endo, M. Itoh, and T. Yatagai, Simultaneous B-M-mode
scanning method for real time Fourier domain optical coherence tomography, Appl. Opt.
45, pp. 1861–1865, 2006.
8. L. An and R. K. Wang, Use of a scanner to modulate spatial interferograms for in vivo
full-range Fourier-domain optical coherence tomography, Opt. Lett. 32, pp. 3423–3425,
2007.
9. R. K. Wang, Fourier domain optical coherence tomography achieves full range complex
imaging in vivo by introducing a carrier frequency during scanning, Phys. Med. Biol.
52, pp. 5897–5907, 2007.
c 2009 SPIE