7

Cellular Level Imaging of the Retina Using

Optical Coherence Tomography
Cherry Greiner and Stacey S. Choi
New England College of Optometry
USA

1. Introduction
Low-coherence interferometry was first demonstrated, in the early 1990s, in imaging the eye
due its tissue transparency and minimal attenuation of the input light. For this reason,
ophthalmic imaging is the most successful application of optical coherence tomography
(OCT). Since its first demonstration, advancements in technology and OCT designs have
resulted in higher resolution images of the retinal layers including retinal photoreceptors,
retinal pigment epithelium (RPE), and retinal nerve fiber layer (RNFL) thus improving the
potential for early diagnosis, the understanding of disease pathology, as well as the
assessment of therapeutic response. This chapter will mainly discuss improvements in OCT
systems over time, particularly developments towards cellular level imaging and their
clinical applications in ophthalmology in the present and in the future.
The success of OCT is in part due to its ability to obtain high structural information at
reasonable depths, bridging the capabilities of confocal and ultrasound imaging. OCT is
analogous to ultrasound pulse echo imaging, using light instead of acoustic waves. Since the
signal from OCT is traveling considerably fast, i.e. at the speed of light, OCT detects optical
echoes from tissue surfaces using low coherence interferometry technique. As a result,
higher resolution cross-sectional images of tissue structures are possible compared to
ultrasound imaging (Fercher et al., 1988; Fujimoto et al., 1986). Early OCT images were
based on time domain approaches with axial and lateral resolutions of less than 20 µm
(Huang et al., 1991; Swanson et al., 1993). Interests in OCT increased resulting in the use
preliminary in clinical studies and eventually to the development of commercial OCT
system for ophthalmic imaging in the mid-1990s. The development of high speed scanners
increased imaging speed capabilities of time domain OCT (TD-OCT). Nonetheless, dramatic
improvements in scan speed was not attained until the development of Fourier domain
detection, i.e. spectral domain (SD-OCT) or swept source OCT. By removing the need to
mechanically scan the reference mirror increased scan speeds by an order of magnitude
from a few thousand to ten to hundred thousand axial scans per second while also
improving sensitivity (de Boer et al., 2003; Lietgeb et al., 2003, Potsaid et al., 2008). In
addition to speed and sensitivity, other important parameters that should be considered for
cellular level imaging are high axial and lateral resolutions.
Developments of wider band light sources ≥ 80 nm, which we referred to as
“ultrawideband” in this chapter, have dramatically improved the axial resolution to ≤ 5 µm

www.intechopen.com
122 Selected Topics in Optical Coherence Tomography

at 800 nm. In addition, ultrawideband light source at other wavelengths, such as those over
1000 nm, provide structural imaging beyond the retinal layers (Povazay et al., 2003;
Puvanathasan et al., 2008). The lower scattering at longer wavelengths is an important
design consideration since it can extend the applicability of OCT technique to certain patient
populations with pre-existing afflictions (Povazay et al., 2003). In parallel to light source
developments to improve axial resolutions, researchers continue to improve lateral
resolutions in OCT ophthalmic imaging. A vital technological advancement in improving
lateral resolution in OCT is the use of adaptive optics (AO) to correct for ocular aberrations.
Multiple variations of AO-OCT currently exist; as AO-OCT systems continue to develop,
they offer great clinical potential for improved clinical assessment of retinal pathologies
great clinical potential for improved clinical assessment of retinal pathologies.
OCT systems have proven its effectiveness in detecting small retinal changes in patients
with various retinal diseases from age-related macular degeneration (AMD) to retinal
detachment. Hence it has become an integral part of diagnostic paradigm in clinical practice
in ophthalmology. By incorporating ultrawideband source and AO in OCT systems, it is
now possible to resolve individual photoreceptors as well as nerve fiber bundle. The power
of cellular level in-vivo imaging is that it allows clinicians to follow retinal and optic nerve
changes over time in patients, which ultimately results in earlier detection and intervention.
Current applications of both OCT and AO-OCT will be reviewed and future direction will
be discussed.

2. Ultrahigh resolution OCT designs

2.1 Ultrawideband light sources
For ophthalmic imaging, high axial resolution in OCT images is particularly important due
to the stratified organization of the retina. In OCT, the interference signal detected is the
field autocorrelation of the light source. Since the envelope of the autocorrelation is the
Fourier transform of the source power spectrum, the axial resolution is determined by
coherence length, lc, of the light source:

2 ln 2 c2
z  lc 

(1)

where c is the center wavelength and  is the spectral bandwidth. Axial resolutions can
therefore be improved by using lower wavelengths and/or increasing the spectral
bandwidth. In selecting the choice of wavelength, it is also important to consider tissue
absorption and scattering that influence OCT imaging contrast and penetration depth.
Hemoglobin and melanin are main tissue absorbers that limit penetration depth of
wavelengths in the visible to near infra-red regime. At longer wavelengths but less than
1800 nm, tissue scattering dominates, which decreases with increasing wavelength. At
wavelength greater than 1800 nm and around 1400 nm, imaging is limited by the absorption
of water. For these reasons and available laser at certain wavelengths, OCT typically uses
800 and 1300 nm center wavelengths. Ultrawideband sources such as superluminscent
diodes (SLD), femtosecond Titanium:Sapphire (Ti:Sapph) lasers and photonic crystal fiber
(PCF) based sources, are typically designed in either wavelength regime. Another important
consideration is the attainable resolution for a given spectral bandwidth at a given center

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 123

wavelength. Fig. 1 shows the required optical bandwidths to achieve axial resolutions at 800
and 1300 nm. Assuming similar spectral shape, the axial resolution is better with 800 nm
light source for the same spectral bandwidth. In addition, the shallower slope for 1300 nm
compared to 800 nm, indicates that the use of longer wavelength will require a larger
increase in bandwidth in order to reach the same axial resolution as with shorter
wavelengths.

Fig. 1. Comparison of optical bandwidths and axial resolutions at two center wavelengths,
800 and 1300 nm.

2.1.1 SLD sources

Light source for conventional OCT imaging is typically SLDs with spectral bandwidths of
20-30 nm resulting in OCT axial resolutions of 10-15 µm. Though this resolution is sufficient
to image retinal layers, identification of individual cells or sub-cellular structures is not
possible. Pushing towards cellular imaging, advanced SLDs have been developed with wide
spectral bandwidths ≥ 80 nm by multiplexing 2 or more SLDs, providing OCT axial
resolutions of < 10 m (Adler et al., 2004; Cense et al., 2009a; Hong et al., 2007; Ko et al.,
2004; Potsaid et al., 2008; Zawadzki et al., 2005). A prototype of this ultrawideband SLD for
OCT imaging was first demonstrated by Ko et al. in 2004. Two independent SLD sources at
center wavelengths 840 and 920 nm were combined with a custom designed fiber coupler
based on fiber GRIN microlenses and miniature dielectric coated mirrors (Adler et al., 2004;
Ko et al., 2004). OCT imaging in the retina with this SLD, which had a 155 nm bandwidth,
resulted in axial resolutions of ~ 3.2 µm. Fig. 2 shows a comparison between images
obtained with an ultrawideband SLD OCT system and a conventional commercial system
from the same subject. As shown, multiple retinal layers were better defined with the
ultrawideband SLD. Moreover, retinal features such as photoreceptor segments, external
limiting membrane and ganglion cell layers were better visualized (Ko et al., 2004). With a
170 nm SLD source, more choroidal vessels were visualized, allowing one to find connecting
vessels between adjacent imaging areas (Hong et al., 2007).

www.intechopen.com
124 Selected Topics in Optical Coherence Tomography

(a) (b)

Fig. 2. Comparison of in-vivo images acquired with (a) an OCT sytem with an
ultrawideband SLD source with an axial resolution of 3.2 µm and(b) a standard resolution
OCT with an axial resolution of 10 µm. Images from Ko et al., 2004.

One of the drawbacks of SLD is that its spectral shape can be less Gaussian-like and more
spectrally modulated. As a result, the axial point spread function (PSF) has sidelobes that
can result in image artifacts. Despite this limitation, spectral shaping may be applied
digitally to approximate a Gaussian envelope and reduce the sidelobes (Potsaid et al., 2008;
Tripathi et al., 2002). SLDs, however, offer advantages over other sources, which can be key
in transitioning high resolution OCT systems from the laboratory and into the clinic. In
particular, they are highly efficient, reliable and portable. Moreover, SLDs are lower in cost,
as much as an order of magnitude in comparison to femtosecond lasers.
In assessing the imaging quality between SLDs and femtosecond sources, a comparative
study has recently been performed by Cense et al. (2009). They used a Femtolasers Integral
Ti:Sapph laser with a spectral bandwidth, center wavelength and power of 135 nm, 800 nm,
and 60 mW, respectively (Cense et al., 2009a). In comparison, the BroadLighter, a
multiplexed SLD, has 110 nm spectral bandwidth, 840 nm center wavelength and 12 mW of
input power (Cense et al., 2009a). The output spectrums from each source are shown in Fig.
3a. Both sources were used on the same OCT system and imaged the same patient on
different days. In-vivo measurements showed similar axial resolutions in tissue specifically,
3.2 µm with the Ti:Sapph laser and 3.3 µm with the SLD (Fig. 3b) (Cense et al., 2009).
Despite more pronounced sidelobes in the axial PSF with the SLD compared to that with the
Ti:Sapph laser (Fig. 3b), the difference in the dynamic range was ≤ 5 dB, which the group
expected to be within the range of variability between individual measurement sessions
(Cense et al., 2009). Finally, a comparison of acquired volumetric images showed that the
wideband SLD can image individual cone photoreceptors, retinal capillaries and nerve fiber
bundles as well as the Ti:Sapph laser source (Fig. 3c-d) (Cense et al., 2009).

(a) (b)

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 125

(c) (d)

Fig. 3. (a) Comparison of spectra from ultrawideband Ti:Sapph (Integral) and multiplexed
SLD (BroadLighter). (b) Normalized coherence function from both sources. (c,d)
Comparison of high resolution imaging with the ultrawideband SLD (c) and Ti:Sapph (d)
light sources. Images from Cense, et al., 2009.

2.1.2 Titanium:Sapphire laser sources

Since OCT was developed in research laboratories, it is not surprising that the first sub-10
µm axial resolution OCT imaging was demonstrated using a laboratory source, femtosecond
Ti:Sapph laser, as opposed to a commercial SLD source (Clivaz et al., 1992). Femtosecond
mode-locked lasers can generate not only wide spectral bandwidth, but also provide
sufficient power for high speed imaging. A novel design is a high power Kerr-lens mode-
locked Ti:Sapph laser with a spectral bandwidth of up to 350 nm at 800 nm center
wavelength developed by Drexler et al. (1999). High performance was obtained with custom
designed chirped mirrors with high reflective bandwidth and intracavity dispersion
compensating prims (Drexler et al., 1999). In-vivo ophthalmic imaging of a normal subject
showed axial resolutions of ~ 2-3 µm (Drexler et al., 2001; Wojtkowski et al., 2004). Later,
improvements resulted in less complex, more compact and lower cost Ti:Sapph laser with
135-176 nm spectral bandwidth (Unterhuber et al., 2003). Spectral broadening was
accomplished by specially designed low dispersive mirrors and the footprint was reduced
by asymmetric folding of the resonator and some double pass configurations (Unterhuber et
al., 2003). In-vivo imaging of patients with macular pathologies showed images with ~ 3 µm
axial resolutions (Unterhuber et al., 2003).

2.1.3 PCF based sources

Another class of wideband light sources for high resolution OCT imaging is based on
nonlinear propagation of pulsed laser into PCF. PCFs are optical fibers made with pure silica
core surrounded by air holes throughout the length of the fiber. The large difference in
refractive index between the air and the silica leads to nonlinear effects resulting in
supercontinuum generation, i.e. extreme broadening of narrow band incident pulse into a
spectrally broadband continuous output. Just a few pulses with nJ of energy are sufficient to
generate supercontinuum (Humbert et al., 2006). Moreover, fiber dispersion can be engineered
to control temporal broadening. Hartl et al. (2001) reported the first supercontinuum based
OCT system. A femtosecond Ti:Sapph laser was launched into a PCF resulting in a spectral
bandwidth of 450 nm in the 1300 nm wavelength region resulting in an axial resolution of 2
µm in tissue (Hartl et al., 2001). Even higher resolutions have been demonstrated.
Modifications including fiber lengths, core sizes and polarization states can be used to
optimize the performance of PCF based sources (Apolonski et al., 2002). With some
optimization, a PCF based source was developed with a spectral bandwidth ~ 400 nm centered

www.intechopen.com
126 Selected Topics in Optical Coherence Tomography

at 750 nm resulting in an axial resolutions of ~ 0.5 µm (Povazay et al., 2002). OCT cross-
sectional imaging was demonstrated using monolayers of human colorectal adenocarcinomas
on coverslips with OCT images showing subcellular structures, such as the nucleoli (Povazay
et al., 2002). Humbert et al. (2006) later demonstrated that supercontinuum can also be
generated by launching Ti:Sapph laser into tapered PCF. Controlled tapering involved the
heating and stretching of a section of the fiber forming a narrow waist (Birks et al., 2000). The
decrease in the fiber pitch or distance between the air holes from the tapering process increases
the non-linear process (Humbert et al., 2006). Using tapered PCF, a maximum spectral
bandwidth of 194 nm centered at 809 nm was developed, resulting in OCT images with an
axial resolution of 1.5 µm in air (Humbert et al., 2006).
OCT imaging with PCF based light source centered at 1100 nm has also been demonstrated.
This wavelength regime is of interest since the absorption of melanin in RPE decreases with
wavelength, therefore imaging with wavelengths greater than 800 nm should allow for better
imaging of the choroid below the RPE layer. In a study by Povazay et al., (2003) a Ti:Sapph
laser was launched into a PCF generating a supercontinuum source from 400 to 1200 nm
which was then passed through long pass filter, resulting in a spectrum with a 165 nm
bandwidth centered around 1050 nm (Povazay et al., 2003). OCT images of ex-vivo pig retina
were acquired and also compared with OCT images using a 165 nm bandwidth Ti:Sapph laser
at 800 nm (Povazay et al., 2003). As expected, better visualization of intraretinal layer was
noted with the lower center wavelength source, i.e. Ti:Sapph laser (Fig. 4a) (Povazay et al.,
2003). However, the PCF based light source showed improved penetration in the choroid, as
evident by more visible choroidal blood vessels as shown in Fig. 4b (Povazay et al., 2003).

(a) (b)

Fig. 4. (a,b) Comparison of OCT images acquired with (a) a Ti:Sapph laser and (b) a PCF fiber
source. Red arrows indicates blood vessels in the choroid. Images from Povazay, et al., 2003.

More recently, 1064 nm was pumped into PCF generating both 800 and 1300 nm
wavelengths simultaneously. Near complete depletion of the pump wavelength created a
double peak spectrum with bandwidth and average power of 116 nm and ~ 30 mW,
respectively at 800 nm center wavelength and 156 nm and ~ 48 mW at 1300 nm center
wavelength (Aguirre et al., 2006). However, coupling loss reduced the power by nearly 50%
(Aguirre et al., 2006). Additionally, power stability was found to be sensitive to
environmental condition (Aguirre et al., 2006). Fiber end facet protection was recommended
to improve long term power stability (Aguirre et al., 2006). The supercontinuum source was
integrated into a TD-OCT system with different couplers, sample arm optics and
photodiode detector for each center wavelength but with a common galvometric scanner
and scanning stage (Aguirre et al., 2006). OCT imaging in hamster cheek pouch showed
axial resolutions of 2.2 µm at 800 nm and 4.7 µm at 1300 nm. Some blurring of the image was
observed at both wavelengths due to side lobe coherence artifacts on the PSF as a result of
spectral modulation at 800 and 1300 nm (Aguirre et al., 2006). To address this issue, further
work will investigate spectral smoothing (Aguirre et al., 2006).

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 127

Another dual band OCT imaging was demonstrated with 840 and 1230 nm center
wavelength with 1060 nm pump wavelength (Spoler et al., 2007). The spectral shape of the
supercontinuum source was tailored with a double pass prism, razor blades and an end
mirror (Spoler et al., 2007). Resulting bandwidths were > 200 nm after spectral filtering with
> 100 mW average power for each wavelength (Spoler et al., 2007). The source was
incorporated in a free space OCT system to launch both wavelengths simultaneously (Spoler
et al., 2007). Dichroic filters separated the two wavelength channels prior to detection
(Spoler et al., 2007). In-vitro OCT imaging of a rabbit eye was performed and showed axial
resolutions <2 and < 4 um at 840 and 1230 nm in tissue, respectively (Spoler et al., 2007). As
expected, 840 nm imaging yielded higher resolution while higher penetration was achieved
with the longer 1230 nm wavelength (Fig. 5a & 5b) (Spoler et al., 2007). Simultaneous
measurements allowed the feasibility to extract spectroscopic information, since differences
in image contrast are influenced by wavelength dependent scattering and absorption of
different types of tissue (Spoler et al., 2007). Differences in intensities are used to construct
spectroscopy images (Fig. 5c) that provided tissue differentiation, such as the conjunctiva
and sclera (Spoler et al., 2007).

(a) (b) (c)

Fig. 5.(a-c) OCT imaging at (a) 840 nm and (b)1230 nm (c) Differential color image based on
scattering intensity from each wavelength range. Higher scattering in 840 nm is shown in
shades of blue while those 1230 nm is in shades of red. Ep: Epithelium; BV: Blood vessel;
CM: Ciliary muscle; St: Stroma; Sc: Sclera; En: Endothelium; DM: Descement’s membrane;
CA: Chamber angle; Co: Conjunctiva. Images from Spoler et al., 2007.

2.2 Adaptive Optics (AO)

In OCT, axial and lateral resolutions are decoupled. The lateral resolution in OCT is similar
to that in standard microcopy:

4 f
x 
D
(2)

where f is the focal length of the focusing lens and the D is the beam diameter on the lens.
Therefore, lateral resolution can be improved by increasing the numerical aperture of the
beam delivery optics by decreasing the focal length and/or increasing the beam diameter. In
ophthalmic imaging, the focusing optics is limited by the eye itself. A 3mm pupil diameter
will yield a lateral resolution of 10–15 µm diffraction limited spot at 800 nm center
wavelength (Drexler, 2004). Increasing the pupil diameter by 1 mm will reduce the focused
beam spot diameter to 4.3 m, in theory. However with larger pupil size, the ocular
aberrations will limit the achievable size of the focus beam. Additionally, the use of
broadband light sources to improve axial resolution will further contribute to chromatic
aberrations.

www.intechopen.com
128 Selected Topics in Optical Coherence Tomography

Adaptive optics (AO) in OCT has been shown to be a successful approach in correcting ocular
and system’s aberrations to improve lateral resolutions. The two essential components of an
AO system for retinal imaging are a wavefront sensor and a corrector placed conjugate to the
pupil plane of the eye as shown in Fig. 6a. To measure ocular aberrations, a Shack- Hartmann
wavefront sensor containing a 2D array of lenslets, each of which with the same diameter and
focal length is used. Light reflected from the retina are distorted by aberrations of the eye
particularly from corneal surfaces and crystalline lens (Yoon et al., 2006). This reflected light is
spatially sampled by the 2D lenslet array on the wavefront sensor, forming multiple beams
focused at the focal plane of the lenses and imaged onto a camera. Light reflected from a
perfect eye, i.e., free of aberrations, emerges as a collimated beam. Therefore, Shack-Hartmann
spots will be imaged at the focal point of each lenslet, forming a regularly spaced grid (Fig. 6b,
left). In comparison, wavefront error from a real eye, i.e., with aberrations such as defocus and
astigmatism, will displace the Shack-Hartmann spots from the optical axis of each lenslet,
resulting in an image of an irregular grid (Fig. 6b, right). Since the displacement is
proportional to the wavefront slope sampled by the lenslet at a particular location, the
aberrations from the eye can be determined. The aberration information is then used to alter
the phase profile of the incident wavefront by changing the physical length that the light
propagates to correct the wavefront error. Different types of wavefront correctors or
deformable mirrors (DMs) have been used including those based on bimorph, piezoelectric
and micro-electro-mechanical systems (MEMS) technologies (Doble & Miller, 2006). An
example of a MEMs based mirror with piston/tip/tilt capability per mirror segment is shown
in Fig. 6c.

Fig. 6.(a)A schematic of an AO system for retinal imaging. (b) Example of a Shack-
Hartmann wavefront sensor images from a perfect eye (left) and from a real eye (right).
(c) Scanning electron microscope of a 163 segment MEMs based mirror. Images courtesy of
Iris AO, Inc.

The first attempt to improve retinal images from scanning laser ophthalmoscope with AO
was shown by Dreher et al. (1989). Using a DM, the group corrected the astigmatism in one
eye based on the patient’s spectacle prescription (Dreher et al., 1989). Several years later, the
Hartmann-Shack wavefront sensor was implemented to measure the wavefront error of the
eye (Liang et al., 1997). With the DM, aberrations from the system and the eye were also
corrected (Liang, 1997). Initially, wavefront compensation was achieved with a closed loop

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 129

feedback system. At each loop, images were acquired and the aberrations of the system and
eye were calculated. The DM partially, ~10%, corrected the wavefront error and loop
iterations were performed until a minimum wavefront error was reached. Later, closed loop
AO systems were developed with faster and automated wavefront sensing and dynamic
correction of ocular aberrations (Hofer, 2001; Fernandez 2001). To date, AO has been
successfully combined with ophthalmic imaging systems such as flood illuminated fundus
camera, scanning laser ophthalmoscope (SLO) and OCT systems for improved lateral
resolution and image contrast (Choi et al., 2005; Choi et al., 2006; Jonnal et al., 2007; Roorda
et al., 2002a; Roorda et al., 2002b; Thorn et al., 2003; Wolfing et al.,2006; Xue et al.,2007).

2.2.1 AO with ultrawideband source OCT

Cellular level OCT imaging can be realized by combining AO with ultrawideband sources
to achieve high lateral and axial resolutions. Hermann et al. (2004) first applied AO with a
130 nm bandwidth Ti:Sapph laser in a TD-OCT system. With AO, the Strehl ratio was
improved by at least a factor of 10 (Hermann et al., 2004). The axial resolution with this
system was 3µm while the lateral resolution was ~ 5-10 µm, 2-3 times better than similar
systems without AO (Hermann et al., 2004). In later developments, AO was combined with
SD-OCT systems, taking advantage of the higher sensitivity, enabling faster data
acquisition. Fig. 7 shows an example design of a SD-OCT system with an AO sub-system.

Fig. 7. A schematic of an AO-OCT system. P: Pupil plane; R: Retinal plane; M: Spherical

mirror;: Telescope magnification. Image from Zawadzki et al., 2005.

www.intechopen.com
130 Selected Topics in Optical Coherence Tomography

The system was designed with a series of telescopes that relayed the pupil conjugate plane
to the Shack Hartmann, bimorph DM and two scanning mirrors, while the retinal conjugate
plane is relayed to the OCT detector. The SLD source used had a spectral bandwidth of 140
nm centered at 890 nm. Axial and lateral resolutions were 6 µm and 4 µm, respectively. With
AO, it was found that image brightness was improved which the group attributed to be due
to astigmatism and defocus correction. Overall improvements allowed for better
visualization of retinal structures and cone photoreceptors (Zawadzki et al., 2005).
Although AO-OCT systems are mostly used to image the photoreceptor layer, it can also be
used to image the RNFL, which is important in the clinical management of glaucoma. With
an AO-OCT system, Kocaoglu et al. (2011) were able to easily visualize individual axonal
bundles in both enface and cross-sectional views due to improved image contrast, as a result
of wavefront error correction. For the first time, the researchers were able to quantify
individual nerve fiber bundles, i.e. width and thickness, from acquired OCT images
(Kocaoglu et al., 2011).

2.2.2 Dual-DM AO-OCT systems

Since wavefront correctors have a limited dynamic range, the use of trial lenses may be
necessary particularly in correcting subjects with large refractive errors. Others, however,
have developed alternative designs to increase the range of wavefront correction of AO-
OCT systems. One example is the addition of a Badal to coarsely correct for large defocus
(Chen et al., 2008). Another design alternative is to incorporate two wavefront correctors
with the first DM (DM1) mainly used to correct for large stroke, low order aberrations while
the second DM (DM2) used to correct for low stroke, high order aberrations (Cense et al.,
2009; Mujat et al., 2010; Zawadzki et al., 2007,2008). DM1 is first used to correct for
aberrations, as calculated by the wavefront sensor, then held static after reaching minimum
wavefront error. The remaining aberrations are again determined by the wavefront sensor,
and then corrected by DM2. Fig. 8 shows the time course of the dual DM AO
implementation. Though DM1 corrected most of the aberrations, these were further reduced
after both DMs were implemented (Zawadzki et al., 2007).

Fig. 8. Time course comparing RMS (left) and Strehl ratio (right) during dual-DM AO
implentation. Images from Zawadzki et al., 2007.

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 131

2.2.3 Correcting motion artefacts in AO-OCT

Despite faster image acquisition, involuntary head and eye movements still occur, despite
the use of bite bars and head restraints, resulting in motion artefacts that distort volumetric
OCT images. Though motion artefacts can be corrected via post-processing software,
hardware based implementations have also been demonstrated. For example, Fergueson et
al. were able to track and correct for retinal motion during or after data acquisition using a
low power tracking beam projected onto a retinal feature, such as a blood vessel junction
(Ferguson et al., 2004). Real time signal processing is used to determine the position and
orientation of the targeted retinal feature to correctly steer the tracking beam (Ferguson et
al., 2004). Since the OCT imaging and the tracking beam share the same pair of scanners, the
OCT beam similarly follows the retina (Ferguson et al., 2004).
More recently, an alternative system design to correct for motion artifact was demonstrated
by Zawadzki et al. by integrating SLO with an AO-OCT. In this combined system, SLO and
OCT retinal images were acquired simultaneously, since both modalities share the same AO
sub-system and vertical scanner, as shown in the optical schematic diagram, Fig. 9a
(Zawadzki et al., 2009). In this system, separate light sources are used for the OCT and SLO
optical paths. The SLO imaging beam is combined into and filtered out of the OCT imaging
path using dichroic filters. Similarly, a dichroic filter is used to separate SLO and OCT signal
in the detection path. To determine motion, the scanning frequencies are such that an SLO
frame is acquired for every OCT B-scan, therefore the researchers were able to correlate
features in the SLO images with disruption of layers from the B-scan images due to lateral
and transverse eye motion (Zawadzki et al., 2009). Fig. 9 b-d show an example of a co-
registered AO-OCT B-scan and AO-SLO frame (Zawadzki et al., 2009).

Fig. 9. (a) Optical schematic diagram of combined AO-SLO-OCT system. (b-d) Example of a
B-scan (b)acquired at the same time as a SLO frame (c) and resulting co-registered image
from both modalities (d). Images from Zawadzki et al., 2011.

Another variation of the combined SLO-OCT system, but without AO, is based on
interlacing spectrally encoded SLO frame with OCT B-scans. In spectrally encoded SLO, the
illumination beam is dispersed by a diffractive element that spectrally encodes each spatial
position (Tao, 2010). The back scattered light is then recombined and directed into a fiber,

www.intechopen.com
132 Selected Topics in Optical Coherence Tomography

which acts as a confocal detection pinhole (Tao et al., 2010). Spatial information is then
decoded by a spectrometer, the same detector used for OCT imaging (Tao et al., 2010). A
schematic diagram of the system is shown in Fig. 10. The slow galvanometer of the OCT
scan (Gy,S) acts as a switch between the spectrally encoded SLO (red) and the OCT path
(blue), directing the beam towards the combined path (green) (Tao et al., 2010). As shown,
the same light source and detector are used by both imaging modalities. During SLO
imaging, the galvanometer switch is flipped to its maximum angle, through the grating then
along the OCT optical path (Tao et al., 2010). The driving signals for the scanning mirrors
are such that a SLO frame is interlaced with an OCT B-scan (Fig. 10 b-c) (Tao et al., 2010).
De-interlacing of SLO and OCT frames were performed prior to image registration (Tao et
al., 2010). Although this imaging technique has yet to be demonstrated with AO sub-system
or the use of ultrawideband light sources, this design has the advantages of potentially
allowing for motion tracking while minimizing cost and footprint.

Fig. 10. (a) Optical schematic diagram of combined spectrally encoded SLO and OCT system.
(b) Timing diagram showing interlacing of SLO and B-scan images. (c) Illustrated scanning
geometry. f: lens, VPHG: Grating, G: Galvanometer, M: Mirror. Images from Tao et al., 2010.

2.2.4 Correcting chromatic aberrations in AO-OCT

In spite of the efforts for increasing transverse resolution, AO-OCT systems are still limited
in correcting for axial (or longitudinal) and lateral (or transverse) chromatic aberration of the
eye, particularly in dilated pupils in combination with the use of broader spectral
bandwidth sources. Axial chromatic aberration is manifested as a change in the focus plane
with wavelength, while lateral chromatic aberration appears from off-axis illumination
laterally shifting the focus with wavelength. The use of an achromatizing lens to correct for
axial chromatic aberration was first proposed by Fernandez et al. (2006). This concept was
later demonstrated in AO-OCT systems with wide spectral bandwidth sources (Fernandez
et al., 2008, Zawadzki et al., 2008). Custom design achromatizing lens was integrated by

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 133

Fernandez et al. in a conjugate pupil plane between the scanning mirrors and the DM, while
Zawadzki et al. placed the achromatizing lens immediately after collimating the input laser.
Both designs showed improved imaging of the individual photoreceptors with the addition
of the achromatizing lens. Fernandez et al. showed that with chromatic aberration
correction, photoreceptor structures between the inner and outer segment junctions appear
elongated, as opposed to rounded as in standard AO, resembling axial morphology of cones
in histology.
Although, an achromatizing lens can correct for axial chromatic aberrations, it does not
correct for lateral chromatic aberrations. Zawadzki et al. (2008) theoretically investigated the
extent of lateral chromatic aberration and its impact on retinal imaging. The group found
that the main contributors to this type of aberration are errors in the lateral positioning of
the eye and off-axis imaging (Zawadzki et al., 2008). Proper alignment of the eye relative to
the OCT camera optical axis should result in lateral chromatic aberration free imaging
(Zawadzki et al., 2008). Based on their system design, the group expects that lateral
chromatic aberration is smaller than the lateral resolution that can be achieved by the AO in
their system and therefore have little impact on image quality (Zawadzki et al., 2008).

2.2.5 AO in functional-based OCT systems

Adaptive optics has also transitioned into functional extensions of OCT, specifically
polarization sensitive OCT (PS-OCT). PS-OCT measures the intensity and polarization state
in tissue, thus allowing depth resolve imaging of birefringent retinal structures such as the
RNFL. The PS-OCT system designed by Cense et al. (2009b) utilized a dual DM AO-
subsystem. The group deduced the improvement in signal contrast and resolution with AO
allowed for a more sensitive measurement of change in phase retardation and hence
birefringence (Cense et al., 2009b). With this system, the group was able to quantify
birefringence of thin retinal nerve fiber bundles in the macula lutea, not measurable with
standard PS-OCT (Cense et al., 2009b). As AO becomes widely implemented, it is expected
that more AO will be integrated in other forms of OCT.

3. Clinical applications
OCT has become an integral part of routine diagnostic paradigm in ophthalmology. Due to
its non-invasive nature and the ability to extract quantitative morphological information, it
has been extensively used in the diagnosis of retinal and optic nerve diseases as well as in
the monitoring of their progression with treatments over time. Here, only some of the
selected representative papers on limited topics are briefly reviewed due to space limitation.

3.1 Ultrahigh resolution OCT

OCT has been able to reveal various microscopic changes ranging from macular edema to
disruptions in photoreceptor layer in patients with various retinal diseases and reduction of
RNFL in patients with glaucoma and other types of optic neuropathy.

3.1.1 Macular edema

OCT imaging is readily used to diagnose and follow macular edema. It occurs in many
disorders including diabetic retinopathy, age-related macular degeneration, retinal vein

www.intechopen.com
134 Selected Topics in Optical Coherence Tomography

occlusion, uveitis and macular hole. Macular edema is the result of an accumulation of fluid
in the retinal layers around the fovea with an increase in retinal thickness, which contributes
to vision loss. Fluorescein angiography (FA) has been and still is the critical diagnostic tool
for macular edema by identifying the characteristic stellar pattern of cystoid macular edema
(CME). Fig. 11 illustrates the appearance of CME through different diagnostic modalities,
namely, conventional fundus photography, FA and OCT.

(a) (b) (c)

(d)

Fig. 11. Cystoid Macular Edema (CME) imaged by 3 different diagnostic modalities. (a)
Conventional fundus picture (b, c) Fluorescein angiography (early and late stages
respectively) – dilation and leakage of capillary are revealed. Fluorescein dye pools in
cystoid spaces and arranged radially from the fovea. (d) OCT image – cystoid spaces are
shown. The central macula is indicated with an arrow. Images from Cunha-Vaz, et al., 2010.

In the early stage of macular edema, a diffuse swelling of the outer retinal layer is observed,
and then leads to the development of cystoid spaces. In the later stage, the large cystoid
spaces can extend from the RPE to the internal limiting membrane, which then eventually
rupture causing macular holes. In diabetic patients, macular edema involves three structural
changes – most commonly, sponge-like retinal swelling (88%), then CME (47%) and then
serous retinal detachment (15%) (Ontani et al., 1999). Fig. 12 shows the fundus pictures and
corresponding OCT images of retinal swelling and serous retinal detachment seen in
diabetic macular edema.
In sponge-like retinal swelling, the OCT image showed low reflective areas in the outer
retinal layers with relatively preserved inner retinal layer with interspersed low reflective
areas. The inner retinal layers were displaced anteriorly by the swollen outer retinal layers.
In serous retinal detachment, the OCT image clearly showed detached retina as a highly
reflective line with the formation of sub-retinal space underneath.
A good correlation has been found between OCT images and visual acuity and FA findings
(Hee et al., 1998; Jaffe & Caprioli, 2004; Ontani et al., 1999). Larger area of increased retinal
thickness and the involvement of the macular area correlated with greater loss of vision. Fig.
13 shows the correlation curve between visual acuity (VA) and retinal thickness

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 135

(a) (b)

Fig. 12. (a) Retinal swelling. The retinal thickness at the fovea is 560 µm. (b) Serous retinal
detachment with retinal swelling. The thickness of the sub-retinal space is 420 µm. The
vertical arrow on the fundus picture indicates the line and direction of scanning. Scan length
= 5 mm. Images from Otani, et al., 1999.

at the fovea in the eyes with CME (Ontani et al., 1999). Twenty eight of 59 eyes (47%)
developed CME, and the foveal retinal thickness and the VA were found to be negatively
correlated with the correlation coefficient of -0.64, P < 0.01.

Fig. 13. Correlation between VA and retinal thickness at the fovea in patients with CME.
Correlation coefficient = -0.64, P < 0.01. Figure from Otani, et al., 1999.

When compared with FA, OCT findings also correlated well with the pattern of edema
noticed in FA. Diffuse leakage is associated with nonspecific retinal swelling in the OCT
image, whereas petaloid-pattern hyper fluorescence is associated with large cystic spaces in
outer plexiform layer (OPL) and outer nuclear layer (ONL) and honeycomb hyper
fluorescence is associated with cystic changes in the OPL, ONL, inner nuclear layer (INL)
and inner plexiform layer (IPL) (Al-latayfeh et al., 2010).

www.intechopen.com
136 Selected Topics in Optical Coherence Tomography

3.1.2 Age related Macular Degeneration (AMD)

OCT has proven extremely useful in both the diagnosis and treatment of AMD. It is even
more so in the initial diagnosis of AMD lesion, as it is less hindered by blood or lipid as it is
in FA (Singh & Kaiser, 2007). A study compared the diagnostic accuracy of OCT with that of
FA in predicting the angiographic pattern of choroidal neovascularization, and found the
sensitivity and specificity to be 94% and 89% respectively when stereo fundus images were
added (Sandhu & Talks, 2005).
OCT has been used to update the drusen classification system; one group identified 3
drusen patterns whereas another group identified 17 different patterns (Khanifar et al., 2008;
Pieroni et al., 2006). Drusen were classified based on shape (concave, convex saw-toothed),
reflectivity (low to high), homogeneity (present or absent, with or without core), and foci of
hyper-reflectivity (present or absent). Soft indistinct drusen were mostly found to be convex
and homogenous with medium internal reflectivity and without overlying hyper-reflective
foci (Khanifar et al., 2008). Another study showed that basal laminar drusen exhibited a
moderately hyper-reflective and saw-toothed pattern on OCT images, and there was an
associated sub-retinal fluid in the macula which could be monitored by using automated
segmentation software in OCT (Leng et al., 2009). Fig. 14 shows FA, infrared and OCT
images of both eyes of a patient with familial drusen. The OCT images show a saw-tooth
appearance of the RPE-Bruch’s membrane complex consistent with the presence of basal
laminar drusen seen on the corresponding FA and infrared images.

Fig. 14. Early-frame FA (left), infrared (center) and OCT (right) images of both eyes of a
patient with familial drusen. Top row = right eye, bottom row = left eye. FA images show
early hyperfluorescence characteristic of basal laminar drusen. Green line on FA and
infrared images marks the area of OCT scans. In OCT images, the green arrows indicate a
saw-tooth appearance of basal laminar drusen. Images from Sohrab, et al., 2011.

The effect of drusen on the overlying retina was investigated using OCT and it was found
that the photoreceptor layer was thinned over 97% of drusen while inner retinal thickness
was unaffected and over at least one druse in 47% of eyes, photoreceptor outer segments
were absent. Furthermore, two types of hyper-reflective abnormalities were observed in the

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 137

neurosensory retina over drusen: (1) distinct hyper-reflective speckled pattern over drusen
(41% of AMD eyes) and (2) prominent hyper-reflective haze in the photoreceptor nuclear
layer over drusen (67% of AMD eyes) (Schuman et al., 2009). Fig. 15 shows the two types of
hyper-reflective abnormalities.

(a) (b) (c)

Fig. 15. (a,b) The sites of prominent diffuse hyper-reflective haze were observed over drusen
in 2 different eyes (shown with arrows). (c) Focal hyper-reflective speckling (arrows) was
visible over drusen. Images from Schuman, et al., 2009.

(a) (b)
(b)

(c) (d)

Fig. 16. FA and OCT images of the right eye in case 17. (a) Early-phase of FA. (b) Late-phase
of FA. (c) OCT image (top) and the extent of the hyper-reflective lesion is outlined (bottom).
(d) OCT C-scan was used to measure the greatest linear diameter (GLD) of the CNVM (top).
OCT B-scans with white lines represent the cut depicted by the C-scan shown above
(bottom). Images from Park, et al., 2010. Copyright [2010] ARVO.

The hyper-reflective foci overlying drusen are thought to represent progression of disease
with RPE cell migration into the retina and possible photoreceptor degeneration or glial scar
formation. The reduction in photoreceptor layer thickness suggests degeneration of
photoreceptors which eventually leads to vision loss.
The choroidal neovascular membranes (CNVMs) have also been detected in all 21 eyes with
exudative AMD by OCT and the anatomic growth pattern of the CNVM was determined
based on the OCT images (Park et al., 2010). A highly reflective sub-retinal and/or sub-RPE

www.intechopen.com
138 Selected Topics in Optical Coherence Tomography

lesion was visualized in the macula. Of twenty eyes, 7 eyes (33%) had > 90% sub-RPE
growth pattern (type 1), 10 eyes (48%) had > 90% sub-retinal growth pattern (type 2) and 3
eyes (14%) had a combined growth pattern. When the size of the CNVM was compared
between FA and OCT images, a good correlation was found between these modalities (i.e.,
for classic CNVM, r = 0.99, P < 0.0001 and for non-classic CNVM, r = 0.78, P < 0.001). Fig. 16
shows FA and OCT images from the right eye of a patient (case 17) with an occult CNVM
associate with exudative AMD. The FA images showed hyper-fluorescence centered over
the macula in the early-phase followed by an intense pooling of the dye into the pigment
epithelial detachment (PED) with ill-defined leakage at the superonasal edge in the late-
phase. The OCT image showed a large PED with an adjacent area of sub-retinal fluid. A
highly reflective sub-retinal lesion above the PED (consistent with type 2 CNVM) and a focal
spot of discontinuity in the RPE (indicated with *) with a possible small extension of the
hyper-reflectivity into the sub-RPE space were observed.

3.1.3 Glaucoma
OCT has shown the greatest potential for imaging glaucomatous structural changes such as
the RNFL. A cross-sectional observational study involving 160 control subjects and 134
patients with primary open-angle glaucoma (POAG) was conducted to evaluate the
accuracy of OCT in detecting differences in peripapillary RNFL thickness between normal
and glaucomatous eyes as well as between different severity groups (Sihota et al., 2006). The
POAG patients were divided into early (n=61), moderate (n=31), severe (n=25) and blind
(n=17) groups. The OCT was reliable in detecting changes in the RNFL thickness between
normal and all glaucoma subgroups. The RNFL thickness was significantly thinner in the
POAG patients compared to the control subjects and also the RNFL thickness continued to
reduce with an increase in the severity of POAG with P < 0.001.
The reproducibility of the RNFL thickness measurements was tested on 51 stable glaucoma
patients using Stratus OCT (Budens et al., 2008). For the mean RNFL thickness, the intra-
session and inter-session intraclass correlation coefficient (ICC) for the standard and fast
scans were 0.98 and 0.96 respectively. The coefficient of variation (COV) ranged from 3.8 to
5.2% (Carpineto et al., 2003). Other study has reported lower ICC of 0.5 and higher COV of
10%. However, the reproducibility is still sufficient to be useful clinically as a measure of
glaucoma progression. Fig. 17 shows a picture of glaucomatous optic nerve head (ONH)
with a dense superior visual field defect. A circle concentric OCT scan around the ONH
shows thinning of the RNFL in the inferiortemporal quadrant, which corresponds to both
the visual field defect and loss of the neural rim of the ONH.

Fig. 17. An ONH picture, a visual field plot and a circle concentric OCT scan around the
ONH from a glaucomatous eye. Green arrows indicate the area of thinning of the RNFL.
Images from Jaffe, et al., 2004.

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 139

3.2 Adaptive optics - OCT

With further improvement in lateral resolution of the system, it is possible to resolve even
smaller details in the retina.
As previously mentioned, Kocaoglu et al. (2011) have shown the first measurements of
RNFL axonal bundles (RNFB) in the living human eyes using their AO-OCT system. Four
normal subjects and one subject with an arcuate RNFL defect were imaged and they were
able to visualize individual RNFB in all subjects. Fig. 18 shows the individual RNFB in one
of the normal subjects (S4). As the RNFB approach fovea, they become thinner and separate.

(a) (b)

(c)

Fig. 18. (a) A wide field SLO, (b) An averaged C-scan and (c) An averaged B-scan images
acquired from subject, S4, at 3º nasal retina (indicated with a white square). The white lines
in (b) denote the area where the B-scans in (c) were obtained. The white solid rectangle in (c)
represents 2x magnification of the area outlined with white dotted rectangle. Images from
Kocaoglu, et al., 2011.

In conjunction with the expected inner retinal changes such as, thinning of the RNFL in
glaucoma and optic neuropathy, AO-OCT imaging has revealed outer retinal changes as
well at the retinal locations with reduced visual function, specifically shortening of cone
outer segments and blurring of the junction between the tip of the cone outer segments and
RPE. These outer retinal changes only occurred when there was a permanent visual field
loss. The same findings were observed in all types of optic neuropathy patients including
glaucoma (Choi et al., 2008; Choi et al., 2011).
Fig. 19 shows AO-OCT images obtained from a nonarteritic anterior ischemic optic
neuropathy (NAION) patient. The AO-OCT images were taken at 2 retinal locations, 2°
temporal 2° superior retina and 4° nasal 4° superior retina. The 4° nasal 4° superior retina
had better visual function than 2° temporal 2° superior retina, hence, the layer labeled 3 (the
junction between the cone outer segment tip and RPE) was better defined and distinct at
that location, and it was not visible at 2° temporal 2° superior retina.

www.intechopen.com
140 Selected Topics in Optical Coherence Tomography

(a) (b)

Fig. 19. AO-OCT images at two retinal locations in the right eye of the patient with NAION.
(a) 2° temporal 2° superior retina and (b) 4° nasal 4° superior retina. 1: ELM, 2: IS/OS, 3:
OS/RPE. Images from Choi, et al., 2008. Copyright [2008] ARVO.

4. Summary and future directions

With technological advancements, OCT has undergone substantial changes towards high
resolution cellular level OCT imaging. Light sources with ≥ 80 nm spectral bandwidth
paved the way for sub-10 µm axial imaging. At the same time, the adoption of AO in OCT
system’s addressed ocular and systems aberrations, improving lateral resolutions. Various
research groups have shown the potential of high resolution imaging with ultrawideband
sources with AO in OCT systems. However for these systems to be widely implemented
into the clinic, additional improvements are needed. Ultrawideband sources must provide
sufficient power and stability, and are compact and inexpensive. Similarly, AO-OCT
systems will also need to consider cost and size in addition to patient comfort and ease of
use. Finally, by understanding and addressing the limitations of AO-OCT systems, can
researchers and clinicians fully exploit the capabilities of AO to better image patients with
various forms of retinal diseases.
The advent of OCT technology has allowed extensive amounts of new anatomic,
physiologic, and pathologic data for the practice of ophthalmology. The ability to image
retinal and ONH structures in detail in-vivo makes OCT an ideal modality for detection of
early disease in screening and prevention for differential diagnosis of various diseases as
well as for quantifying therapeutic effects and identifying recurrence during follow-up. The
practicality of OCT in terms of obtaining motion artifact free images and presentation of
quantified data in a user friendly way requires further improvement. Also, with an increase
in the number of OCT users and commercial systems, there is a need to establish adequate
normative data and common clinical standards to allow for consistency across devices and
comparison between patients and diseases.

5. Acknowledgment
Financial support is acknowledged to the Department of Defense (DOD) Telemedicine and
Advanced Technology Research Center (TATRC), W81XWH-10-1-0738.

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 141

6. References
Adler, D.; Ko, T.; Konorev, A. et al. (2004). Broadband Light Source Based on Quantum-
well Superluminescent Diodes for High-resolution Optical Coherence
Tomography. Quantum Electronics, Vol. 34, No. 10, pp. 915-918, ISSN 1063-
7818
Aguirre, A.; Nishizawa, N. & Fujimoto, J. (2006) Continuum Generation in a Novel Photonic
Crystal Fiber for Ultrahigh Resolution Optical Coherence Tomography at 800 nm
and 1300 nm. Opt Express, Vol. 14, No. 3, (Feb 2006), pp. 1145-1160, ISSN 1094-
4087
Al-latayfeh, M.; Sun, J. & Aiello, L. P. (2010). Ocular coherence tomography and diabetic eye
disease. Semin Ophthal, Vol. 25, No. 5-6, (Sept – Nov, 12010), pp. 192-197, ISSN 1744-
5205
Apolonski, A.; Povazay, B.; Unterhuber, A. et al. (2002) Spectral shaping of supercontinuum
in a cobweb photonic-Crytal Fiber with Sub-20-fs-pulses. J Opt Soc Am A, B Vol. 19,
No.9, (Sept 2002), pp.2165-2170, ISSN 1084-7529
Birks, T.; Wadsworth, J. & Russel, P. (2000). Supercontinuum Generation in Tapered Fibers.
Opt Letters, Vol. 25, No. 19, (Oct, 2000), pp. 1415-1417, ISSN 0146-9592
Budenz, D., Fredette, M., Feuer, W. et al. (2008). Reproducibility of peripapillary retinal
nerve fiber thickness measurements with Stratus OCT in glaucomatous eyes.
Ophthalmology, Vol. 115, No. 4, (Apr 2008), pp 661-666, ISSN 1549-4713
Carpineto, P.; Ciancaglini, M.; Zuppardi, E. et al. (2003). Reliability of nerve fiber layer
thickness measurements using optic coherence tomography in normal and
glaucomatous eyes. Ophthalmol, Vol. 110,No 1, (Jan 2003) pp 190-195, ISSN 1549-
4713
Cense, B.; Koperda, E.; Brown, J. et al. (2009a). Volumetric Retinal Imaging with Ultra-high
Resolution Spectral-Domain Optical Coherence Tomography and Adaptive Optics
Using Two Broadband Light Sources. Opt Express, Vol. 17, No. 5, (Mar 2009), pp.
4095-4111, ISSN 1094-4087
Cense, B.; Gao, W.; Brown, J. et al., D.(2009b). Retinal Imaging with Polarization-Sensitive
Optical Coherence Tomography and Adaptive Optics, Opt Express, Vol. 17. No. 24,
(Nov 2009), pp. 21634-21651, ISSN 1094-4087
Choi, S.; Doble, N.; Lin, J. et al. (2005). Effect of Wavelength on in Vivo Images of the Human
Cone Mosaic. J Opt Soc Am A, Vol. 22, No. 12, (Dec, 2005), pp. 2598–2605, ISSN
1084-7529
Choi, S.; Doble, N.; Hardy, J. et al. (2006) In vivo imaging of the photoreceptor mosaic in
retinal dystrophies and correlations with visual function. Invest Ophthalmol Vis Sci,
Vol. 47, No. 5, (May 2006), pp. 2080-2092, ISSN 1552-5783
Choi, S.; Zawadzki, R.; Keltner, J. et al. (2008). Changes in cellular structures revealed by
ultra-high resolution retinal imaging in optic neuropathies. Invest Ophthalmol Vis
Sci, Vol. 49, No. 5, (May 2008), pp. 2103-2119, ISSN 1552-5783
Choi, S.; Zawadzki, R.; Lim, M. et al. (2011). Evidence of outer retinal changes in glaucoma
patients as revealed by ultrahigh-resolution in vivo retinal imaging. Br J
Ophthalmol, Vol. 95, No. , (Jan 2011), pp. 131-141, ISSN 0007-1161

www.intechopen.com
142 Selected Topics in Optical Coherence Tomography

Clivaz, X.; Marquis-Weible, F. & Salathe, R. (1992). Optical Low Coherence Reflectometry
with 1.9 µm Spatial Resolution. Elect Letters, Vol. 28, No. 16, (Jul 1992), pp.1553-
1555, ISSN 0013-5194
Coen, S. & Haelterman, M. (2001). Continuous-wave Ultrahigh-repetition-rate Pulse-train
Generation Through Modulational Instability in a Passive Fiber Cavity. Opt Letters,
Vol. 26, No. 1, (Jan 2001), pp. 39-41 , ISSN 0146-9592
de Boer, J.; Cense, B.; Park, B. et al. (2003). Improved Signal-to-Noise Ratio in Spectral
Domain Compared with Time-Domain Optical Coherence Tomography. Opt
Letters, Vol. 28, No. 21 , (Nov 2003), pp. 2067-2069, ISSN 0146-9592
Dreher, A.; Billie, J. & Weinreb, R. (1989). Active Optical Depth Resolution Improvement of
the Laser Tomographic Scanner. Appl Opt, Vol. 28, No. 4, (Apr 1989), pp. 804–808,
ISSN 0003-6935
Drexler, W.; Mogner, U.; Kartner, F et al. (1999). In vivo Ultrahigh-resolution Optical
Coherence Tomography. Opt Letters, Vol. 24, No. 17, (Sept 1999), pp. 1221-1223,
ISSN 0146-9592
Drexler, W.; Morgner, U.; Ghanta, R. et al. (2001). Ultrahigh-resolution Ophthalmic Optical
Coherence Tomography. Nat Med, Vol. 7, No. 4, (Apr 2001), pp. 502-507, ISSN
1078-8956
Drexler, W.; Sattman, H.; Hermann B. et al. (2003). Enhanced Visualization of Macular
Pathology With the Use of Ultrahigh-Resolution Optical Coherence
Tomography. Arch of Ophthalmol, Vol. 121, No. 5, (May 2003), pp. 695-706, ISSN
0003-9950
Drexler, W. (2004). Ultrahigh-resolution Optical Coherence Tomography. J Biomed Opt, Vol
.9, No. 1, (Jan/Feb 2004), pp. 47-74, ISSN 1083-3668
Doble, N. & Miller, D. (2006). Wavefront Correctors for Vision Science, In: Adaptive Optics for
Vision Science, J. Porter, H. Queener, J. Lin, K. Thorn, A. Awwal (Ed) , pp. 82-117,
ISBN 978-0-471-67941-7, Hoboken, NJ, USA.
Fercher, A.; Mengedoht, K. & Werner, W. (1988) Eye-length Measurement by
Interferometry with Partially Coherent Light. Opt Letters, Vol. 13, No. 3, (Mar 1988),
pp. 186-188, ISSN 0146-9592
Fernandez, E.; Iglesias, I. & Artal, P. 2001. Closed-loop Adaptive Optics in the Human Eye.
(2001). Opt Express, Vol. 26, No. 10, (May, 2001), pp. 746-748, ISSN 0146-9592
Ferguson, R.; Hammer, D.; Paunescu, L. et al. Tracking Optical Coherence Tomography. Opt
Letters, Vol. 29, No. 18, (Sept 2004), pp. 2139-2141, ISSN 0146-9592
Fujimoto, J; De Silvestri, S.; Ippen, E. et al. (1986). Femtosecond Optical Ranging in
Biological Systems. Opt Letters, Vol. 11, No. 3, (Mar 1986), pp. 150-152, ISSN 0146-
9592
Hartl, I.; Li, X.; Chudoba, C. et al. (2001). Ultrahigh-resolution Optical Coherence
Tomography Using Continuum Generation in a Air-Silica Microstructure Optical
Fiber. Opt Letters, Vol. 26, No. 9, (May 2001), pp. 608-610, ISSN 0146-9592
Hee, M.; Puliafifito C.; Duker, J. et al. (1998). Topography of diabetic macualr edema with
optical coherence tomography. Ophthalmology, Vol. 105, No. 2, (Feb 1998), pp. 360-
370, ISSN 1549-4713

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 143

Hofer, H.; Chen, L.; Yoon, G. et al. (2001). Improvement in Retinal Image Quality with
Dynamic Correction of the Eye’s Aberrations. Opt Express, Vol. 8, No. 11, (May
2001), pp. 631–643, ISSN 0146-9592
Hong, Y.;Makita, S.; Yamanari, M. et al. (2007). Three-dimensional Visualization of
Choroidal Vessels by Using SStandard and Ultra-high Resolution Optical
Coherence Tomography. Opt Express, Vol. 15, No. 12, (Jun 2007), pp. 7538-7550,
ISSN 0146-9592
Huang, D.; Swanson, E.; Lin, C. et al. (1991). Optical Coherence Tomography. Science, Vol.
254, No.5035 , (Nov 1991), pp.1178-1181, ISSN 0036-8075
Humbert, G.; Wadsworth, W.; Leon-Saval, G. et al. (2006). Supercontinuum Gneration
System for Optical Coherence Tomography based on Tapered Photonic Crytal
Fibre. Opt Express, Vol. 14, No. 4, (Feb 2006), pp. 1596-1603.
Jaffe, G. J. & Caprioli, J. (2004). Optical coherence tomography to detect and manage retinal
disease and glaucoma. Am J Ophthalmol, Vol. 137, No. 1, (Jan 20044), pp. 156-169,
ISSN 0002-9394
Jonall, R.; Rha, J.; Zhang, Y et al. (2007). In Vivo Functional Imaging of Human Cone
Photoreceptors. Opt Express, Vol. 15, No. 24, (Nov 2007), pp 16141-16160, ISSN
0146-9592
Jungtae, R.; Jonnal, R.; Thorn, K., et al. (2006). Adaptive Optics Flood-Illumination Camera
for High Speed Retinal Imaging. Opt Express, Vol. 14. No. 10, (May 2006), pp. 4552-
4569, ISSN 0146-9592
Khanifar, A.; Koreishi, A.; Izatt, J. et al. (2008). Drusen ultrastructure imaging with
spectral domain optical coherence tomography in age-related macular
degeneration. Ophthalmology, Vol. 115, No. 11, (Nov 2008), pp.1883-1890, ISSN
1549-4713
Ko, T.; Adler, D.; Fujimoto, J. et al.. (2004). Ultrahigh Resolution Optical Coherence
Tomography Imaging with a Broadband Superluminscent Diode Light Source. Opt
Express, Vol 12, No. 10, (May 2004), pp. 2112-2119, ISSN 1094-4087
Kocaoglu, O.; Cense, B.; Jonnal, R. et al. (2011). Imaging Retinal Nerve Fiber Bundles Using
Optical Coherence Tomography with Adaptive Optics. Vision Res, Vol. 51, No. 16,
(Aug 2011), pp. 1835-1844, ISSN 42-6989
Leng, T.; Rosenfeld, P.; Puliafito, G. et al. (2009). Spectral domain optical coherence
tomography characteriatics of cuticular drusen. Retina, Vol. 29, No. 7, (Jul-Aug
2009),pp. 988-993, ISSN 1539-2864
Liang, J.; Williams, D.; & Miller, D. (1997). Supernormal Vision and High-Resolution Retinal
Imaging Through Adaptive Optics. J Opt Soc Am A, Vol. 14 No. 11, (May, 2007), pp.
2884–2892, ISSN 1084-7529
Lietgeb, R.; Hitzenberger, C. & Fercher, A. (2003). Performance of Fourier Domain vs. Time
Domain Optical Coherence Tomography. Opt Express, Vol. 11, No. 8, (Apr 2003),
pp.889-894, ISSN 0146-9592
Mujat, M.; Ferguson, D.; Patel, A. et al. (2010). High Resolution Multimodal Clinical
Ophthalmic Imaging System. Opt Express, Vol. 18, No. 11, (May 2010), pp. 11607–
11621, ISSN 0146-9592

www.intechopen.com
144 Selected Topics in Optical Coherence Tomography

Ontani, T.; Kishi, S. & Maruyama, Y. (1999) Patterns of diabetic macular edema with optical
coherence tomography. Am J Ophthalmol, Vol. 127, No. 6,(Jun 1999) pp. 688-693,
ISSN 0002-9394
Park, S.; Truong, S., Zawadzki, R. et al. (2010). High-resolution Fourier-domain Optical
Coherence Tomography of Choroidal Neovascular Membranes Associated with
Age-related Macular Degeneration. Invest Ophthalmol Vis Sci., Vol. 51, No. 8, (Aug
2010), pp. 4200-4206, ISSN 1552-5783
Pieroni, C.; Witkin, A.; Ko, T et al. (2006). Ultrahigh resolution optical coherence
tomographyin non-exudative age-related macular degeneration. Br J Ophthalmol,
Vol. 90, No. 2, (Feb 2006), pp. 191-197, ISSN 0007-1161
Potsaid, B.; Gorczynska, I.; Srinivasan, V. et al. (2008). Ultrahigh Speed Spectra/Fourier
Domain OCT Opthalmic Imaging at 70,000 to 312,500 Axial Scans per Second.
Ophthalmol Express, Vol. 16, No. 19, (Sept 2008), pp. 15149–15169, ISSN 0146-9592
Povazay, B.; Bizheva, K.; Hermann, A. et al. (2003). Enhanced Visualization of Choroidal
Vessels using Ultrahigh Resolution Ophthalmic OCT at 1050 nm. Opt Express, Vol.
11, No. 17, (Aug 2003), pp. 1980-1986, ISSN 0146-9592
Puvanathasan, P.; Forbes, P.; Ren, Z. et al. (2008). High-speed, high resolution Fourier-
domain optical coherence tomography system for retinal imaging in the 1060 nm
wavelength region. Opt Letters, Vol. 21, No. 21, (Nov 2008) pp. 2479-2481, ISSN
0146-9592
Roorda, A. & Williams D. (2002a). Optical Fiber Properties of Individual Human Cones. J
Vis, Vol. 2, No. 5, (Apr, 2002), pp. 404–412, ISSN 1534-7362
Roorda, A.; Romero-Borja, F.; Donnely, W. et al. Adaptive Optics Laser Scanning
Opthalmoscopy. (2002b). Opt Express, Vol. 10, No. (May 2002), pp. 405-412, ISSN
0146-9592
Sandhu, S. & Talks S. (2005). Correlation of optical coherence tomography, with or without
additional colour fundus photography, with stereo fundus fluorescein angiography
in diagnosing choroidal neovascular membranes. Br J Ophthalmol, Vol. 89, No. 8,
(Aug 2005), pp. 967-970, ISSN 0007-1161
Schocket, L.; Wikins A.; Fujimoto, J. et al. (2006). Ultrahigh-Resolution Optical
Coherence Tomography in Patients with Decreased Visual Acuity after Retinal
Detachment Repair. Ophthalmol, Vol. 113, No. 4, (Apr 2006), pp. 666-692, ISSN
0161-6420
Schuman, S.; Koreishi, A.; Farsiu, S. et al. (2009). Photoreceptor layer thinning over drusen in
eyes with age-related macular degeneration imaged in vivo with sepctral domain
optical coherence tomography. Ophthalmol, Vol. 116, No. 3, (Mar 2009), pp. 488-496,
ISSN 1549-4713
Sihota, R.; Sony, P.; Gupta, V. et al. (2006). Diagnostic Capability of Optical Coherence
Tomography in Evaluating the Degree of Glaucomatous Retinal Nerve Fiber
Damage. Invest Ophthalmol Vis Sci, Vol. 47, No. 5, (), pp 2006-2010, , ISSN 1552-
5783
Singh, R. & Kaiser, P. K. (2007). Advances in AMD imaging. Int Ophthalmol Clin, Vol. 47, No.
1 (Winter 2007), pp. 65-74, ISSN 0020-8167

www.intechopen.com
Cellular Level Imaging of the Retina Using Optical Coherence Tomography 145

Spoler, F.; Kray, S.; Grychtol, P. et al. (2007). Simultaneous Dual-band Ultra-high Resolution
Optical Coherence Tomography. Opt Express, Vol. 15, No. 17, (Aug 2007), pp.
10832-10841, ISSN 1094-4087
Srinivasan, V.; Monson, B.; Wojtkowski, M. et al. (2006). High-Definition and 3-dimensional
Imaging of Macular Pathologies with High-speed Ultrahigh-Resolution Optical
Coherence Tomography, Ophthalmology, Vol. 113, No. 11, (Nov 2006), pp. 2054-
2065, ISSN 1549-4713
Srinivasan, V.; Monson, B.; Wojtkowski, M. et al. (2008). Characterization of Outer Retinal
Morphology with High-Speed, Ultrahigh-Resolution Optical Coherence
Tomography. Invest Ophthalmol Vis Sci, Vol. 49, No. 4, (Apr 2008), pp. 1571-1579,
ISSN 0146-0404
Swanson, E.; Izatt, J.; Hee, M. et al. (1993). In Vivo Retinal Imaging by Optical Coherence
Tomography. Opt Letters, Vol. 18, No. 21,(Nov 1993), pp. 1864-1866, ISSN 0146-
9592
Tao, Y.; Farsiu, S. & Izatt, J. (2010). Interlaced Spectrally Encoded Confocal Scanning Laser
Ophthalmoscopy and Spectral Domain Optical Coherence Tomography. Biomed
Opt Express, Vol. 1, No. 2, (Sept 2010), pp. 431-440, ISSN 0146-9592
Unterhuber, A.; Povazay, B; Hermann, B. et al. (2003). Compact, Low-cost Ti:Al2O3 Laser
for In Vivo Ultrahigh-resolution Optical Coherence Tomography. Opt Letters, Vol.
28, No. 11 (Jun 2003), pp. 905-907, ISSN 0146-9592
Witkin, A.; Ho, T.; Fujimoto, J, et al. (2006). Ultra-high Resolution Optical Coherence
Tomography Assessment of Photoreceptros in Retinintis Pigmentosa and Related
Diseases. Am J Ophthalmol, Vol. 142, No. 6, (Dec 2006), pp.945-952, ISSN 0002-
9394
Wolfing, J.; Chung, M.; Carroll, J. et al. (2006). High Resolution Imaging of Cone-Rod
Dystrophy with Adaptive Optics. Ophthalmology, Vol. 113, No. 6, (Jun 2006), pp.
1014-101, ISSN 1549-4713
Xue, B.; Choi, S.; Doble, N. et al. (2007). Photoreceptor counting and montaging of en-face
retinal iamges from an adaptive optics fundus camera. J Opt Soc Am A, Vol. 24,
Issue 5, (Apr 2007) , pp. 1364-1372, ISSN 1084-7529
Zawadazki, R.; Jones, S.; Olivier, S. et al. (2005). Adaptive-Optics Optical Coherence
Tomography for High-Resolution and High-Speed 3D Retinal In Vivo Imaging. Opt
Express, Vol. 13, No. 21, (Oct 2005), pp. 8532-8546, ISSN 0146-9592
Zawadzki, R.; Choi, S.; Olivier, S. et al. (2007). J Opt Soc Am A, Vol. 24, Issue 5, (May 2007) ,
pp. 1373-1383, ISSN 1084-7529
Zawadzki, R.; Cense, B.; Zhang, Y. et al. (2008). Ultrahigh-resolution Optical Coherence
Tomography with Monochromatic and Chromatic Aberration Correction. Opt
Express, Vol. 16, No. 11, (May 2008), pp. 8126-8143, ISSN 1094-4087
Zawadazki, R., Jones, S., Chen, D. et al. (2009). Combined Adaptive Optics – Optical
Coherence Tomography and Adaptive Optics – Scanning Laser Ophthalmoscopy
System for Retinal Imaging, Proceed of SPIE, ISBN 9780819474094, San Jose, CA,
USA, Jan 24, 2009

www.intechopen.com
146 Selected Topics in Optical Coherence Tomography

Zawadzki, R.; Jone, S.; Pilli, S.; Balderas-Mata, S.; Kim, D.; Olivier, S. & Werner, J. (2011)
Integrated Adaptive Optics Optical Coherence Tomography and Adaptive Optics
Scanning Laser Ophthalmoscope System for Simultaneous Cellular Resolution In
Vivo Retinal Imaging. Biomed Opt Express, Vol. 2, No. 6, (May 2011), pp. 1674-1686,
ISSN 0146-9592

www.intechopen.com
 Selected Topics in Optical Coherence Tomography
Edited by Dr. Gangjun Liu

ISBN 978-953-51-0034-8
Hard cover, 280 pages
Publisher InTech
Published online 08, February, 2012
Published in print edition February, 2012

This book includes different exciting topics in the OCT fields, written by experts from all over the world.
Technological developments, as well as clinical and industrial applications are covered. Some interesting
topics like the ultrahigh resolution OCT, the functional extension of OCT and the full field OCT are reviewed,
and the applications of OCT in ophthalmology, cardiology and dentistry are also addressed. I believe that a
broad range of readers, such as students, researchers and physicians will benefit from this book.

How to reference
In order to correctly reference this scholarly work, feel free to copy and paste the following:

Cherry Greiner and Stacey S. Choi (2012). Cellular Level Imaging of the Retina Using Optical Coherence
Tomography, Selected Topics in Optical Coherence Tomography, Dr. Gangjun Liu (Ed.), ISBN: 978-953-51-
0034-8, InTech, Available from: http://www.intechopen.com/books/selected-topics-in-optical-coherence-
tomography/cellular-level-imaging-of-the-retina-using-optical-coherence-tomography

InTech Europe InTech China

University Campus STeP Ri Unit 405, Office Block, Hotel Equatorial Shanghai
Slavka Krautzeka 83/A No.65, Yan An Road (West), Shanghai, 200040, China
51000 Rijeka, Croatia
Phone: +385 (51) 770 447 Phone: +86-21-62489820
Fax: +385 (51) 686 166 Fax: +86-21-62489821
www.intechopen.com