GO
GO7. New optical imaging techniques to measure
structure and function in the eye
7.1 Optical coherence tomography to measure thickness
7.2 Optical redox fluorescence studies of cellular function
7.3 Three-dimensional visualization of the eye
7.4 Multi-photon excitation microscopy and spectroscopy
of ocular tissue
7. New Optical Imaging Techniques To Measure Structure and Function In The Eye
7.1
Optical Coherence Tomography To Measure Thickness
Optical coherence tomography (OCT) is an optical imaging technique based on a Michelson interferometer and a low coherence light source (Fercher, 1996; Puliafito et al. 1996, Masters 1999). It can obtain micron resolution cross-sectional imaging of biological tissue. The technique is analogous to ultrasound B-mode imaging; the difference is that with OCT light is used instead of acoustic waves.
The light source is typically a superluminescent diode operating in the infrared. Two-dimensional images of ocular structures are generated. Image contrast is dependent on differences in refractive index between adjacent structures; this is an analog of A-scans generated by ultrasound imaging instruments. If multiple OCT images are acquired by a series of laterally adjacent positions then an image is produced which is analogous to the B-mode in ultrasound. Tomographic or cross sectional imaging can be achieved from a successive set of axial range scans. Optical coherence tomography is noninvasive in that is operated in a non-contact fashion. It has been used to measure corneal thickness, measure retinal thickness, and for ocular biometry (Fercher, 1996; Puliafito et al., 1996).
Structures in the eye are imaged on the basis of their reflectivity
of infrared light. Structures with similar reflectivity are not imaged
as separate. Typically the instruments generate sagittal sections across
ocular tissue. What is measured is the optical distance which is a product
of the local refractive index and the mechanical distance. In order to
obtain the mechanical thickness of a structure in the eye it is necessary
to determine the refractive index at each region within the structure.
This is not always known and therefore an average refractive index is used
to calculate the corresponding mechanical distance.
The axial images are gray scale maps of intensity of reflected
light at the optical interfaces of the structure. Since the eye is in motion
it is necessary to use digital image processing techniques to warp the
acquired images and to correct for the movement of the eye. Ocular motion
varies the depth of each axial reflectivity profile; therefore it
is necessary to digitally correct for the motion artifact.
The gray scale intensity images are further processed with a false color representation. Typically the logarithm of the optical reflection or intensity of back scatter is represented by different colored pixels. The different colored pixels represent different intensities of reflected light in the biological structure; this is not the same as different morphological structures. Therefore, it is critical that the colored imaged be interpreted correctly; colored images have very different meaning than those from conventional histopathology using stains and dyes.
Images obtained with a clinical confocal microscope are en face images; i.e. they are in the plane of the tissue. Images obtained with OCT are sagittal slices through the tissue.
Optical coherence tomography has promise as a useful noninvasive technique for biometry of the anterior segment. It has been applied to measure corneal thickness in a clinical setting.
7.2 Optical Redox Fluorescence Studies of Cellular Function
Light can be used as a noninvasive probe to monitor cellular function in cells, tissues and organs (Masters and Chance, 1993). Cellular metabolism can be noninvasively optically monitored by the technique of redox fluorometry (Masters, 1990). Redox fluorescence imaging is a noninvasive method to noninvasively measure cellular oxidative function ( Masters, 1984a; 1984b; 1985; 1988; 1990c; 1993c; Masters et. al., 1982; 1983; 1989; 1991; 1993).
The probes are the intrinsic fluorescent pyridine nucleotides, and the flavoproteins. The pyridine nucleotides, NAD(P)H are excited in the region 365 nm and fluoresce in the region 400-500 nm. The flavoproteins are excited in the region 450 nm and fluoresce in the region 500-600 nm. Typically the fluorescence is imaged following ultraviolet excitation, and the changes in fluorescence intensity follow changes in cell and tissue oxidative metabolism.
Two-dimensional metabolic imaging based on redox fluorometry has two limitations. Firstly, there is photobleaching of the fluorescent molecules during the measurement period. Secondly, the short wavelengths preclude deep penetration of highly scattering tissues. Our previous works on the cornea in which we have used redox fluorometry is aided by the low light scatter of the normal cornea (Masters and Chance, 1993; Masters, 1993).
Cells contain two intrinsic fluorescent probes which can be used
for functional metabolic monitoring together with two-photon excitation
microscopy. The fluorescent intensity from intrinsic oxidized flavoproteins
present in mitochondrial of cells is a noninvasive measure of cellular
respiratory function. The fluorescence from the reduced pyridine
nucleotides NAD(P)H is another intrinsic probe to study cellular metabolism.
The fluorescence intensity from these intrinsic probes provides a noninvasive
optical method to monitor cellular respiration. The fluorescence
from NAD(P)H has be used to study cellular metabolism in many tissues and
organs due to the strong fluorescence intensity. The NAD(P)H fluorescence
intensity occurs in two compartments, the mitochondrial and the cytosolic;
this complicates the interpretation of the fluorescence studies.
The main advantage of measuring the fluorescence intensity from
the oxidized flavoproteins is that the fluorescence is localized in the
mitochondrial space. The light absorption of oxidized flavoproteins
has a broad maximum at 460 nm and extends from 430 to 500 nm. The
fluorescence intensity from oxidized flavoproteins in the cornea epithelium
occurs in the region from 520 nm to 590 nm with a broad maximum at 540
nm.
An example of a application of redox fluorometry is an investigation of the relationship between the oxygen transmissibility of different contact lens materials and the steady state oxygen concentration in the tear film under the contact lens (Masters, 1988). It was also possible to estimate the degree of cellular hypoxia in the anterior epithelium under the various contact lenses. Further development of the technique may provide an optical, noninvasive method to investigate cellular metabolic function in the eye.
7.3 Three-Dimensional Visualization of The Eye
Our ultimate interest is to visualize the living, in situ eye in three dimensions, and investigate the temporal changes of the three dimensional structures (four dimensional visualization) (Masters 1998a, Masters 1998b). Furthermore, these techniques to can be used to combine two different imaging modalities, i.e. backscattered light confocal microscopy and redox fluorometry (cellular oxidative function measured from mitochondrial fluorescence).
Once the stack of en face optical sections are converted to a three-dimensional volume in the computer it is possible to observe the volume from any arbitrary angle and direction. For example, although the confocal microscope acquired the confocal images as en face image in the plane of the tissue, it is possible to view the three-dimensional volume from a direction that is different from the plane of data acquisition. It is possible to view the in vivo human cornea as a stack of sagittal sections.
The two dimensional stack of confocal images form the eye is the input data set for the three dimensional visualization. In the ideal case the resolution in all three orthogonal directions is equal, i.e. x = y = z axis resolution. Then each volume element in the reconstructed three dimensional image produced from the two dimensional slices would be a cube. In practice, this is not the case and the volume elements, or voxels are not cubes but cuboids. The image processing operation of interpolation can be used to resize the slices and produce a volume formed from cubic voxels.
The principles of volume rendering techniques are as follows. The method is based on the assignment of a color and a partial opacity to each voxel. The three dimensional rendering is formed by merging voxels that project to the same pixel on the image plane. The technique has merit as compared to surface rendering methods since there is no requirement for making a binary classification of the data. The method also has problems; mainly high computational costs and difficulties to merge surface and volume data in a composite visualization.
There are many applications of confocal imaging and three dimensional
reconstruction of the living eye. Many application involve the clinical
observation of the living eye in human subjects or animals. Examples
of the authors prior work in three-dimensional visualization of stacks
optical sections acquired with a confocal microscope include the following.
The three-dimensional visualization of the full thickness of the ex vivo
rabbit cornea (Masters and Paddock, 1990a; 1990b; Masters and Farmer, 1993;
Masters, 1992c, 1993c). The in vivo human fundus and optic
nerve were visualized in three dimensions from a stack of optical sections
acquired with a modified scanning laser ophthalmoscope (Fitzke and Masters,
1993). The rabbit ocular lens has been visualized in three-dimensions (Masters,
1991c, 1992b, 1993b, 1993c, 1993d, 1994a). The in vivo human lens
has been visualized in three dimensions from a series of optical slices
acquired with a rotating Scheimpflug camera (Masters et al., 1996;
Masters, 1996a; 1996c; Masters and Senft, 1997, Masters 1997). Another
example is the shape visualisation of the anterior and posterior human
cornea in vivo (Masters, 1997). A final example of the three-dimensional
visualization of living biological tissue is the three-dimensional microscopy
of in vivo human skin (Masters et al., 1996).
These example serve to demonstrate the power of the three-dimensional
visualization technique. Below are some potentially useful applications.
The following represent some of the clinical applications:
Observation of donor tissue from an Eye Bank.
Clinical observation of the anterior segment of the human eye.
Clinical observation of the ocular lens.
Real-time, serial observation of wound healing.
Observation of the residual haze in the cornea following laser
refractive surgery.
Observation of nerve fibers in the anterior segment of the
eye.
Observation of metal objects in the eye.
Observation of ocular tissue that is semi-opaque.
7.4
Multi-Photon Excitation Microscopy and Spectroscopy of Ocular Tissue
Confocal microscopy uses a spatial filter (pinhole), located in front of the photodector, which results in increased axial resolution. It is this increase in axial resolution which permits optical sectioning of thick tissue and its subsequent three-dimensional computer reconstruction. An alternative technique is two-photon excitation microscopy which used a high-power near-infrared laser to focus the light into a diffraction limited spot at which two-photon excitation of the chromophores occurs. This technique also results in increased axial resolution, but it differs from confocal microscopy since it does not require a spatial filter.
Maria Göppert-Mayer, who worked in Göttingen, developed the theoretical foundation for two-photon excitation processes as the subject of a PhD thesis (Göppert-Mayer, 1931). Denk and co-workers invented a laser scanning microscope for fluorescence imaging based on two-photon excitation microscopy (Denk et al., 1990). Their work is a major advance in optical microscopy.
The principle of two-photon excitation processes is the simultaneous absorption of two photons; two photons with a wavelength in the near infrared can induce an electronic transition that normally requires a ultraviolet photon. The experimental verification of the two-photon absorption process is that the number of electronic transitions from the ground state to the excited state, and therefore the fluorescence intensity, is proportional to the square of the instantaneous intensity of the excitation light.
In general fluorophores have two-photon absorption peaks at approximately two times their one-photon absorption wavelengths; however, this general finding should be experimentally verified for each fluorophore. In contrast to one-photon excitation, with two-photon excitation the fluorescence emission occurs at wavelengths much shorter than the excitation wavelength. For example, two-photon excitation at 700 nm could produce fluorescence at 450 nm. therefore, it is not necessary to use ultraviolet excitation (for example at 366 nm) to produce the fluorescence.
The diffraction-limited focusing of the laser beam and the temporal
concentration of the 100-fsec pulses generated by the mode-locked laser
result in two-photon excitation processes only in the region of focus.
Outside of the focal region there is insufficient photon density, and therefore
no absorption process occur. Therefore, outside of the focal region there
is no signal; this is the origin of the optical sectioning. The optical
sectioning is derived from the physics of the optical system; therefore,
an aperture (pinhole) in front of the light detector is not required to
obtain the optical sectioning capability of the microscope.
Prior to the development of two-photon excitation microscopy
in order to excite chromophores which absorb in the ultraviolet region
it was necessary to use ultraviolet excitation light. When ultraviolet
fluorescence microscopy is used to image living cells and tissues it was
observed that cell viability was compromised and that there was photobleaching
of the chromophore.
The technique of two-photon excitation laser scanning microscopy
has been used to overcome the two problems cited previously; photobleaching
and tissue penetration during imaging of the redox states of the ex vivo
cornea (Masters 1996d; Piston, Masters, and Webb, 1995).
The technique of two-photon excitation microscopy is an important
imaging technique which minimized the damaging effects of phototoxicity
in living cells. Both the photodamage and the photobleaching are limited
to the focal slice. This is very different from a conventional confocal
microscope system in which the phototoxicity and photobleaching are
not limited to the focal slice (Masters, 1995)
The practical instrument consists of a laser scanning microscope
that is used for scanning the laser beam and for detection of the specimen
fluorescence. The typical light source is a self-mode-locked Ti:sapphire
laser, which is pumped by a CW argon-ion laser. This provides light in
the range 700 nm to about 1050 nm. A mode-locked titanium sapphire laser
which is pumped with a high-power argon ion laser is a common light source
for two-photon excitation microscopy. This system provides femtosecond
laser pulses in the wavelength range of 700 nm to 1000 nm. A cheaper,
alternative light source is a solid state system in which the argon laser
is replaced with a diode-pumped laser.
A recent advance in the application of multiphoton-excitation microscopy combined the technique of redox fluorescence metabolic imaging of cells and spectroscopy of in vivo human skin (Masters et al.1997, Masters et al 1998). In this study multiphoton excitation microscopy at 730 nm and 960 nm were used to image in vivo human skin autofluorescence from the surface to a depth of about 200 microns. The spectroscopic and fluorescence lifetime data suggest that reduced pyridine nucleotides, NAD(P)H, are the primary source of skin autofluorescence at 730 nm excitation. The emission spectra of the in vivo human skin had a peak at 450 nm which corresponds to fluorescence from NAD(P)H.
The applications of multi-photon excitation microscopy have been
demonstrated for the ex vivo human lens, the ex vivo rabbit cornea, and
in vivo human skin. The high peak power of the light used in multi-photon
excitation microscopy probably precludes its use in the living human eye
as a clinical imaging technique, although there may be other potentially
useful applications.
GO