Abstract.
We are developing an imaging system with the ability of imaging a 3-D
object plus its color spectrum information. The system makes use of
the spatial and wavelength selectivity of volume holograms, which act
as multiple focal-length lenses and color filters to separate 2-D slices
with different color from the 3-D object into various detectors. The
holographic microscope will be a powerful tool for imaging application
in cell-biology, biochemistry, materials research and any other 3-D
imaging application.
Motivation. Fluorescent probes are widely used in cell-biology,
biochemical, medical applications and other materials research, because
of its high detecting efficiency, compatibility with microscope optics,
fast response, and sensitivity to the molecular structures and local
environment. They are used as labels of different proteins and ion indicators,
and for the measurement of the membrane potential, the PH-factor, etc.
Within these applications, sometimes it is interesting to know all the
4-D information, including the accurate 3-D spatial location and the
color spectrum that is sensitive to the local chemical environment (see
Figure 1).
Figure 1. Human chromosome analysis.
A conventional microscope system can be used to image a 2-D slice of
a 3-D object. If the 4-D information (3-D spatial and color spectrum)
is to be collected, a scanning process is required for different depths
and wavelengths, such as the confocal microscope with various color
filters or the 3-D scanning spectroscope (see Figure 2). The scanning
process, besides usually being time-consuming, may bleach the fluorescent
probes due to the use of high intensities.
Figure
2.
Architecture of a confocal microscope: 3-D imaging is achieved by
scanning the microscope across the volume of the object.
One volume
hologram can act as an imaging lens with an inherent color filter, due
to the spatial selectivity of the hologram and the wavelength Bragg-matching
property. However, within the dynamic range of the holographic medium,
multiple holograms can be recorded into the same medium, which makes
it possible to generate a holographic imaging element that behaves as
a combination of imaging lenses and color filters (see Figure 3). Each
hologram will image a 2-D slice of a 4-D (3-D spatial dimensions plus
color information) object onto the detector. With the presence of multiple
holograms, multiple 2-D slices of the object can be collected and imaged
onto different areas on the detector at the same time, capturing the
full 4-D information of the object.
Figure
3. Sketch of a holographic imaging system: Multiple 2-D images are
obtained from different locations and colors of the object.
Research
and Achievements. The goals of this project are to understand the
imaging resolution of the holographic imaging system and compare it
with other current technologies; to investigate the system design for
optimal performance; and to validate the theoretical model through experimentation.
The basic performance of two holographic architectures is simulated
with computer and then compared with experimental results. Parameters
like image quality, the spatial resolution of the object, the color
selectivity and the photon efficiency for the incoherent fluorescence
are evaluated.
The first architecture is the 90-degree holographic recording between
a spherical wave and a plane wave. The advantage of this architecture
is that the holograms themselves replace the objective lenses in the
microscope, therefore no expensive optical elements are required. However,
both the simulation and the experiments indicate that the system suffers
from aberrations in the degenerate direction of the hologram, which
will distort the image and decrease the spatial resolution, as seen
in Figure 4.
Figure
4. Experiment and theoretical simulation of the image of a point
source in degenerate direction of a 90-degree hologram recorded with
a spherical wave and a plane wave.
The
second architecture overcomes this problem by using transmission-geometry
holographic recording with a microscope objective as collimating lens
(Figure 5). This implementation eliminates the aberration in the degenerate
direction, and also gives the flexibility to adjust the thickness of
the crystal for optimal resolution and efficiency trade-off. Figure
6 shows the simulation results for the depth resolution of a 700micron-thick
hologram read out with incoherent chromatic light source.
Figure
5. Structure of a holographic imaging system with transmission holograms.
Figure
6. Theoretical simulation of the depth selectivity for the transmission-geometry
holographic microscope. Each curve corresponds to a different chromatic
component pof the object.
Using
only one hologram recorded in a LiNbO3 crystal in transmission geometry,
2-D imaging of a mask is demonstrated for monochromatic light source
and chromatic light, as shown in Figure 7.
Figure 7. 2-D image of a mask using the transmission-geometry holographic
microscope. Top: Image of the mask when it is illuminated with monochromatic
light. Only a line in degenerated direction can be imaged by the hologram.
Bottom: Image of the same mask when it is illuminated with a broadband
light source. The horizontal axis corresponds to different color components
from the object.
Figure 8 shows the comparison between the direct optical microscope
imaging and the holographic imaging. The specimen used in the experiment
consists of 15micron-diameter fluorescent spheres arranged in two layers
at different depths. The horizontal direction of the holographic image
also gives the corresponding color spectrum of the fluorescence.
