 |
Description  |
|
|
BACKGROUND OF THE INVENTION
This invention relates generally to the field of optical measuring and
testing, and more specifically to apparatus incorporating interference
fringe pattern generators for retinal acuity and related testing.
Ophthalmologists use a variety of techniques to measure ophthalmic and
related functions and characteristics. Some of these measurements indicate
retinal acuity at both the central and peripheral retinal regions. Others
measure neurological response to a range of visual stimuli.
For example, ophthalmologists use apparatus of the type that implements
either Moire or interference techniques to test and measure retinal
acuity. This measurement is obtained by varying the "fineness" of the
fringes projected onto the retina and monitoring the patient's ability to
resolve them. The patient's ability to resolve a fringe pattern of a
certain "fineness" converts directly into a measurement of retinal acuity.
Tests of peripheral vision can lead to an early diagnosis of glaucoma.
Prior instruments of this general type used to measure the acuity of the
central field of retina have not been employed successfully to measure the
acuity of the eccentric region of the retina which is the area associated
with peripheral vision. This is mainly because of their inability to
project interference fringe patterns onto those eccentric regions of the
retina, i.e. they do not have a sufficiently wide field. Resultantly,
today, testing of peripheral vision is accomplished by flashing light at a
variety of locations oblique to the patient's line of sight. The patient's
ability or inability to detect those flashes at different points within a
peripheral field of view is directly related to the size of the patient's
visual field, but not necessarily to the acuity of the peripheral or
eccentric regions of the retina. Therefore, such testing does not really
provide an accurate indication of peripheral acuity.
Measurements of neurological response to spatially and temporally varying
visual stimuli are useful in diagnosing other problems including
retinal-neurological dysfunction. During testing, evoked potentials from
the brain are produced in response to a visual stimulus. The most common
visual stimulus today is a phase-reversing checkerboard or bar pattern
displayed on a television screen.
All the foregoing tests and measurements using many current techniques
require clear ocular media with reasonably normal refractive properties.
If the media are not clear, as in the case of a patient afflicted with
cataracts, the tests are not always valid. However, if a procedure were
available for performing these tests independently of the opacity and
refractive properties of the eye, better diagnosis could be made.
Generally, laser produced interference fringe patterns provide a basis for
instruments that measure retinal acuity because they can be projected onto
the retina independently of ocular refractive errors and minor ocular
media opacities.
There are two basic methods for producing fringe patterns: (1) an
interferometric technique that utilizes interference phenomena, and (2) a
Moire technique that utilizes shadow casting and/or pattern
multiplication.
There are a wide variety of measuring and testing procedures that utilize
interference fringe patterns and there are many ways to produce and
control interference fringes. Generally, an interference fringe pattern is
produced when at least two coherent beams of light are brought together
and interact. When two coherent beams interact, they destructively
interfere to produce dark spots or bands and constructively interfere to
produce bright spots or bands.
Moire fringes are produced when two similar, geometrically regular patterns
consisting of well defined clear and opaque areas are juxtaposed and
transilluminated. Some examples of geometrically regular patterns used to
generate Moire fringes include (1) Ronchi rulings, (2) sets of concentric
circles, and (3) radial grids. The generation of Moire fringes can be
considered as shadow casting; that is, the shadow of the first pattern
falling onto the second pattern produces the Moire fringes. The
mathematical function describing Moire fringes is obtained by multiplying
the intensity transmissions or irradiances of the overlapped geometrically
regular patterns.
Fringes generated by both interference and Moire techniques are used by
ophthalmologists for testing retinal acuity. In one such apparatus, light
from a laser is divided into two coherent beams by an optical element
consisting of two adjoined dove prisms. These two beams are converged and
directed into the eye where they interact to produce an interference
fringe pattern on the retina.
In another apparatus used in the field of ophthalmology, a laser source and
an ordinary Ronchi ruling form an interference fringe pattern. The laser
source produces a laser beam that is directed to the Ronchi ruling. The
Ronchi ruling splits the incident beam into multiple coherent beams of
widely varying strengths. It is necessary to use complicated motions of
numerous optical and mechanical components to select only two coherent
beams and to control the spacing of interference fringes eventually
projected onto the retina. In yet another ophthalmic apparatus, two Ronchi
rulings are used that produce Moire fringes which are eventually imaged
onto the retina.
Certain disadvantages exist in apparatus that utilize the interferometric
techniques to form fringe patterns in ophthalmic applications. For
example, in such apparatus the two light beams generally travel through
different light paths that contain distinct optical elements. If the
elements in each path are not matched optically, aberrations distort the
fringe pattern. Matched optical elements can eliminate the aberration
problem; however, they significantly increase the overall expense of the
apparatus. Moreover, this apparatus is subject to various outside
influences, such as vibration and thermal change. These influences can
cause fringe pattern motion or noise and lead to improper measurements.
Moire techniques also have many limitations. When small spacings and high
accuracies are required, the geometrically regular patterns used to
generate Moire fringes are quite difficult and expensive to produce. In
applications where one ruling moves next to a fixed ruling, the spacing
between the rulings must be held constant or errors result. Also, Moire
fringes are localized, i.e., they exist in a very small region of space,
and additional optical components are often required to image the Moire
fringes into desired regions.
Recently, an amplitude grating and a spatially coherent,
quasi-monochromatic light source have been used to generate interference
fringes. An amplitude grating is a generally transparent to
semi-transparent media whose opacity is altered in accordance with some
spatially periodic pattern. An amplitude grating "breaks up" or diffracts
an incoming beam of light into a series of diffracted cones or orders. The
strength, or amount, of light in each order depends upon the exact shape
of the periodic opacity of the amplitude grating. Although various
diffracted orders could be approximately the same strength, scalar
diffraction theory for a thin amplitude grating predicts that the dominant
strength will lie in the zero order undiffracted light and that the
strength of other diffracted orders will vary. Indeed, practical
applications bear out this prediction.
In U.S. Pat. No. 3,738,753, issued June 12, 1973, Huntley proposes to pass
light from a source through an amplitude grating to produce different
order cones of diffracted light: for example, zero order and first order
cones. To compensate for the different intensities, the diffracted light
cones are reflected back through the grating. After the second passage
through the grating, the zero order cone of the reflected first order cone
and the first order cone of the reflected zero order cone have equal
strengths and are combined to form a high contrast interference fringe
field. This double pass system is quite stable because it closely
approximates a common path interferometer. In a common path
interferometer, the interfering beams traverse the same optical path.
Therefore, perturbations affect both beams simultaneously and do not
distort the output fringe pattern which is sensitive only to differences
between the two optical paths. However, problems in such a double pass
system do occur because it is difficult to control grating substrate
aberrations and mirror-grating separation.
Further improvements have been made with the advent of holographically
produced amplitude gratings. Holographic amplitude gratings are produced
by exposing a high resolution photographic emulsion to the precise
interference pattern of a laser two-beam interferometer. During ordinary
photographic processing, the photosensitive silver halide in the emulsion
converts into opaque metallic silver to form the amplitude grating.
In an application of one such holographic grating, a double frequency
holographic grating produces a so called "shearing" pattern. See U.S. Pat.
No. 3,829,219, issued 1974 to Wyant, and U.S. Pat. No. 4,118,124 issued
Oct. 3, 1978 to Matsuda. This grating is produced by sequentially exposing
a single photographic emulsion to a first laser interference pattern of a
first spatial frequency, f.sub.1, and then to a second laser interference
pattern of a second spatial frequency, f.sub.2. Equal amplitude
transmission modulations at both frequencies f.sub.1 and f.sub.2 are
achieved by adjusting the exposure to the first and second laser patterns.
Ordinarily, the two sequential exposures are identical, but if f.sub.1 and
f.sub.2 are very different or if one laser pattern is in red light and the
other is in green light, the sequential exposures must be compensated for
the spectral and frequency responses of the photographic plate. These
exposure adjustments to achieve equal amplitude transmission modulations
in f.sub.1 and f.sub.2 are usually done by trial and error.
Upon illumination with spatially coherent, quasimonochromatic light, this
double frequency grating produces two first order light cones of equal
strength, one light cone being associated with each of the f.sub.1 and
f.sub.2 frequencies. These two first order light cones interact to form a
very stable, high contrast fringe pattern. Such a double frequency
holographic shearing interferometer also is a common path interferometer.
It is simple to construct. However, in this interferometer it is necessary
to separate the zero order cone from the interacting first order cones.
This separation requirement limits the f/number of the input light cone
and the amount of shear obtainable. Moreover, if the two first order cones
have high diffraction angles an astigmatic distortion of the output fringe
field exists. In addition, the efficiency, or ratio of output fringe field
power to input power, is only about 2%.
For many years people have bleached photographically recorded amplitude
gratings to obtain "phase gratings". One basic type of such bleaching,
known as volume bleaching, chemically converts the opaque silver in the
photographic emulsion into a transparent, high index silver salt. A second
type of bleaching, known as tanning, chemically removes the developed
silver within the emulsion and leaves a void. A tanned phase grating has a
corrugated surface. Whereas an amplitude grating selectively absorbs
light, a bleached phase grating selectively introduces phase delays across
the input light beam. As a result, a phase grating is much more efficient
than an amplitude grating; that is, the ratio of first order power to
input power is greater.
However, bleached gratings are generally characterized by substantial
problems. They are very noisy and also may deterioriate physically back
into amplitude gratings upon extended exposure to light. Bleached gratings
also have a lower spatial frequency response than amplitude gratings.
Although volume bleached gratings are less noisy and have a higher spatial
frequency response than their tanned counterparts, they generally are
weaker and less efficient.
The efficiency of a volume bleached grating can be increased by increasing
its thickness. However, any substantial increase in thickness drastically
changes the basic diffraction properties of the grating. Any amplitude or
phase grating can be considered optically thick when the optical thickness
of the emulsion is more than five times the grating spacing. A grating can
be considered optically thin if the optical thickness of the emulsion is
less than half the grating spacing. Properties of thick gratings are
accurately predicted by electromagnetic theory, while properties of thin
gratings are described by scalar diffraction theory. For example, a thick
phase grating output consists of only the zero order and one first order
diffracted cones. In addition, diffraction takes place only for a plane
wave input at a certain specified angle with respect to the grating. On
the other hand, a thin grating of the same spacing produces multiple
orders (i.e. the 0, .+-.1, .+-.2, .+-.3, etc. orders) with either a
spherical wave or plane wave input at an arbitrary angle with respect to
the grating.
Distinctions between optically thin amplitude and optically thin phase
gratings are accurately predicted by scalar diffraction theory. When a
pure sinusoidal amplitude transmission perturbation exists in a thin
amplitude grating, only the zero and .+-.1 diffracted orders exist. When a
pure sinusoidal phase perturbation occurs in a thin phase grating, many
orders (e.g., the 0, .+-.1, .+-.2, .+-.3, and other orders) are observed.
The strengths of the phase grating orders are proportional to the
normalized Bessel functions [J.sub.n (m/2)].sup.2, where n is the order
number (e.g., n equals 0, .+-.1, .+-.2, . . . ) and m is the strength, or
magnitude, of the phase perturbation in radians. When the amplitude
grating perturbation departs from a pure sinusoidal form, additional
diffracted orders are generated. The strengths of these additional orders
are directly related to the strengths of the Fourier components associated
with the grating perturbation function.
With a phase grating, the diffracted orders associated with a
non-sinusoidal phase perturbation are predicted by convolving the
individual outputs from each Fourier component of the phase perturbation.
Such a multiple convolution reveals complicated phase relationships
between multiple orders associated with just one particular Fourier
component. In addition, diffracted orders corresponding to sum and
difference frequencies are generated when the phase perturbation consists
of more than one fundamental spatial frequency. For example, one might
consider bleaching the previously discussed double-frequency holographic
grating to improve its poor efficiency. Although bleaching will increase
the overall efficiency of such a grating, the bleached grating, in
accordance with the convolutional operation, produces sum and difference
frequency diffraction cones that are in addition to and that interact with
the desired fundamental frequency diffraction cones. It is then possible
for the sum and difference frequency diffraction cones to destroy the
fringe field.
SUMMARY
Therefore, it is the object of this invention to provide an improved
holographic phase grating for producing a high contrast interference
pattern that is useful in ophthalmic applications.
Another object of this invention is to provide an improved holographic
grating that is useful in the testing of retinal acuity.
Still another object of this invention is to provide an improved
holographic phase grating that is useful in the testing of peripheral
vision.
Still yet another object of this invention is to provide an improved
holographic phase grating that is useful in the testing of visual evoked
responses.
Yet another object of this invention is to provide apparatus for testing
retinal acuity.
Yet still another object of this invention is to provide apparatus for
testing peripheral retinal acuity.
Yet another object of this invention is to provide apparatus for testing
visually evoked responses.
A further object is to provide apparatus in the nature of a focimeter for
measuring the focal length of an optical element such as a lens and
testing it for aberration.
In accordance with my invention, I use a single frequency holographic phase
grating in ophthalmic testing equipment. A spatially coherent light source
illuminates the grating to produce diverging diffractions, in conical or
rectangular form, of different order. By "different order", I mean
diffractions whose order numbers have different absolute values. In two
diffractions of different order, the diffractions have equal strength and
overlap thereby to produce a bright, high constrast, low noise
interference pattern. I place a focusing element between the light source
and grating to produce a spatially coherent source of light at a focal
point that is slightly displaced from the grating. Other optical elements
positioned in the resulting interference fringe field project the
interference pattern through the eye and onto the retina.
Various controls in the optical path enable many ophthalmic measurements,
including visual evoked response measurements and visual acuity
measurements in the central and eccentric regions of the retina even in
the presence of corneal or eye lens opacities known as cataracts, or other
refractive effects.
The above and further objects and advantages of this invention may be
better understood by referring to the following description taken in
conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a diagram that illustrates the apparatus for producing a
holographic grating in accordance with this invention;
FIG. 2 is a chart that depicts the various basic steps for processing a
holographic grating in accordance with this invention;
FIG. 3 is a diagram of an interferometer constructed in accordance with one
aspect of this invention for producing fringe patterns;
FIG. 4 is a diagram of apparatus constructed in accordance with this
invention for measuring retinal acuity;
FIG. 5 depicts typical fringe patterns that are produced in the retinal
acuity apparatus shown in FIG. 4;
FIG. 6 is a diagram for an alternate embodiment of retinal acuity testing
apparatus constructed in accordance with this invention;
FIG. 7A is a perspective view of a retinal acuity testing apparatus
constructed in accordance with this invention;
FIG. 7B is a detailed perspective view of the apparatus shown in FIG. 7A
with the housing partially removed;
FIG. 8 depicts, in diagrammatic form, a retinal acuity tester that includes
a white-light source;
FIG. 9 depicts, in diagrammatic form, ophthalmic testing apparatus for
measuring retinal acuity in the central and peripheral regions of the eye,
with constant and variable contrast and for providing visual stimuli for
visually evoked response measurements;
FIG. 10 is a diagram that is useful in understanding peripheral acuity
measurements;
FIG. 11 depicts, in diagrammatic form, another embodiment of ophthalmic
testing apparatus for measuring acuity and for providing stimuli for
visually evoked response measurements with variable contrast; and
FIG. 12 depicts, in diagrammatic form, another embodiment of ophthalmic
testing apparatus for providing another type of stimuli for visually
evoked response measurements.
DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS
A. Holographic Grating
FIG. 1 depicts, in diagrammatic form, the arrangement of apparatus
necessary for exposing a photographic plate during the production of a
holographic phase grating. The holographic phase grating produced in
accordance with the arrangement shown in FIG. 1 and the procedures
outlined in FIG. 2 are essential to the operation of the diverse
embodiments of the invention that are shown in the other figures.
Specifically, this apparatus includes a laser source 10 which directs
light along an axis 11. The other apparatus in FIG. 1 splits the light
into parts that travel over two separate paths and are then brought back
together to expose a photographic plate 12.
A conventional beamsplitter 13 separates the light into two parts. A first
part travels along a first path that includes mirrors 14 and 15 for
reflecting the light into an objective lens and pinhole 16, thereby to
produce a spherical wave that emanates from a point source at the pinhole.
The wave appears in a cone 17 and is directed toward the photographic
plate along an axis 18. The second path established by the beamsplitter 13
includes a mirror 20 and an objective lens and pinhole 21 that produce a
spherical wave cone 22 that emanates from a point source at that pinhole
along an axis 23. The light waves from these two point sources combine;
they destructively interfere to produce dark bands and constructively
interfere to produce bright bands at the photographic plate 12.
The photographic plate 12 mounts on a rotary table which positions the
photographic plate 12 and accurately establishes an angle .theta. between
the axes 18 and 23. The spatial frequency, .delta., of the interference
pattern at plate 12 is closely approximated by the equation
##EQU1##
where .lambda. is the laser wavelength. Although the fringes produced at
the plate 12 are slightly hyperbolic, they are excellent approximations to
rectilinear bands and therefore are shown as such in various figures.
Increasingly better approximations to rectilinear bands are achieved by
increasing the distance along the axes 18 and 23 between the plate 12 and
the pinholes 16 and 21, respectively.
The apparatus diagrammed in FIG. 1 has been used to manufacture gratings
having the desirable properties that characterize my invention. The
equipment is simple and relatively inexpensive. For example, the laser 10
can comprise a TEM.sub.00 mode laser; the beamsplitter 13, a conventional
variable density beamsplitter that enables the intensity of the two beams
to be equalized. The mirrors 14, 15 and 20 are standard planar mirrors.
The objective lens comprises a conventional 10.times. microscope
objective, and the pinhole matches that objective lens. The distances 18
and 23 are approximately 2 meters. With this specific arrangement, I am
able to obtain a 500 line-per-millimeter interference fringe pattern over
a 3".times.3" area with maximum fringe displacement error of about 0.00254
millimeters.
Once the apparatus in FIG. 1 is arranged, the emulsion on the photographic
film can be exposed to the interference pattern as shown as Step 1 in FIG.
2. During this exposure step, certain controls must be exercised to assure
a holographic grating of good quality. For example, the exposure should be
made in an environment that is not subjected to vibrations. Thermal
disturbances should be minimized as any air flow between the beamsplitter
13 and the photographic plate 12 can distort the resulting fringes. In
applications where very high densities and minimal distortions are
required, the distances along axes 18 and 23 must be increased to 5 or
even 10 meters. Precise determinations of .lambda. and .theta. must be
made. Although this basic apparatus can be used to produce highly accurate
holographic phase gratings, the maximum accuracy ultimately then will be
determined by the accuracy of angular measuring equipment, the stability
of the single frequency laser, the optical table stability, and the
atmospheric and thermal controls that are exercised.
In order to produce a phase grating with special properties that enable the
construction of the various disclosed embodiments, it is first necessary
to produce an amplitude grating. Given the various properties of
commercially available photographic emulsions and developers, a thin
emulsion photographic plate and a chemically compatible developer are
selected. A process of heavily overexposing and underdeveloping the
emulsion reduces the optical thickness of the processed emulsion to a
fraction of its original physical thickness. Thus, by utilizing the
controls set forth in Steps 1 and 2 of FIG. 2, one produces an amplitude
grating characterized by having:
1. an optically thin emulsion conforming to scalar diffraction theory;
2. a specific form for the absorbtion function which converts to a
correspondingly specific phase transmission function after bleaching; and
3. a specific amplitude or strength of the absorbtion function which
converts to a specific peak-to-peak phase modulation after bleaching.
Specific plate types, exposures, development times and developers are
discussed later.
Once the development of step 2 is complete, the photographic plate is
washed in an acid short-stop solution in Step 3. The solution contains an
acid hardener. A two-minute treatment in a hardening bath produces
acceptable results.
In Step 4, the emulsion of the photographic plate is fixed and hardened. A
standard fixing bath and acid hardener have been used successfully, the
plate being immersed in the bath for about ten minutes.
Next (Step 5), the emulsion is prewashed for thirty seconds and
hypo-cleared in a hypo clearing bath for about two minutes. In Step 6, the
emulsion is washed (e.g., twenty minutes ih filtered water) and then
soaked in a methanol bath until all residual sensitizing dye is removed
(Step 7). Once the methanol bath has been completed, the plate is dried in
a light blow air drying operation.
All the foregoing steps are conventional photographic processing steps that
utilize commercially available chemicals. Upon completion of Step 7, an
amplitude grating has been produced. Steps 8 and 9 then convert this
amplitude grating into a phase grating having the desired characteristics.
More specifically, after the photographic plate is dried thoroughly in step
7, it is bleached during Step 8 in a bromine vapor until the plate is
clear. Once the bleaching operation has been completed, the plate is
rinsed in a methanol bath to remove residual Br.sub.2 and dried thoroughly
by a light blow air drying operation in Step 9.
It now will be beneficial to discuss certain characteristics of these
holographic phase gratings that are particularly desireable. First, the
exposure and development times and the emulsion have been chosen to
produce "thin" gratings. As a specific example, I have made 393.7
line-per-millimeter gratings on Kodak 131-01 plates according to the
foregoing processing procedure using an average exposure of 200
ergs/cm.sup.2 and a development time of 15 seconds in standard Kodak D-19
developer at 80.degree. F. Uniform development is achieved by using a
large development tank and rapid manual agitation of the plate. After
complete processing in accordance with the steps of FIG. 2, the resulting
thin phase grating diffracts both input spherical waves as well as input
plane waves; as previously stated, a thick grating diffracts only input
plane waves incident at a particular angle with respect to the grating.
Measurements have shown that a thin phase grating manufactured according to
the foregoing process has a pure sinusoidal phase transmission function
whose peak-to-peak phase delay produces equal strength zero and .+-.1
diffraction orders. The 200 ergs/cm.sup.2 exposure produces an average
amplitude transmission of approximately 0.45 for the developed, but
unbleached, Kodak 131-01 plates. Experimental data has confirmed that a
pure sinusoidal phase transmission function is maintained when the thin
grating has an average amplitude transmission of 0.5 or less in its
developed but unbleached state. The strength or peak-to-peak phase delay
of the final phase grating is adjusted by controlling the initial exposure
(Step 1, FIG. 2) within the limits set by an average amplitude
transmission of 0.5 (measured after Step 7 in FIG. 2). A very weak phase
grating produced with low exposure levels exhibits a strong zero order
diffraction, a weak first order, and an even weaker second order
diffraction. Stronger gratings produced with higher exposure levels
exhibit increasingly more powerful first and second order diffractions and
decreased zero order diffraction. Equal strength zero and .+-. 1
diffraction orders or equal strength zero and .+-.2 diffraction orders are
achieved by a trial and error adjustment of the initial exposure.
The advantages of such a thin phase grating that produces two different
diffraction orders of equal strength will now become apparent in the
following discussion of an interferometer that utilizes such a phase
grating.
B. Interferometer
Referring now to FIG. 3, an interferometer is depicted in schematic form
that includes a helium neon laser 30 which directs light along an axis 31
to a negative lens 32. The negative lens 32 expands the beam slightly so
that it completely fills a microscope objective 33. The microscope
objective 33 focuses this light at a focal point FP displaced a distance
Z.sub.1 from a holographic grating 34 constructed as described above. The
laser 30, negative lens 32 and microscope objective 33 constitute a source
of a quasi-monochromatic diverging spherical wave that emanates from the
focal point FP. In one embodiment, the cone from the focal point FP is an
f/2 cone.
When the spherical wave from the point source at the focal point FP strikes
the grating 34, it produces a number of cones of diffraction. According to
scalar diffraction theory, the strength of the diffracted cones is
governed by the Bessel function [J.sub.n (m/2)].sup.2 where n is the
diffraction order number and m is the grating transmission function
peak-to-peak phase delay in radians. The previously specified exposure and
development times for a Kodak 131-01 plate yield a value of m=2.870 at
.lambda.=6328 .ANG.. The zero and first order diffraction cones are of
equal intensity because [J.sub.0 (1.435)].sup.2 =[J.sub.1 (1.435)].sup.2.
Moreover, the diffraction angles are such that the zero order cone
overlaps both first order cones, while the first order cones merely abut
each other. At some point at a distance Z.sub.2 from the grating 34, an
output such as is shown in FIG. 3 is produced. The zero order cone appears
as planar circle 35; first order cones appear as planar circles 36A and
36B. Areas 37A and 37B are areas of overlap and the fringes are produced
in those areas. Moreover, the fringes in the areas 37A and 37B are out of
phase with each other. Thus, if the centrally located fringe in area 37A
is a dark band, the corresponding fringe in area 37B is a light, or
bright, band. By "light" and "dark" bands, I do not mean bands having the
same intensity across the band, as the bands are shown in the drawings.
The fringe intensity actually varies smoothly and is proportional to the
square of a sine function, although the eye may perceive distinct
alternating bands under some illumination conditions.
| | |
