 |
Description  |
|
|
FIELD OF THE INVENTION
This invention relates in general to optical instruments and methods, and
more particularly to an instrument for scanning a surface or other
structure with an optical beam, detecting the light emitted from the
structure, and generating a two-dimensional representation of an image of
the structure.
BACKGROUND OF THE INVENTION
In the art of optical instruments, it is known to scan a surface to be
imaged with a small light source, collect the light reflected from the
illuminated spot and direct it to a detector which provides an output
signal varying in time in correlation with the scanning of the illuminated
spot across the surface. The detector output can be stored in a permanent
storage medium or provided directly to a scanning display device, such as
a television raster or a cathode ray tube display. By synchronizing the
scanning operation of the illuminating source with the scanning of the
display signals, a two dimensional image is produced.
One such instrument is a scanning ophthalmoscope which produces an image of
the fundus of the eye. It has been found that the use of a laser light
source provides improved imaging in an ophthalmoscope. A laser scanning
ophthalmoscope is described in U.S. Pat. No. 4,213,678. One problem
associated with ophthalmoscopes of the type described in U.S. Pat. No.
4,213,678 is that the light collected, at the time the laser is
illuminating a particular area on the retina, includes not only light
reflected directly from that area, but also light scattered from other
surfaces and materials within the eye. This scattered light can cloud or
fog the image, since it represents light contributions from other than the
specific illuminated area. In an ideal system, each small illuminated area
of the target object being examined produces a corresponding image area in
the output display, with a brightness or intensity related only to light
reflected directly from that target area. In some situations, on the other
hand, the scattered light by itself, to the degree that it can be
separated from the light directly reflected from the iluminated target
area, is useful for diagnostic purposes.
In a device as described in the noted patent, the entrance pupil for the
scanning laser beam has a small cross sectional area within the pupil of
the eye, typically 0.5 mm in diameter, whereas the exit aperture for the
reflected light is the overall pupil of the eye, which typically is nine
mm in diameter. The detector is placed in a plane conjugate to this exit
aperture. In the embodiment described in the patent, the scanning is
effected by deflection galvanometers. The horizontal galvanometer is
driven at 15.75 kHz. in order to match the horizontal scan frequency of a
conventional television sweep, which preferably is used to display the
output image. The vertical galvanometer is driven at 60 Hz to produce 525
lines per frame of the output image, again corresponding to the generation
of a conventional television raster.
In a scanning ophthalmoscope of this type, the resolution in the raster
display of the retinal image directly corresponds to the cross sectional
area of the laser spot as it scans the retina. The contrast of the
ultimate image depends, at least in part, upon the proportion of light
received by the detector which is directly reflected from the illuminated
area. Thus, to the extent that scattered light indirectly reaches the
detector at the same time as it receives the light directly reflected from
the illuminated area, the image is fogged and the contrast is reduced. The
term "reflected" is used herein in a broad sense to refer to all optical
energy returned by the target structure, it hence includes returned
optical energy that results from both specular and diffuse reflection.
One technique used in some optical instruments to improve contrast for
images of this type may be termed double scanning. According to this
technique, the optical system is arranged to provide that the light
reflected from the illuminated target area is selected with a
scanning-like action related to the scanning of the incident illumination
in such a manner that, at a given instant, the reflected light received by
the detector is only that which is reflected from the illuminated target
area. In effect, as applied to an ophthalmoscope, the fundus conjugate
plane thereby allowing discrimination, at the conjugate retinal plane,
between the light directly reflected from the retinal locus and that
scattered either anteriorly or positiorly, i.e. within the retina. This
approach, however, has been deemed to be unsuitable for an instrument like
the laser ophthalmoscope of the type described, because in that instrument
the exit aperture for the reflected light is so large that the returning
reflected beam was deemed to require an unduly large scanning element.
Since, at the driving frequencies associated with a television raster, a
deflection galvanometer is limited by mass considerations to a very small
surface, in the order of three millimeters, a reflection galvanometer
large enough to encompass the returning image has been deemed not
feasible.
Another deflection element which has been used for scanning optical
instruments is a multifaceted rotating polygon, which would have to rotate
at sufficiently high speeds to produce a horizontal scan matching the
television frequencies. However, once again the size of the facet required
to encompass the image received from the eye's exit aperture is
prohibitively large in terms of fabricating a polygonal reflector to
rotate at the required speeds.
The acousto-optical deflector is also not available in a form considered
suitable for the reflected beam in such an instrument, due to aperture
limitations.
OBJECTS OF THE INVENTION
It accordingly is an object of the present invention to provide an optical
system for producing a two-dimensional representation of the reflection
characteristics of a scanned structure and having relatively high
resolution and contrast.
Another object of the invention is to provide an optical instrument having
double scanning, i.e. of both incident and reflected light, at high
frequencies such as are conventional in a television-type raster display.
It is also an object to provide an ophthalmological instrument for
providing a two-dimensional representation of reflection characteristics
of structure within an eye essentially in response only to light reflected
from the eye structure in a selected manner. In one particular embodiment,
the image is created in response essentially to directly reflected light;
an in another embodiment in response to indirectly reflected light.
It is another specific object of this invention to provide an
ophthalmological instrument for providing a two-dimensional representation
of the reflection characteristics of the fundus of an eye wherein the
contrast of the ultimate image is enhanced by enabling essentially only
directly reflected light to generate that image.
It is another object of the invention to provide a confocal scanning
ophthalmoscope utilizing an infrared laser beam to scan the eye fundus.
It is still another object of the invention to provide a confocal scanning
ophthalmoscope which produces a graphic image on the retina during the
scan.
Other objects of the invention will in part be obvious and will in part
appear hereinafter.
SUMMARY OF THE INVENTION
It has been found, in one practice, that a double scanning optical
instrument can be constructed utilizing a laser source and a multifaceted
polygonal reflector for horizontal scan, with a reflection galvonometer or
other scanning element for vertical scan, where the facet size in the
direction of scan for the polygonal reflector is necessarily small and the
reflected beam from the exit aperture of the system is substantially
larger than that facet dimension. In the illustrated embodiment described
below, the small facets of the polygonal reflector intercept less than 20%
of the reflected light from the exit aperture. However, unexpectedly,
under these circumstances the instrument attains a significant improvement
in contrast over a single scan system, despite the significant loss of
throughput.
It has thus been found that an optical instrument, of the type which
responds to light energy responsive to a scanned incident beam, can be
provided with double scanning with at least one scan element having such a
small size that the exit beam overfills it. That is, this scan element is
of such small size that it intercepts only a portion of the exit beam. In
spite of the resultant loss of exit beam energy, the double-scanning
instrument attains images having significant improvements over those of
prior instruments. An instrument according to the invention attains this
improved performance even when configured to have a large optical exit
aperture, as is often desired.
DESCRIPTION OF THE DRAWINGS
For a fuller understanding of the nature and objects of the invention,
reference may be made to the following description and the accompanying
drawing, in which:
FIG. 1 is a diagrammatic representation of one embodiment of a scanning
ophthalmoscope according to the invention;
FIGS. 2 and 3 are explanatory ray diagrams of optical beam features of the
embodiment illustrated in FIG. 1;
FIGS. 4 and 5 are explanatory ray diagrams of optical scan features of the
embodiment of FIG. 1;
FIG. 6 is an explanatory ray diagram of the embodiment of FIG. 1, where the
optical system includes a diaphragm stop and the detector is repositioned.
FIG. 6a is a view of the diaphragm stop of FIG. 6;
FIG. 7 is a diagrammatic representation of a modification to the embodiment
of FIG. 1;
FIG. 8 is a block diagram of a portion of the ophthalmoscope of FIG. 7;
FIG. 9 is a diagrammatic representation of another embodiment of a scanning
ophthalmoscope according to the invention.
FIGS. 10 and 10a are diagrammatic representations of a telescope magnified
for insertion in any of the embodiments illustrated.
FIG. 11 is a diagram of the relationship between the axis of rotation of
the scanning element of the embodiment of FIG. 1 and the allowable
movement of the ophthalmoscope apparatus.
FIG. 12 is a block diagram of an electronic circuit employed in the
practice of the invention, and
FIG. 13 is a block diagram of another electronic circuit employed in the
practice of the invention.
DESCRIPTION OF THE PREFERRED EMBODIMENT
FIG. 1 shows an embodiment of the invention in the form of an
ophthalmoscope 10. A laser illumination source 11 produces a narrow
incident light beam 12 which passes through a shaping lens system 13 which
produces a slightly converging beam that impinges on a small turning
mirror 14. The mirror 14 directs the incident laser beam onto facets of a
multi-faceted rotating polygonal reflector scanner 15, which provides a
horizontal scanning motion of the incident beam. The incident beam is
reflected from this first stage scanning element onto a focusing mirror
16, which directs the beam onto the reflecting surface of a galvanometer
reflector scanner 17 to produce a vertical scanning motion. From the
galvanometer reflector scanner 17, which is a second stage scanning
element, the laser input beam is directed onto a second focusing mirror
18, for focusing it onto the fundus 19a of the eye 19 of a subject. The
incident beam enters the eye at the crystalline lens 19b.
The reflected light from the fundus 19 is directed back over a common
portion of the foregoing optical input path, which includes focusing
mirror 18, the second stage scanner 17, focusing mirror 16 and the first
stage scanner 15. All of these common elements are mirrors and hence do
not contribute reflections of the input beam back to the detector as noise
background. The reflected output beam from the first stage scanner 15 in
large part passes by the turning mirror 19 and hence separates from
further traverse along the incident optical path. The output beam instead
is directed through a focusing lens 20 and onto an optical detector 21.
The detector 21 is electrically connected to an electrical instrumentation
unit 22 which provides electrical control signals to the laser source 11
and electrical drive signals to the scanning deflection elements 15 and
17. In essence, the instrumentation unit provides synchronization of the
signals received at the scanning elements 15 and 17 so that the temporal
order of the signals produced by the detector 21 can be correlated with
the location of the scanned incident laser beam on the surface of the
fundus. The control and synchronization which the instrumentation unit
provides enables a two-dimensional display device 23, such as a television
raster device, to form a two-dimensional display of an image of the eye
fundus 19a, in response to the electrical signal which the detector
produces in response to the reflected optical energy it receives. The
detector signal may be applied to a long term storage element 24, such as
a video tape recorder, for subsequent readout and display. For a
description of a suitable electrical timing and control circuit, reference
is made to U.S. Pat. No. 4,213,678 which is incorporated herein by
reference.
THE LASER GENERATOR
The laser 11 can be any suitable laser light source which provides emission
at frequencies yielding appropriate contrast for the fundus, or other
target. Typically, the laser 11 is an Argon-Krypton laser or Helium-Neon
laser operated at a power level to produce an illumination irradiance of
one hundred microwatts per square centimeter at the fundus. The laser 11
may also be selected to emit in the infrared wavelength region to provide
a scanning beam which does not require that the eye pupil be medically
dilated to obtain an image of the fundus. For color imaging two lasers of
different wavelengths may be employed and converted into a single beam
with a dichroic beam splitter.
The laser beam, when it is emitting in the visible wavelength can also be
arranged to present a graphic image, such as a cross in the scanning of
the retina. FIG. 7 illustrates a modification to the embodiment of FIG. 1
in which additional elements are inserted between the laser 11 and the
horizontal scanner 15. This embodiment includes an acousto-optic modulator
(AOM) 30 for performing the graphic imaging function. The AOM receives a
control input from a program control unit 34, which is typically a
computer programmed to provide a signal timed to direct the laser beam
emerging from the AOM away from the scanning path, thus blanking the
scanning beam appropriately, to produce the image, a suitable computer
being an IBM PC-XT made by International Business Machines, Yorktown
Heights, N.Y. with a Revolution 512.times.8 graphics peripheral card with
gen lock, made by Number Nine Computer, Cambridge, Mass. A program
available for the graphic control is Media Cybeunetics' Halo, by Media
Cybernetics of Takoma Park, Md.
A visible graphic image may also be provided when an infrared laser is
employed, by utilizing an incandescent light beam incident on the AOM. The
incandescent source will have sufficient intensity to stimulate the
patient's retina but will not affect the scanned output image.
Prisms 32 are placed in the beam between the laser and the AOM and afer the
AOM to allow lasers of different wavelength to be used, while preserving
the same Bragg angle relationship for the different wavelengths within the
AOM to maintain the output beams from the AOM on the same optical axis.
FIG. 8 is a block diagram of the acousto-optic modulator.
The acouto-optical modulator includes a driver unit 40 coupled to a
transducer unit 41. The driver unit 40 includes an RF oscillator 42
typically operating at 40 MHz, followed by a buffer 46 which couples the
oscillator to a balanced modulator 47. The output from the balance
modulator 47 is coupled through an RF power amplifier 48 as the modulated
RF output to the transducer element 41. The transducer element 41 is
typically a glass crystal having Piezoelectric elements bonded to it to
produce acoustic waves in the glass crystal. Optical waves incident on
this crystal are then diffracted when the balanced modulator provides an
output and remain undiffracted when there is no output from the balanced
modulator, that is when the output from the driver 40 is blanked. The
driver and transducer unit is commercially available from IntraAction Inc
of Bellewood, Ill., under the trade designation AEM40 & MOP402B. In the
conventional acousto-optic modulator a video input is provided to the
balanced modulator 47 thus controlling the output signal to effect the
optical modulation, control signals are applied to this input from the
graphics program control. In this invention, however, a blanking input to
the balanced modulator must be presented at times, not directly associated
with the presentation of the graphics, in order completely to turn off the
laser beam during the retrace of the display raster. This is done so that
the laser beam does not impinge upon the patient's eye during this period,
thus avoiding unnecessary irradiation of the patient's eye on
distractingly visible retrace lines. This "absolute blanking" does not
suffer from the requirement of careful adjustment typical of the balanced
modulator circuit, but cannot achieve gray-scale modulation, since it is
essentially on or off. In order to ensure that the inputs from the
graphics control to the video input of the balanced modulator do not
interfere with this retrace blanking, a retrace blanking signal is
provided directly to buffer 46 to decouple the RF oscillator output 42
from the balanced modulator 47, thus disabling the driver unit 40 during
this period.
The AOM diverts some of the beam energy into a first order (Bragg
diffraction) beam at an angle typically about 15 mradians from the zero
order beam. Either the original beam or the diffracted beam can be used to
form the flying spot on the retina and its intensity is controllable over
about three orders of magnitude by th AOM drive. The Bragg diffraction on
which the modulation depends is from acoustic waves in a glass, of
frequency 40 to 100 MHz. Two complications occur: because this is
diffraction, it is inherently chromatic; and modulation of a high
frequency carrier introduces other frequency components.
As above described the chromaticity is compensated with prisms 32 placed
around the AOM. This brings both red an green beams to the glass at their
preferred Bragg angles. The second prism is after the AOM to cause the
beams to exit together, but the two prisms can be combined into one
without serious problems. Minor adjustment at the combining dichroic beam
splitter brings the two rasters into perfect alignment.
The trouble caused by the modulation itself is more subtle. When the RF
carrier of, say, 40 MHz is turned off or on, lower frequencies are present
for a few cycles. Lower frequencies deflect the beam at smaller angles,
and a few cycles may well be a whole pixel. So, if the deflection is
perpendicular to the fast (horizontal) scan direction, the beam moves off
the raster line as it is turned off. In this orientation, all the line
segments acquire little bends. This problem is solved if the AOM
deflection angle is horizontal. The bends are still here, but they stretch
the line segment out or tuck back into it, so that the only perception of
them is that the leading edge of the segment is slightly softer and the
trailing edge slightly sharper than expected.
THE INPUT OPTICAL SYSTEM
The purpose of the input optical system is to scan the fundus with a narrow
optical beam to sequentially illuminate small segmental areas across the
fundus surface in a known pattern so that the reflected light detected in
time sequence can be electrically converted to a two-dimensional
representation of the reflection characteristics of the fundus. In one
illustrative instrument, the input optical system forms the incident laser
beam with a cross sectional area of substantially 0.5 mm diameter at the
entrance pupil of the eye and focussed on the fundus to produce a spot
approximately twelve microns in diameter. The horizontal scanning motion
in the illustrated preferred embodiment is provided by a rotational
scanner which is shown in the preferred embodiment as a multi-faceted
polygonal reflector scanner 15 which is rotated by an electric motor at
speeds sufficient to produce a scanning frequency of 15.75 kHz to be
compatible with a TV sweep frequency. A polygon of (m) facets turns the
incident laser beam through a scan angle of 720/m degrees. Thus, if, for
example, there are twenty-four facets on the polygon, it must rotate at
40,000 rpm in order to generate the 15.75 kHz scanning frequency. In order
to rotate at this speed the moment of inertia of the polygon must be kept
small. In one practical embodiment, each facet is six mm wide. The
polygonal rotating reflector of the scanner 15 can be obtained
commercially from Lincoln Laser (Phoenix, Ariz., No. PO-24 (A grade, G
grade). A holographic disk scanner, scuh as made by Holotech, Inc., which
has spaced holographic facets may be substituted for this polygon
reflector. The scan angle can be changed optically by any of the
subsequent optical elements. One approach to modifying the field of view
is to set the vertical scanner to the same 28.8 degrees and then to modify
the whole field of view at once. Resolution depends on the ratio of scan
aperture to scan angle, so proper optical modification after the scan
preserves the original resolution. With an input beam diameter of about 1
mm at the polygon, the available resolution is 794 spots, of which only
667 are used because of the 84% TV duty cycle. This is, in fact, about all
that the available TV bandwidth can use. Once the resolution is fixed, at
the polygon, the field of view can be increased or decreased by simple
optical magnification. Increase of the field results in concurrent
decrease of the beam diameter at the pupil, and this increases the spot
size at the retina, so the resolution is unchanged.
One method of changing the field size is to add an external telescope. This
approach is illustrated in FIGS. 10 and 10a. FIG. 10 is an illustration of
the beam diagram of this system while FIG. 10a illustrates the return
reflection envelope. In FIG. 10 lenses 40 and 41, and are placed between
the eye pupil and the mirror 18. Lens 40 is typically a 28-diopter
ophthalmoscope lens and lens 41 may be a 14 diopter ophthalmoscope lens.
By reversing the position of the lenses the field of view can be made
smaller. If the distance between lenses 40 and 41 is adjusted to be
unequal to the sum of the lenses' focal length, then refractive errors in
the patient's eye can be compensated.
An advantage of this arrangement is that the telescope spacing adjusts beam
focus, providing an independent compensation for patients' refractive
error. This telescope system does however, produce reflections. Four
refractive surfaces intercept the incoming laser beam and reflect it back
to the detector. Some of these reflections can be blocked by appropriately
placed stops and some can be diminished or displaced by suitable choice of
surface bend and tilt. The residuum can at least be localized to one small
area of the picture. Since it is a moving picture, the clinician can
easily look around such a single reflection. Finally, in the tightly
confocal arrangement (small aperture at the retinal conjugate), the
reflections substantially vanish, so that it is only in the afocal mode
that they are a problem.
Another method of changing field size avoids these problems: element 18 can
be placed in a position to increse or decrease the field at the retina.
This is inconvenient to implement, but is a preferred embodiment if
reflections are a problem.
The vertical scanning motion in the illustrated preferred embodiment is
introduced by a deflection galvanometer 17 that provides a scan action
which corresponds with the television vertical scan of 60 Hz. Galvanometer
controls, such as those manufactured by General Scanning of Watertown,
Mass., are suitable for driving and controlling the position of the
galvanometer mirror. The mirror 17 can, for example, be a type G120D
General Scanning mirror.
With this structure and optical alignment in the instrument 10, the
illustrated laser beam of 0.5 mm in diameter which it produces underfills
each mirror facet of the polygon scanner 15, which, in the same
illustrative embodiment, is six mm wide. The beam scanning pivots about a
point in the plane of the eye's pupil.
The laser beam must be in focus at the retina, and the scan waist must be
located (approximately) at the eyes of the pupil. Under these
circumstances the spot size is appropriate for the available resolution,
and the image will appear in focus at the TV screen even if it is not in
focus at the confocal aperture. It is the focus of the incident beam which
determines the picture's resolution and the focus of the return beam (at
the confocal stop) which controls contrast. The fact that these controls
are largely orthongonal is what allows flexibility as to mode of view.
The turning mirror 14 preferably is a stationary mirror reflector. It is
small in size in order to produce a minimal shawdow in the output beam,
and hence preferably is only large enough to intercept the input beam
which the focusing element 13 directs, via the turning mirror, to the
first stage scanner 15. In the configuration shown the turning mirror acts
as the beam separator between the input and reflected return beam.
In the embodiment illustrated in FIG. 9 the laser beam is originally
directed toward the polygon scanner 15. A mirror 38 with a central hole
allows the laser beam to pass through it. The return reflected beam from
the scanner 15 is then reflected by the annular portion of mirror 38 to
the detector 21.
FIGS. 2 and 4 illustrate features of the input optical system. FIG. 2
represents the input beam with the scanners assumed to be stationary in a
neutral, non-deflecting, position. The narrow collimated incident beam 12
from the laser is, in this partial representation, shaped by the optical
elements 13, 14, 16 and 18, aside from the eye 19 of the subject. The
incident beam is in focus at the retina 19a. The limiting aperture formed
in this instance by the entrance pupil of the eye 19 is conjugate at the
scanners 15 and 17.
FIG. 4, which represents scan features of the input system, illustrate the
input beam instantaneously as a single ray which each scanning element
moves, as a function of time. The drawing shows, in effect, a time
exposure. In the illustrated envelope, the beams intersect at the scanners
and their conjugates, which, for the scanned input beam includes the
entrance pupil. The scan angle is the full angle of this envelope in the
plane of the scan.
The mirror 18 is large and spherical. Large, so that even at f/2 (for the
scan) the eye's pupil is far back from the optics. With human subjects
there are some inflexible dimensions. The mirror is spherical because no
aspheric is correct for both beam and scan systems at all points. That
constraint can be understood by noting that the beam on one side of this
mirror may be always collimated, no matter where it hits the mirror. So
the mirror must have everywhere the same local curvature--which implies a
sphere. Since the mirror is used off-axis, the scan system is then
astigmatic.
The scan system astigmatism can be corrected by adjusting the separation
between the horizontal and vertical scanners along the system's optic
axis. The small spherical mirror 16 is used as a relay between the two
scanners, for more flexibility. This mirror only focuses a line scan, so
it can be tilted in the orthogonal plane, contributing no astigmatism.
Both mirrors contribute coma, of course, so tilt angles are kept small.
THE OUTPUT OPTICAL SYSTEM
As noted, a major portion of the output optical system has a common optical
path with the input system. This common path includes both of the scanning
elements 15 and 17. In the illustrated instrument, it also includes the
two focussing elements 16 and 18. However, in the output system, the light
reflected from facets 15a of the rotating polygon scanner 15 passes around
the turning mirror 14 and is incident on the detector optical system,
which includes lens 20 and detector 21.
FIG. 3 represents the output beam without regard to the scanning elements
15 and 17, i.e. in the same manner as the representation in FIG. 2. As
illustrated, the reflected beam from the fundus has an exit aperture large
compared to the cross section of the scanning beam, preferably
substantially the entire pupil of the eye, with a diameter of as much as
nine mm. The image of this aperture at its conjugate plane also is nine
mm. Absent magnification, the reflected output beam from the illuminated
area on the fundus likewise is approximately nine mm in diameter at any
conjugate of the exit pupil, which is where the scan elements 15 and 17
are located.
In this configuration the central region of the eye's pupil is used as an
entrance pupil and the remaining annulus an exit pupil, thus conforming to
Gulstrand's principle. This means that the scanners, optically conjugate
to the pupil, need to be big enough to intercept that larger return beam.
For the vertical scanner which moves as a 60 Hz sawtooth, a 10-15 mm
mirror is suitable.
With the polygon, however, the available aperture (the facet) both rotates
with respect to the beam and moves across it. The incident 1 mm beam and a
6 mm facet on the polygon combine to give just about the 84% duty cycle
required for a TV raster. But the return beam may be as much as 15 mm in
diameter, overfilling the facet even at the center of its sweep. This does
lose light, but the facet is filled with signal light over most of its
duty cycle, and therefore a very uniform fraction of the light from the
annular exit pupil is recovered
The ophthalmoscope 10 can have a small entrance pupil, as described above,
due to the large radiance of the incident beam. The output beam, however,
has relatively low radiance, and hence the provision of this large output
pupil is desired to collect a maximal amount of output light energy. The
large exit aperture hence enhances the high efficiency of the instrument.
It also facilitates viewing a large portion of the eye fundus.
FIG. 3 also illustrates, with exaggerated scale, that the output beam
passes around the turning mirror 14, which hence casts a small shadow
generally of low significance.
It is desirable to separate the incident and return beams as close to the
polygon facet as possible in order to place the incident beam in the
center of the return beam and thus stop direct reflection from the cornea
(and spectacles if desired) from reaching the detector.
FIG. 5 represents scan aspects of the output beam, in the same manner as
the scanned input beam representation in FIG. 4. The scanned output rays
intersect, and the envelope of the scanned rays has minimal cross-section,
at the pupilary plane of the eye 19 and at the scanning elements 15 and
17; this is the same as for the scanned input beam, FIG. 4. The former is
at the plane of the exit pupil and the latter are at planes conjugate to
it.
As also illustrated in FIG. 3, the relatively large cross-section of the
output beam overfills each facet on the polygonal reflector scanner 15.
With the six mm facet width of the illustrated embodiment, this overfill
corresponds to a loss of throughput of approximately 80%. However, the
reflected output light beam which the scanners 15 and 17 direct to the
detector 21 is directly reflected substantially exclusively from the
illuminated segmental area of the fundus. The detector 21 hence receives a
minimal level of scatter or other unwanted light energy. These features
enable the instrument to attain a resultant improvement of contrast at the
detector which is unexpectedly high, and to yield a substantial
improvement in contrast in the resultant image.
The placement in the instrument 10 of the detector 21 at the retinal
conjugate plane, as apparent in FIG. 3, is advantageous because it allows
the detector to have a small aperture. Optical detectors of this type have
numerous advantages over large-aperture detectors. In particular, an
avalanche diode detector detector 21 is highly suitable for use as the
detector in this system.
FIGS. 6 and 6A illustrate an alternative embodiment in which a diaphragm
stop 26 is placed in the return beam path at the retinal conjugate plane
and the detector 21 is moved to the pupillary conjugate plane. In FIG. 6
the envelope of the return beam is diagrammed. For convenience the
diaphragm stop 26 can be formed as a disk with varying size openings (as
illustrated in FIG. 6a) so that the size of the diaphragm stop 26 may be
varied. Detectors are best placed at pupillary conjugates, since pupils
tend to be about detector size (a few millimeters) while the retinal spot
size is likely to be ten times smaller. In the invention described herein,
the detector of choice is a semiconductor, typically a 1 mm avalanche
diode with an integral amplifier such as RCA C30950E (RCA, Ste Anne de
Bellvue, Quebec, Canada). When this detector is placed at the pupillary
plane the retinal conjugate plane can be used for the placement of the
aperture which limits the amount of retinal surface the detector receives
light from. Since the retinal conjugate is a magnification (about ten
times) of the retina a 1 mm aperture at the retinal plane restricts the
retinal area seen to approximately 0.1 mm. On the other hand, if the
aperture is made 10 mm, the retinal area seen is so much larger than the
illuminating spot that the system is really afocal. A third option of
invest is to use a 10 mm aperture with a central 1 mm stop, giving a "dark
field" view of the retina, in which only light indirectly reflected is
detected. With a rotatable aperture disk 26, as shown in FIG. 6a the view
can be varied from tightly confocal to afocal or dark field. The same disk
can be arranged to carry filters for various wavelengths. Following the
retinal plane a simple 10X microscope objective (not shown) can be used to
bring the pupil back down to 1 mm for a match to the avalanche diode.
If the polygonal reflector 15 is formed with twenty-five facets,
distortions due to facet-to-facet and other variations remain stationary
in the displayed raster image, since it is evenly divisable into 525
television lines. For this reason, it is deemed preferable that the
polygonal sacnner have a number of reflective facets equal to an integral
multiple of twenty-five. For different raster scan frequencies, a
different number of facets would be appropriate. The controlling factor is
that the number of reflecting facets should be integrally divisible into
the number of raster lines. Further, as described above, there is a common
optical path from the horizontal scanner 15 to the target object (in this
example, the fundus of the eye) for the scanning beam and for the
reflected light. Under these circumstances any reflection of the input
laser beam from elements in the common optical path will appear as a noise
signal to the detector. Accordingly the focusing elements 16 and 18, as
well as scanning elements 15 and 17 are, front-surface mirrors.
While the instrument 10 has been described in terms of the advantages of
de-scanning to produce signals corresponding only to light reflected
directly from the illuminated target area, there are situations in which
it is advantageous to look only at indirectly reflected light. This can be
accomplished by moving the detector off the optical axis of the system to
that it is in effect looking at target areas displaced from the direct
illumination of the input beam. It has been found that information
provided from these reflections also is useful in determining
characteristics of an eye fundus. An alternative arrangement for attaining
this response to only indirect illuminating is to image on the detect a
target area concentric with, and larger than, the illuminated area, and to
mask light reflected from the illuminated area, e.g. with a dark-field or
central stop.
Moreover, if the detector is moved axially, the plane of the image can be
moved to positions anterior to the retinal surface and thus various types
of floaters, such as vitreous spots and strands may become visible in the
image. Similarly, movement of the image plane to posterior, sub-surface
positions enables the instrument to image interior structure of the eye
fundus.
The 15.75 kHz horizontal scan frequency an the 60 Hz vertical scan
frequency described above for the illustrated embodiment are for use with
television standard adopted for the USA. These values can be selected to
suit other standards in practice in other countries. For example, the
standard which operates with 625 lines per frame, requires the same 15.75
kHz horizontal scan frequency and a 50.4 Hz vertical scan frequency.
As a practical matter it is desirable to leave the patient confortably
stationary (in a head rest) and for the physician to move the
ophthamoscope to change the angle of the entrance beam. This means moving
sources, detectors, optics and scanners. With a polygon rotating at
40,000, RPM gyroscopic considerations must be addressed. To avoid
gyroscopic torques the polygon must be moved only parallel to or
perpendicular to its axis of spin. In the present embodiment a
conventional fundus camera mount is used to support the ophthalmoscope and
its motion can be controlled over short distances by a joy stick. Since
the mount translates the polygon along X, Y, or Z axes, and rotates it
about the Z axis, as illustrated in FIG. 11, none of the motions tilt the
spinning axis Z of the polygon 15 and consequently there are no gyroscopic
torques on the bearings.
In FIG. 12 there is illustrated an oscillator clock supply which provides a
polygon driver output signal and a vertical clock output signal. The
oscillator includes a crystal controlled master clock 60, typically
operated at 4.032 mHz. The output of the clock 60 is provided to a binary
counter 61 and a 126 kHz signal from the binary counter is provided to a
divide by 25 circuit 62, the output of which provides the polygon driver
output signal. A second signal is taken from binary counter 61 at 31.5 kHz
and this is the vertical clock output signal.
FIG. 13 is a block diagram illustrating the manner in which the start of
scan pulses generated from the pin diode 50 are processed to produce the
blanking input for the AOM, the composite synch output for the monitor and
other peripherals, and the vertical scanner drive. The output from the pin
diode 50 is supplied through amplifier 51, delay circuit 54 and blank
width control element 55 as one input to NOR gate 70. The delay unit 54 is
arranged to equal the time required for the facet to rotate from the
sensing position into the position where it intercepts the laser beam for
scanning. Width circuit 55 provides for a pulse which is adjusted to be
wide enough to cause blanking from the time one ra | | |
