Background pf the invention

The present invention relates to eye measurement instruments generally and more specifically to instruments for measuring intraocular pressure.

A variety of devices currently exist that measure intraocular pressure. Initially, intraocular pressure was measured by tonometers that used a mechanical probe to applanate the cornea and measure the force required for a specific applanation distance. The intraocular pressure was then calculated from the measured force. However, applanation or contact tonometers require a topical anesthetic and cause varying degrees of distress and discomfort to patients. These drawbacks stimulated the development of noncontact tonometers that use a pulse of high pressure air to applanate the cornea. Although air pulse tonometers eliminate any contact with the eye as well as the need for a topical anesthetic, the audible sound and the high pressure burst of air directed toward the eye can still create fear and discomfort in some patients.

A new generation of devices is currently under development that measure intraocular pressure by modulating acoustic waves directed toward the cornea and monitoring the response characteristics of the wave. These devices generally operate by causing the cornea to vibrate by directing an acoustic wave of specific amplitude, frequency and phase toward the cornea. A second acoustic or light wave with its own unique characteristics is also directed toward the cornea. The vibrating surface of the cornea reflects this second wave and modulates the characteristics of the second wave. A detection device receives the signal of the modulated, reflected, second wave and correlates the change in wave characteristic to a measurement of intraocular pressure.

One such apparatus is described in U.S. Patent No. 4,928,697 to Hsu, which is incorporated herein by reference in its entirety. This patent uses a speaker to excite the cornea with a low frequency acoustic wave, a transmitter to direct a high frequency acoustic wave toward the vibrating surface of the cornea, a receiver to acquire the modulated high frequency wave reflected from the vibrating surface of the cornea and an electronic device for detecting and displaying any modulation in the amplitude of the high frequency wave. This apparatus also includes a chin rest or saddle to

help hold the patient's head in a fixed position. However, testing of dual acoustic wave devices similar to that disclosed in this patent has produced only a weak correlation between the measurement of amplitude modulation and intraocular pressure, and it is has not been shown that the accuracy of this technology is acceptable or equal to the accuracies available with existing, contact and noncontact tono etry equipment, possibly because of interference of the perturbing wave on the measuring acoustic wave, as well as limitations in resolution of the acoustic measuring wave.

A method of correlating the mean normal response frequency of the cornea with intraocular pressure has also been suggested. See J.M.

Hamelink and G.L. Cloud, "Ocular Tonometry Through Sonic Excitation and Laser Doppler Velocimetry," Journal of Bio echanical Engineering. 101, pp. 267-70 (Nov. 1979). The Hamelink, et al., article describes experiments in which low frequency, acoustic cornea! excitation and laser Dόppler velocimetry were used to develop an empirical relationship between the mean normal response frequency and intraocular pressure. Mean normal response frequency is defined as the frequency at which the ratio of response amplitude to input amplitude is at a maximum, and this parameter was selected because it is observable using laser Doppler velocimetry techniques. Although the experimental setup was fairly crude and intraocular pressure was the independent variable in this experimentation, the observed scatter in the frequency data suggested that intraocular pressure measurements with accuracies equal to existing clinical tonometry equipment could be obtained. Although the Hamelink and Cloud research furthered amplitude modulation techniques by using laser Doppler velocimetry, it did not actually provide a method for measuring intraocular pressure but merely suggested that phase modulation might be a more accurate technique for intraocular pressure measurement.

Despite the new generation noncontact tonometry equipment described above, a definite need still exists for a research tool and clinical instrument that provides accurate, efficient monitoring of the magnitude of cornea velocity without causing physiological or psychological discomfort to the subject.

Brief Summary of Invention

The present invention improves upon prior art tonometry and eye dynamics monitoring equipment by permitting the accurate, efficient monitoring of eye dynamics through an apparatus that does not contact the eye, requires no topical anesthetic, creates no physiological discomfort to the patient and minimizes the likelihood of psychological concern in the patient. The invention accomplishes these objectives by using a laser interferometer and advanced microprocessor signal processing methods to measure the velocity of a perturbed cornea surface. The interferometer presents to a detector the phase modulation of a coherent laser beam reflected off the surface of a cornea as the cornea is perturbed by a low frequency, low pressure, acoustic wave. The frequency of the phase modulated signal is proportional to the magnitude of the cornea velocity. Briefly, the invention includes an interferometer having a Helium-Neon (HeNe) coherent laser light source and a photocell detector mounted on a horizontal mounting plate that is supported on a larger structural plate by an electro-mechanical positioning platform capable of providing x, y, and z-axis adjustment. The large structural plate is supported by rigid, pneumatically damped legs to minimize surrounding background vibration. A fixture or saddle to support and position the patient's head is mounted on damped rods fixed to the large plate. These damped rods minimize vibration and provide coarse adjustment in the y-axis (vertical) direction. A hand¬ held, electronic joystick actuates DC servo motors to properly align the interferometer optics with a subject's eye. The HeNe laser beam is directed through a beam expander to produce a larger interfering wavefront requiring less demanding fixture alignment tolerances. A beam splitting cube divides the laser light beam into a reference beam and a measurement beam. The measurement beam is directed through an achromatic lens that focuses the measurement beam approximately to the center of curvature of the cornea, 8 millimeters (mm) posterior to the cornea's epithelial layer, minimizing the redirecting effect the cornea's spherical power has on the reflected measurement beam. The reference beam is directed through an attenuator to insure that the reflected measurement beam and the reference beam are of equal intensity prior to being recombined by the beamsplitter. The recombined beam passes through a second dichroic beamsplitter that reflects radiation with a

wavelength of between 400-600 nanometers (nm) and transmits radiation with a wavelength of between 600-800 nm through an imaging lens and onto the photocell detector. A target projector illuminated by a fiber optic illuminator projects a target image through a third beamsplitter and onto the surface of the patient's cornea where it is seen as a blue-green target on the red field of the HeNe laser beam. The target provides the patient with an object to focus on and reduces random eye movements.

Recombining the reference and measurement beams creates an optical effect called a dynamic fringe interference pattern. When the recombined beam is reflected onto the photocell detector, the detector uses optical and electronic components to convert the fringe interference pattern into an electrical fringe pattern signal that is transmitted to a preamplifier with an extremely high signal-to-noise ratio. The axial fringe pattern signal can be displayed on an oscilloscope to confirm fringe acquisition. The fringe pattern signal produced by the photocell detector and preamplifier is digitized by microcomputer hardware and software. This digitized signal is processed by software algorithms to de-modulate and filter the signal into cornea velocity. A specialized "matched" filter routine computes the maximum value of the magnitude of the cornea velocity in response to the acoustic excitation. The algorithm minimizes the effects of eye motions unrelated to the motion forced by the acoustic wave. Intraocular pressure or other physical properties of the eye can then be determined from the maximum cornea velocity.

Accordingly, one objective of the apparatus of the present invention is to provide a device for measuring the dynamics of an eye.

Another objective of the apparatus of the present invention is to provide a device for measuring the velocity of a perturbed cornea.

Still another objective of the apparatus of the present invention is to provide a device for measuring the velocity of a perturbed cornea while minimizing patient discomfort.

Still another objective of the apparatus of the present invention is to provide a device for measuring intraocular pressure.

A further objective of the apparatus of the present invention is to provide a noncontact device for measuring intraocular pressure. Another objective of the apparatus of the present invention is to provide a device for measuring the dynamics of an eye using laser interferometry, including tear-film layer break-up time.

Another objective of the apparatus of the present invention is to provide a device for measuring intraocular pressure using laser interferometry. * Another objective of the apparatus of the present invention is to

5 provide a method of measuring the velocity of a perturbed cornea.

Still another objective of the apparatus of the present invention is to provide a method of measuring the velocity of a perturbed cornea while minimizing patient discomfort.

Still another objective of the apparatus of the present invention is 10 to provide a method of measuring intraocular pressure.

A further objective of the apparatus of the present invention is to provide a noncontact method of measuring intraocular pressure.

Another objective of the apparatus of the present invention is to provide a method of measuring the dynamics of an eye using laser is interferometry, including tear-film layer break-up time.

Another objective of the apparatus of the present invention is to provide a method of measuring intraocular pressure using laser interferometry.

These and other objectives and advantages of the apparatus of the 20 present invention will be apparent to one skilled in the art from the detailed description, drawings and claims that follow.

Brief Description of the Drawings

FIG. 1 is a side elevational schematic view of the present invention with elements deleted for clarity. 25 FIG. 2 is a top plan schematic view of the interferometer assembly of the present invention.

FIG. 3 is a front elevational view of the head support assembly illustrated in FIG. 1.

FIG. 4 is an electrical schematic of a preamplifier that can be used ^ 30 in the present invention.

FIG. 5 is a flow chart of signal acquisition and processing software that may be used with-the present invention.

FIG. 6 is a representation of the axial fringe pattern signal display produced by the present invention.

FIGS. 7-10 are flow charts of computer software that may be used with the present invention.

FIG. 11 is a pair of graphs illustrating an uncorrupted acoustic response model and spectrum as functions of time and frequency.

Detailed Description of Invention

As shown in FIGS. 1-3, the cornea velocity measurement device 10 of the present invention generally includes a vibration minimizing support table 12, a head fixture assembly 14, an instrument support and adjustment assembly 16, interferometer assembly 18 and biomicroscope 20. As can be seen in FIGS. 1 and 3, head fixture assembly 14 has adjustable chin rest 24, head band 26 and head supports 28. Adjustable chin rest 24 helps position the patient's head along the y-axis while head band 26 and head supports 28 help hold the patient's head in a fixed position. Head supports 28 may be made of any suitably compliant material and design so as to hold the patient's head in a fixed position, but

"wedge" shaped padded foam is preferred. Chin rest 24, head band 26 and head supports 28 are mounted to mechanically damped rods 29 that are mounted on table 12. Coarse adjustments to chin rest 24 and rods 29 can be made by turning adjusting screws 31 and 33, respectively. Alternatively, chin rest 24 may be adjusted by use of a DC servomotor driven linear platform (not shown). Suitable chin rests 24, head bands 26, rods 29 and adjusting screws 31 and 33 are commercially available from sources such as Nikon Corporation and Newport Corporation, 18235 Mt. Baldy Circle, Fountain Valley, California. Support table 12 contains optical breadboard 30 and pneumatically damped legs 32. To minimize external movements that may affect the measurements obtained by device 10, breadboard 30 should be relatively massive and stiff, for example, weighing at least 500 pounds. Suitable breadboards 30 and legs 32 are commercially available from sources such as Newport Corporation.

Instrument and adjustment assembly 16 contains optics plate 34, x- axis translation stage 36, z-axis translation stage 38 and joystick controller 40. Stages 36 and 38 are driven by DC servomotors (not shown) that are controlled by joystick 40. If a DC servomotor is used to drive

adjusting screws 31 thereby adjusting chin rest 24 along the y-axis, joystick 40 may contain switch 41 to control this additional servomotor.

As can be seen in FIGS. 1 and 2, interferometry assembly 18 generally contains laser 42, beam expander 44, beamsplitter 46, lens 48, attenuator 50, beamsplitter 52, relay lens 54, beamsplitter 56, lens 60, detector 62 and preamplifier 63. Optics plate 34 is preferably an aluminum alloy. Laser 42 is preferably a Helium-Neon laser with a discrete wavelength, an output power approximately between 2 and 3 milliwatts (mW) and a beam diameter of approximately 0.8 mm, however, other suitable lasers such as Argon, diode and frequency double YAG lasers may also be used. Suitable lasers 42 are commercially available from a wide variety of well-known sources. Beam expander 44 preferably is a 20x Galilean beam expander, but other suitable beam expanders 44 may also be used. Beam expander 44 expands laser beam 64 to approximately 16 mm to produce a larger interfering wavefront requiring less demanding fixture alignment tolerances. Iris diaphragm 66 controls the diameter of expanded beam 64a. Beamsplitter 46 splits expanded beam 64a into reference beam 65 and measurement beam 68. Lens 48 focuses measurement beam 68 approximately 8 mm posterior to the center of curvature of the cornea, thereby minimizing the redirecting effect the cornea's spherical power has on the reflected measurement beam 68. Lens 48 is preferably an achromatic, F/2 lens with a 32 mm focal length. Attenuator 50 is preferably a neutral density filter with an optical density of 0.8. Attenuator 50 reduces the intensity of reference beam 65 by transmitting approximately 15.7% of the reference beam per pass. Attenuation of reference beam 65 is necessary because the patient's cornea surface reflects only about 2.5% of measurement beam 68 back to beamsplitter 46 and both reference beam 65 and measurement beam 68 must be of approximately equal intensity to produce a usable interference pattern such as that illustrated in FIG. 6. If reference beam 65 is not attenuated, the intensity of reference beam 65 will not maximize the contrast or amplitude modulation of the interference pattern caused by modulated measurement beam 68. Imaging lens 60 focuses beam 72 through interference filter 74 and onto detector 62. The amount of detector shot noise (NEP) is proportional to the detector surface area. By using lens 60 to focus recombined beam 72 on detector 62, a detector 62 having a small surface area and, thus, a higher sensitivity, can be used. Lens 60 preferably has a focal length of 100 mm and filter 74 is preferably a 1 nm

full width at half maximum (FWHM) optical interference bandpass filter. Filter 74 reduces the out-of-band optical noise reaching detector 62. Detector 62 converts the incident photons of recombined beam 72 into an output signal that is directed to preamplifier 63. Detector 62 is preferably a PIN or silicon photodetector that has its peak sensitivity at the wavelength of laser 42, and suitable detectors 62 are commercially available from sources such as EG&G Electro-Optics, Salem, Massachusetts and Silicon Detector Corporation. Preamplifier 63 may be any suitable preamplifier and a schematic diagram of one suitable preamplifier 63 is illustrated in FIG. 4. Other preamplifiers 63 may also be used.

Interferometer assembly 18 also contains biomicroscope 20 and target projector 58. Target projector 58 is illuminated by light from fiber optic cable 76 and projects a rotating green point source onto beamsplitter 56. Beamsplitter 56 preferably is 50% reflective and 50% transmissive and reflects green target beam 77 through relay lens 54 and onto beamsplitter 52, which is preferably transmissive of radiation in the 600-800 nm range but reflective of radiation in the 400-600 nm range, thereby allowing recombined beam 72 to pass through but reflecting target beam 76 back through beamsplitter 46, lens 48 and into the patient's eye, where it appears as a rotating green target within the red field of measurement beam 68. This rotation helps to make a subject's eye saccadic motions more consistent and predictable. While the patient fixates on the green target, an operator observes the subject's cornea through binocular microscope 20, which may be any suitable biomicroscope such the Model 900 BM stereo biomicroscope available from Haag Streit, and uses joystick 40 to position optics plate 34 so that measurement beam 68 is reflected off of the center of the patient's cornea and aligned with reference beam 65. Alignment by the operator is aided by the use of fiber optic illuminator 82.

In use, a patient places his or her head in head support 14 and interferometer assembly 18 is aligned as discussed above. An acoustic wave, for example, a 20 Hertz (Hz) acoustic wave generated by speaker 79, is directed at the patient's cornea through acoustic port 78 via line 81 that is preferably a 18 AWG cannula and is prevented from contacting the patient's eye by cone 80, as can be seen in FIG. 2. The acoustic excitation is not a standing wave pattern but more of a pulsating jet of air. The use of the 18 AWG cannula maximizes the applied acoustic pressure by decreasing the exposed surface area of the cornea. This has the

additional benefit of decreasing the drying sensation of the cornea experienced by the subject. The cannula should have no bends to ensure laminar flow. The acoustic pressure applied to the cornea is limited by the comfort level of the subject. The low frequency acoustic wave is inaudible to the patient and causes the cornea to vibrate at the same frequency as the acoustic wave. As the cornea is vibrating, laser 42 emits beam 64 that passes through variable neutral density filter 84 that reduces the energy in beam 64 to less than 170 nanowatts, thereby avoiding color saturation in the patient's eye, and is directed by mirror 86 through beam expander 44, iris diaphragm 66 and beamsplitter 46 where it is divided into measurement beam 68 and reference beam 65.

Measurement beam 68 is directed through lens 48 where it is reflected off of the vibrating surface of the patient's cornea and back through lens 48 and beamsplitter 46. Reference beam 65 passes through attenuator 50, reflects off mirror 70, passes through attenuator 50 a second time and is recombined within beamsplitter 46 with reflected measurement beam 68 into recombined beam 72. The reflection of measurement beam 68 off of the vibrating cornea surface causes beam 68 to be slightly out of. phase with beam 65 so that the recombination of beams 65 and 68 results in an optical effect called a dynamic fringe interference pattern as illustrated in FIG. 6. This fringe pattern has the appearance of a series of concentric, alternately dark and light rings, with each pair of rings representing the sinusoidal intensity variations of recombined beam 72. The intensity of the recombined beam will increase whenever the phase of beams 65 and 68 are identical, resulting in a light ring, and decrease whenever the phase of the two beams are opposite, resulting in a dark ring. As the cornea surface is displaced, each ring sinusoidally alternates in intensity. The phase of this intensity variation is directly proportional to displacement. An alternance from light to dark represents an optical path length variation of 1/2 wavelength of the radiation source. In the case of a HeNe laser having a wavelength of 632 nm, each alternance represents a relative cornea movement of 316 nm.

The function of measuring these light to dark alternances is initiated by detector 62, which converts the optical signals to electrical signals. The electrical signals are amplified and conditioned by pre¬ amplifier 63. Once amplified and conditioned by pre-amplifier 63 the electrical signal can be routed to an oscilloscope for monitoring and/or to

any suitable data acquisition and processing equipment, such as a micro¬ computer using software similar to that illustrated in FIGS. 7-10. FIG. 5 illustrates the general approach to the more specific flow diagrams illustrated in FIGS. 7-10. When the distance traveled by the cornea is measured as a function of time, the peak velocity of the vibrating cornea can be calculated easily because the velocity is directly proportional to the frequency of the dynamic fringe pattern signal. The peak velocity in response to acoustic excitation is determined by frequency demodulating and filtering the dynamic fringe signal. The degree of motion by the perturbed cornea to be measured is extremely small. Peak displacements less than approximately 2 microns and velocities less than approximately 12 microns per second for an acoustic excitation of 20 Hz are possible. This makes device 10 sensitive to motions such as random eye motions (saccades) as well as head and neck motions. These "noise" motions can be greater than the desired response to the 20 Hz acoustic excitation. Therefore, the signal must be further processed to separate the signal from the noise. One possible method uses "matched" filter algorithm 100 (FIG. 5) incorporated within the signal processing software. Matched filter algorithm 100 has its spectral characteristics matched identically to those of the uncorrupted acoustic response signal. The desired acoustic response is modeled as shown in FIG. 11. Filter algorithm 100 functions by cross-correl ting the spectrum of the uncorrupted response (the desired signal spectrum) with the spectrum of the detected signal (desired acoustic response plus "noise" motions) and dividing by the power spectral density of the noise. The result is an estimate of the peak velocity of the acoustic signal response. This value is proportional to intraocular pressure ("IOP") and thus can indicate the level of IOP. However, as with any filter, a small amount of noise will pass through. If the energy of the noise which passes matched filter algorithm 100 is greater than the energy of the desired signal, then the output will be a poor estimate of the desired velocity signal peak. Thus the "noise" motions must be minimized such that detection of the acoustic signal can occur. Matched filter algorithm 100 is implemented within a program written in the ASYST™ programming language, called PLAN.A1 (FIG. 7). This program acquires, saves and processes the raw fringe signal and outputs the estimated peak velocity of the response to the 20 Hz acoustic pulse.

Once optical alignment is obtained, the operator initiates the acquisition process by depressing "fire" button 39 on joystick 40. Program PLAN.A1 acquires a total of six seconds of raw fringe data. The first second of data is taken with acoustic port 78 disabled. This data is used as an estimate of the noise for computation of the noise power spectral density. Acoustic port 78 is then permitted to transmit the 20 Hz acoustic wave and the remaining five seconds are acquired. The raw fringe data can be saved if desired.

Program PLAN.A1 processes each of the five second sample periods separately, resulting in an estimate of the peak cornea velocity for each, and determines the mean and standard deviation for the five samples. The mean is the desired estimate of the magnitude of the peak cornea velocity.

Once the magnitude of the peak cornea velocity has been determined, this velocity can be correlated to intraocular pressure or other physical properties of the eye. For example, peak cornea velocity can be correlated to axial cornea displacement or cornea acceleration using methods well- known in the art.

This description is given for purposes of illustration and explanation. It will be apparent to those skilled in the relevant art that modification may be made to the invention as described above without departing from its scope and spirit.