JPH04166149A - Device for displaying electronic image of nonstationary object - Google Patents

Abstract

PURPOSE: To automatically stop laser irradiation when a laser cannot be matched with a target by providing object tracking means to track motion of a nonstationary object, inspection means to form electronic expression of an optical image to generate an optical image of the nonstationary object and display means to convert the electronic expression. CONSTITUTION: A main computer 62 transmits an ignition signal on a system bus 64 to an illumination/shutter controller 1. The controller 1 transmits an ignition signal on a line 168 to a laser 170. Even in any case in laser treatment, since a tracking sub-system can pursue motion of eyes, a postprocessing part 152 transmits a tracking error signal on the system bus 64. The illumination/ shutter controller 1 reacts to an error signal by directly transmitting a cutoff signal on the line 168 to the laser 170. Transmission of laser energy is not resumed until a doctor once again confirms a treatment plan displayed on a screen. Therefore, an unexpected injury of eyes of a patient which is possibly caused when laser irradiation is started in a condition where a laser beam is not matched with a target, can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a device for displaying electronic images of non-stationary objects. The device according to the invention relates to an equipment for ophthalmological treatment which includes a device for delivering laser energy. In particular, the device according to the invention tracks the retina to maintain a constant real-time image of the fundus and automatically directs a laser beam to a defined target comprising the fundus to treat fundus pathologies. It has something to do with what you do.

[Prior Art and Problems to be Solved by the Invention] In the actual practice of ophthalmology, involuntary movements or sudden movements of the patient's eyes have conventionally occurred, hindering effective diagnosis and treatment of the fundus. was.

The continuous occurrence of sudden and other eye movements prevents physicians from obtaining constant real-time images of the eye to diagnose the human fundus and prescribe treatment for medical conditions.

Also, when treating an eye with laser energy, sudden movements of the eye can cause the laser to miss its target and damage healthy retinal tissue.

The effectiveness of laser treatment of eye diseases depends on the ability to track eye movements during the period that laser treatment is being delivered and to prevent the initiation of irradiation when the laser is not on target. If the physician must manually aim and initiate the laser at a non-stationary target, the effectiveness of the treatment depends on reflexes, physician hand-eye coordination, and patient-optics alignment. Dependency is inevitable.

Both doctors and patients have similar concerns that sudden movements of a patient's eye can cause the doctor to apply laser therapy to the wrong location on the retina, thereby damaging healthy tissue. Laser therapy typically requires a physician to manually aim a laser at a moving target. Once the laser is aligned with the selected treatment area, the physician activates the laser. Due to current therapy delivery technology deficiencies, the results of laser therapy can often be ineffective or even harmful. For example, improper application of laser treatment can cause disease recurrence, overtreatment can cause unexpected bleeding, and heat transfer during laser treatment can cause patchy separation after surgery.

Instead of successful treatment, deviation from the intended laser target can cause permanent, irreparable damage to the optic nerve or other parts of the retina. Such risks have often discouraged the use of laser therapy for patients with limited vision, who may otherwise benefit from the delivery of laser therapy. It is.

According to the system, the movements of the human eye are tracked within the limits, thereby generating a clear, constant, real-time image of the fundus of the eye for diagnosing diseases. However, such tracking systems do not coordinate the video display with the execution of the laser energy treatment. In some conventional systems, after diagnosis, the patient's eyes must be viewed by a separate system and then subjected to laser treatment. This approach requires locating the target area for treatment a second time. Additionally, differences in the type of irradiation and viewing equipment used to diagnose and treat a patient's disease can change the appearance of the fundus as viewed by the physician and complicate the treatment process.

Other systems have been proposed that can combine the delivery of laser energy with automatic tracking for viewing images for viewing. Such a system is
While providing a steady target for aiming the laser energy delivery device, it has a number of drawbacks. First, because of the single wavelength laser illumination, the system is limited to producing monochromatic images in the color of the illuminating laser, resulting in less than optimal images and different features are not necessarily distinguishable. Second, the system does not provide a means for automatic aiming and delivery of laser therapy. Instead, a user manually aims and applies laser energy for the length of time necessary to effect the intended treatment. Even though the system is able to track eye movements, the success of treatment ultimately depends on the accuracy of the user's laser aiming and determining the appropriate length of time to deliver laser energy to the user's target. Depends on.

Additionally, the system provides the physician with a direct image of the fundus of the eye, thereby exposing the physician's eye to the patient's eye and the harmful laser energy reflected by the optical system and returned to the image forming system. It will be exposed.

These systems that integrate combined tracking and laser illumination also rely on tracking of salient features of the fundus. This tracking technique requires the user to perform operations such that the features that consist meet the constraints of the operating characteristics of the tracking system. Requires operation such that optimal tracking performance characteristics are achieved. Diagnosis and treatment time is thereby extended by requiring the user to locate orthogonal features within the field of view of the video tracking system. Locating the feature can be a stressful and fatiguing process for the user, often resulting in less than optimal tracking performance characteristics.

Furthermore, the system's tracking algorithm results in inaccurate tracking. The tracking algorithm cannot determine the actual distance required to match the current image of the fundus with the desired image. The system calculates a ratio of a first value representing an intensity profile of a portion of a line of pixels to a second value representing an intensity profile of a portion of the line of pixels. The calculated ratio of the first value and the second value is compared to a previous ratio. The adjustable tracking reflector is adjusted by one unit in one direction based on the relative values of the current and previous ratios. If the image moves more than one unit between comparisons, the image cannot be returned to its previous position within a single tracking cycle that includes image comparison and calculated adjustments. The unit adjustment must be large enough to track high velocity motion. However, if the unit is too large,
Coordination for small movements is no longer accomplished.

The general object of the invention is therefore to overcome the drawbacks of the prior art.

In particular, one object of the present invention is to generate a true color image of the fundus, automatically aim a laser at an intended target on the fundus, and track the movement of the target during the period of laser irradiation. , and to provide a system that automatically stops laser irradiation when the laser no longer aligns with the target.

Another object of the present invention is to enhance the ability of laser treatment of the fundus of the eye, with automatic aiming of the laser and automatic tracking of eye movements within the period of laser irradiation.

A related object of the present invention is to reduce the risk of ocular damage associated with laser treatment of the fundus of the eye.

Another object of the present invention is to prevent undertreatment or overtreatment of a portion of the fundus of the eye with laser energy.

Another object of the invention is to provide a tracking system that does not rely solely on tracking individual features.

A related object of the invention is to provide a system that is easy to configure and has improved reliability.

Another object of the invention is to fully compensate for fundus motion within a single tracking cycle by generating a variable distance adjustment signal.

[Means to solve the problem]

The foregoing and other objects are achieved by the present invention, which provides an improved, integrated system for the diagnosis and treatment of diseases of the ocular fundus. The system includes a high velocity responsive tracking system that generates a variable magnitude optical path adjustment vector and maintains a constant image of a moving target over a large range of angular velocities. include. Target tracking is an interesting sight (target area)
The comparison of the selected reference image with the current image of the image is facilitated. Also provided is a technique for indirectly observing the fundus with digitized, addressable, true color video footage. An interactive user program allows the physician to create a treatment plan based on the video monitor.

Finally, signals transmitted by the workstation's processor to the laser aiming device direct laser energy to the patient's fundus according to the target displayed on a real-time image of the fundus on a color video monitor.

The system of the present invention includes an automatic tracking subsystem. The user selects the desired view of the fundus. As the patient's eyes move, the automatic tracking subsystem calculates and transmits a set of adjustment vectors to the servo-reflector positioner. The servo reflector adjusts the optical path of the image-producing optical train to optically compensate for movement of the fundus in the image plane, thereby maintaining a constant real-time image of the fundus. Servo reflectors are capable of high frequency response due to improved image processing hardware, image processing algorithms, and reflector servo control hardware. More specifically, one array of systolic parallel processors is used to compare the current image with a desired reference image previously stored in the array processor. One method of comparison, based on statistical correlation analysis, is to store current real-time images generated by a CCD camera that have been hypothetically displaced by different amounts in several directions. Compare with the reference image stored in . After said comparison has been made, a post-processor (digital signal processor) identifies the best match in magnitude and directional displacement between the stored video and the current video, and determines the best match against an acceptable threshold. Measure correlation. If the alignment is within the working range, an adjustment vector is transmitted to the reflector positioner, which aligns the current image with the reference image.

In one form of use of the system of the invention, the physician:
Define the optical coagulation treatment plan by marking the laser treatment target on the fundus image displayed on a true color video display. Coordinates and parameters for the treatment are stored on the computer workstation. A laser aiming subsystem, controlled by the same workstation, automatically directs the laser beam to a selected target displayed on the screen in treatment mode.

The physician then activates the laser by entering a confirmation signal using the control keypad.

During treatment, if the tracking subsystem loses track of the target area, the laser will automatically shut off and
Laser treatment is not resumed until the laser beam and target are realigned and the physician restarts treatment. This prevents accidental damage to the patient's eye that could occur if the laser beam was not aligned with the target and the laser was started.

The present invention avoids the need for the physician to use a direct optical viewing point interface with the optical subsystem that would allow stray reflected laser light to damage the physician's eye. The video interface allows the doctor to view images of the fundus of the eye directly through an optical viewing point.
A video monitor allows for indirect observation.

Indirect display of the fundus by video display isolates and protects the physician's eye from laser energy reflected towards the viewing channel. In one embodiment, the fundus is red;
A color image of the fundus is generated by illuminating with monochromatic green and blue light and capturing and displaying an electronically constructed color image on a high resolution color monitor.

The present invention also provides video display and digital storage of a patient's retinal image data, facilitating the use of an expert system to alert physicians to risky treatment plans and suggest common treatment patterns. The doctor can accept the proposed treatment or develop an alternative treatment plan using the video footage of the fundus. A set of markers is displayed on the video screen to represent the therapeutic laser target.

The laser delivery subsystem of the present invention includes a surgical laser interface that uses a servo reflector controlled by a processor to direct emitted light energy to a target.

A processor causes a signal to be transmitted to a positioning device coupled to the reflector. The signal is responsive to markers displayed on the video screen that indicate where to direct the laser. The doctor starts irradiating the laser, and the processor
The laser is directed to the next target and in this way all targets represented by the markers displayed on the screen will have been treated.

As the patient's eye moves while the laser is being applied, the tracking servo reflector is adjusted to change the beam path, thereby maintaining alignment of the laser beam with the target. If for any reason the tracking subsystem is unable to maintain tracking of the fundus, the laser is automatically shut off, thereby preventing the laser from irradiating unintended locations on the fundus.

The present invention also connects computing hardware to a computing workstation, thereby facilitating computer-aided diagnosis and treatment of the fundus and storage of data related to patient diagnosis and treatment. Such hardware may include a CPU, graphics processor, keyboard, joystick/mouse/trackball device, tape storage, and hard disk drive. A modem or similar communication link is used to provide a remote physician with a real-time image of the patient's fundus and to provide the possibility of defining and performing therapeutic operations from a remote location. can be

〔Example〕

Referring specifically to page 7 of FIG. 1, the illumination/shutter controller 1 sends an activation signal on line 3 to the xenon arc lamp 2. As shown in FIG. Xenon arc lamp 2 is visible illumination fiber optic 4
The light beam is sent to the filter wheel 6 at the top. Filter wheel 6 is comprised of three angularly spaced color filters for selectively transmitting red, green, or blue color bands on optical path 8 to imaging delivery optics 10 . The chromaticity of the light sent on optical path 8, whether red, green or blue, depends on the rotational position of filter wheel 6. The light beam that has passed through the filter continues on the optical path 12,
It passes essentially unchanged through the beam splitter 14. The light beam that has passed through the filter continues on the optical path 16,
It passes essentially unchanged through the beam splitter 18. The light beam passing through the filter continues on optical path 20 and is reflected onto optical path 24 by beam splitter 22 . The light beam on the optical path 24 that has passed through the filter is reflected from the tracking servo mirror 26 and proceeds along the optical path 28 to the scanning eye optics 30. The filtered light beam continues on optical path 32 to a contact lens 34 suitable for the patient's eye 38 . Contact lens 34 produces an image of the filtered light along optical path 36 on the patient's retina 40 . Furthermore, the light that has passed through the filters alternately can be sent directly from the xenon lamp 2 into the optical path 36 via an optical fiber connected to the contact lens 34 .

The retina 40 partially reflects the light that has passed through the filter.

The light that passes through the filter returns to the contact lens 34 along an optical path 36. Contact lens 34 forms an image of the reflected, filtered light, which continues on optical path 32 to scanning eye optics 30 . The reflected light continues on optical path 28 , is reflected by servo tracking mirror 26 , and continues on optical path 24 to beam splitter 22 . Half of the light traveling away from tracking servo mirror 26 on optical path 24 is sent by a beam splitter on optical path 42 to dichroic beam splitter 44 . Substantially all of the filtered imaging light passes through beam splitter 44 and continues on optical path 46 to zoom lens 48 . Focusing light emitted from the zoom lens on the optical path 50 is filtered by an infrared cutoff filter 52 . The focused light formed from the image-forming light that has passed through the filter is transmitted to the image-forming CCD.
It continues on the optical path 54 to the camera 56.

The optically active surface of the CCD camera 56 is the tracking mirror 26
is aligned with the image plane of the array of scan-to-image forming optical systems formed by the optical path taken from the light reflected by the image forming CCD camera 56 to the image forming CCD camera 56.

The CCD camera 56 consists essentially of a 512 x 512 CCD array that sequentially polls each cell and generates an analog signal on a line 66 representing the intensity of light energy incident on a given cell. It has appropriate access circuitry to output the signal. The operation of the CCD camera 56 is governed by control signals transmitted from the camera controller 60 onto the line 58. The operation of camera controller 60 is governed by instructions sent from workstation/main computer 62 to controller 60 on system bus 64 .

Still referring to FIG. 2, a CCD camera 56 sequentially bores each cell in the CCD array. is transmitted to the frame grabber 68. Frame grabber 68 converts the digital data stream received on line 66 into 512 x 512 byte frames;
Frame-directed digital image data on line 70 is transmitted to bit mapped addressable video storage 72 .

Multiplexer 74 routes the gray scale data received on line 70 to lines 76, F8 or 80 and ultimately to respective color video memory modules 82, 84 or 86. The selection of the video memory module to receive grayscale data is determined by control signals sent on the system bus 64 from the light/shutter controller 1 to a multiplexer in the video memory 72.

Returning to FIG. 1, the filter wheel 6 is equipped with a position sensor (not shown) which generates a signaled position according to the position of the filter wheel 6 and thus according to the collar sent thereby. This position signal is carried on a line 89 to the illumination/shutter controller 1 which indicates the color of the illumination light being sent from the filter wheel 6 on the optical path 8.

Light/shutter controller 1 sends control signals on system bus 64 to video memory 2. multiplexer 74
A video memory module 82.8 is connected via selection line 75 to receive a grayscale video image from
By selecting one of 4 or 86, video memory 72 responds to the control information. Memory module 82
. exceeds 10 frames per second.

Graphics video memory module 88 receives and stores graphics data from main computer 62 via system bus 64 . Graphics memory 88 provides video graphical overlays that modify portions of the CRT display to display user-defined features such as laser targets as well as display on-screen menus. Incorporation of such graphics into color display images may be accomplished in any suitable manner as known in the art. A high resolution true color cathode ray tube 90 receives image data on lines 92, 94, 96 and 98 from video memory module 73 in any suitable manner as is well known in the art. The CRT 90 displays color images and graphic overlays at an update rate of 60 frames per second. The system for generating and displaying true color images of the visual base also includes an image copressor 91 that receives encoded graphical data to form a graphical image, such as a laser target mark. It converts the converted coordinates and image data into a bit mapped image. Coprocessor 91 then transfers the bit-mapped images on system bus 64 to graphics memory 88 in any suitable manner as is well known in the art. Since all images and processing data have been converted to digital data, future It is possible to create an electronic log of the diagnostic/treatment session for reference. The workstation includes a disk controller 190 connected by a bus 192 to a tape 194 for storing program and image data, a winchester disk 196, and a floppy 198 drive.

The physician first sets the field of view by selecting the magnification level and initial orientation of the servo mirror 26, hereinafter represented as Xo and Yo. The clinician enters the desired magnification and position information via the keypad 102, which consists essentially of a joystick pointing mirror and function keys to set the magnification level. Main computer 62 sends a magnification adjustment signal on system bus 64 to zoom lens control unit 104 . Zoom lens control unit 104
is well known in the art to route appropriate control signals to control line 1.
06 to the zoom lens 48. The zoom lens 24 is adjusted to provide an ocular field of view between 10 and 60 degrees.

Main computer 62 sends orientation signals Xo and Yo on system bus 64 to servo digital signal processors 108 and 110, respectively. Furthermore, referring to Fig. 3,
The tracking servo mirror 26 utilizes a standard gimbal arrangement controlled by two sets of linear actuators 214 and 230 positioned on vertical axes, as is well known to those skilled in the art. Initial orientation signals Xo and Yo are stored in registers 200 and 202, respectively. register 2
00 is the X on the track 204.

is sent to the servo controller 208. The servo controller 208 is
o Convert flaw signal into digital positioning signal.

Servo controller 208 sends digital positioning signals on line 210 to digital to analog converter 212 . Digital to analog converter 212 sends an analog adjustment signal to X position actuator 214 of servo mirror 26 . The actuator 214 is controlled by a feedback system in which the diode laser beam is
A zero position photodetector (PSD) 216, which springs up from the reflective surface 25 on the back side of the primary reflective surface 27 of the servo mirror 26, detects the reflected beam in a manner well known in the art. PSD 216 sends the sensed position of the mirror on line 217 to analog to digital converter 218 . The absolute X position of the mirror is sent on line 219 to servo controller 208, which compares the actual position to the desired position and transmits error-corrected positioning data on line 210 accordingly. In operation, the servo mirror control feedback system provides a frequency response of several hundred kilohertz.

The method and apparatus for positioning the servo mirror on the Y-axis are essentially the same as those described above for adjusting the X-axis of the tracking servo mirror 26. Yo is received by register 202. The position of Yo is transferred to servo controller 224, which performs a set of control steps before transferring the digital adjustment signal on line 226 to digital to analog converter 228.

The output of transducer 228 is sent to Y-axis actuator 230. The position feedback signal is sent from PSD 232 on line 233 to analog-to-digital converter 234. Analog to digital converter 234 sends the absolute position of the servo mirror on line 235 to servo controller 224 . Servo controller 224 compares the absolute position to the desired position of servo mirror 26 and sends appropriate control signals to Y-axis positioning of servo mirror 26 to make further adjustments.

Once the clinician or user has identified a particular field of view at the base of the eye 40, the automatic tracking system illustrated in FIG. 4 operates to maintain a constant image of the field of view despite the patient's eye movements. . Xenon arc lamp 2 transmits a narrow band of light in the 80 nanometer wavelength range via fiber 20 to emission optics 122 .

This light is subsequently sent on optical path 124 to a narrow infrared (IR) bandpass filter 26. As a tracking beam in the future
The infrared light that passes through the filter 126 described continues on an optical path 128 to the beam splitter 14 .

Beam splitter 14 superimposes the tracking beam onto the illuminated illumination light beam on optical path 16 . The tracker beam tracks the equivalent ridge to the retina as the illuminating light beam path described above.

The net M40 partially reflects the weak tracking beam.

The light for the tracker passes through the contact lens 34 along the optical path 36.
to return to. Contact lens 34 images the reflected portion of the tracker beam that continues on optical path 32 to scanning eye optics 30 . The reflected tracker beam passes through the optical path 28
continues above, is reflected by the tracking servo mirror 26,
It continues on optical path 24 to beam splitter 22 . As previously mentioned, half of the light leaving the tracking servo mirror 26 on the optical path 24 is directed by the beam splitter 22 on the optical path 42 to the gicroic beam splitter 44. Substantially all of the transmitted tracker beam is reflected by splitter 44 in optical path 130 toward tracking miscellaneous optics 132 . Tracking miscellaneous optics 132 focuses the collimated infrared beam on optical path 138 to tracker CCD camera 140 by first passing the tracker beam on optical path 134 through a narrow band infrared filter 136. do. What should be noted is that the tracker CCD camera 140
The optically active surface of is aligned with the image plane of light passing from tracking servo mirror 26 to tracking miscellaneous optics 132. Furthermore, narrowband green light can also be used in place of infrared light for tracking purposes, with appropriate modifications to the apparatus for the filter.

Returning now to FIG. 4, when the user relinquishes control of the servo mirror 26 by indicating the desired field of view of the retina 40, a sample portion of the desired image of the retina 40 is processed by a geometric array parallel processor (GAPP). ) stored in unit 149. In the preferred embodiment, the dimensions of the stored reference image are 24 x 12 CCD cells. The selection of dimensions is chosen to maximize the speed of volume calculations resulting in image specimens large enough to generate reliably distinguishable images.

If the image is not large enough, the sample may not contain a discernible image and eye movements may not be reliably detected; Analog-to-digital conversion can be completed using standard integrated circuits. The tracker subsystem illustrated in FIG. The current image of the retina 40 is periodically compared to a reference image. The statistical correlation analysis technique described below is used to track eye movements.

Tracking of the movement of the eye 38 is accomplished in the following manner. As explained above, first, the reference image is the tracker CCD camera 140.
along optical path 138.

Each cell of tracking CCD camera 140 generates a signal corresponding to the intensity of light energy incident on the cell surface. The formatter 144 arranges the 6-bit grayscale data into slices of data, which are arranged in a GAPP array 149 (6 bits per bit slice).
bits (from the lowest bit to the highest bit). G.A.
The PP array 149 consists essentially of 3x6 or 3x37 arrays of NCR45CG F2 type cardiac systolic array integrated circuit devices. Processing engine 148 includes a systolic array integrated circuit along with summing circuitry (not separately shown). In addition to the above formatting procedures,
Formatter 144 also converts the 6-bit grayscale data to a two-level black/white signal that is used in a multiple-choice comparison algorithm whose exclusive OR is the two-level intensity values of the cells of the current image and the reference image.

CCD cell output is high (HIGH) or low (LOI)
II) is controlled by a user-defined modifiable threshold and is encoded by the formatter 14 by the main computer 62 on the system bus 64.
Transferred to 4. The digital signals representing the 24.times.12 cell reference images are then transferred on line 146 to GAPP unit 149 in bit slices of up to 64 bits at a time.

Once the reference image is stored in the GAPP array 149, the entire (24×12 cell) reference image is periodically compared to the current image to check and compare eye movements. In the preferred embodiment, the current image is
000 homologous books will be published. In the preferred embodiment 64 x
A 64-cell CCD camera 140 receives tracking images. However, only the 30x18 cell portion of camera 132 is actually used for tracking eye movements, thus allowing tracking within a 3 cell perimeter around the last known position of the reference image, as a reference image. do.

FIG. 5 illustrates the relationship of tracking image 300 to the last known position of reference image 302. The dimensions shown for the tracking image as well as the reference image are preferred, but the selection of other suitable dimensions will be apparent to those skilled in the art. The analog intensity signal is converted to digital data in the manner described above for encoding the reference image. Camera 140 then transfers the digital data on line 142 to formatter 144 .

To find the best match between a small region of 24X12 cells in a 30X18 cell tracking image and a 24X12 reference image, the reference image is oriented in either direction or a combination of directions from the center of the 30X18 window. 0.1.2
Or each 30x18 window displaced a distance of 3 cells is continuously compared with a subregion of 24x12 cells.

The format i'! The bit slices of digital image data on 146 are transferred to the GAPP unit 149 for each of the 49 possible 24×12 cell subregions of the 30×18 cell tracking image. The 49 small areas of the tracking image are transferred to GAPP 149,
It is compared to the 24x12 cell reference image in the appropriate order.

The preferred tracking subsystem illustrated in FIG. 4 utilizes correlation tracking techniques to calculate and transmit correction vectors to compensate for eye movements.

The reference value of the tracking image and the correlation value for the selected small area are stored in the control memory 1 that can be written on the track 156.
GAPP array under the control of signals transmitted from 54
149. For each subregion transferred to GAPP array 149 in the preferred tracking algorithm, the following operations are performed.

First, the CAPP array 149 has 64 cells (the width of the data bus 146 between the reference image and the current tracking image) between the reference image and the selected small area of the current tracking image. Simultaneously calculate differences in grayscale values.

Second, the difference between the gray school values of each cell in the reference image and the tracking image is then squared in parallel in the APP array 149 in a manner well known to those skilled in the art.

Third, the squared differences for all 288 cells (12 multiplied by 24) are then squared in parallel to arrive at the correlation value between a particular subregion of the tracking image and the reference region. will be added. A collection of 49 correlation values, referred to as a correlation matrix, is a reference image generated by the GAPP array 149 during each iteration of the tracking algorithm in which a mirror position correction vector is calculated by the tracking subsystem of FIG. This algorithm can be calculated as follows: i=1...12 k=-3...+3j=1...
24 1=-3...+3 Fourthly, the correlation matrix is transferred to the post-processing section 152 on the line 150,
This post-processing unit converts the value of the correlation matrix to the lowest correlation value (
The comparison is made to find a small area of the tracking image that has the highest degree of similarity to the reference image.

Fifth, the correlation value obtained from the fourth step is compared against the threshold C1 value to determine whether the "best match" subregion and the reference image are It determines whether there is sufficient similarity between them. If the threshold is exceeded, it is concluded that the tracking system has missed tracking, and the post-processing unit 152 sends the system bus 6 to the main computer 62 to report the error condition.
The error signal is transmitted on 4. However, if the threshold is not exceeded, the post-processing unit 152 sends correction vectors on lines 154 and 158 to correction registers 156 and 160 at the X and Y positions, respectively. Registers 156 and 16
0 sends the correction vectors on lines 206 and 222 to servo control units 208 and 224, respectively. The servo control units 208 and 224 are configured so that the current image is the sixth one.
The correction vector is used to determine a new set point for the servo mirror 26 so that it can be aligned with the reference image as shown.

As shown in FIG. 6, the "best match" of the current image 304 is two cell distances in the X direction and one cell distance in the Y direction from the position of the reference image 306. Displaced. Therefore, the post-processing unit in the illustrated correction cycle applies a correction vector of −2 in the X direction to Y
A correction vector of -1 is sent in the direction to align the current image with the reference image.

Still alternatively, the preferred tracking subsystem performs a tracking subsystem by It is possible to compare small regions of the image to a reference image. The encoded values are summed in a manner similar to that described above, and the lowest sum of the correlation matrix is compared to a threshold and registered in register 1 by the post-processor on lines 154 and 158, respectively.
The correction vectors transferred to 56 and 160 are determined.

This algorithm for comparing the reference image to a subregion of the tracking image can be calculated as follows: i=1...12 k=-3...+3j=1...
24 1=-3...+3 Due to the higher information content of the plan for encoding grayscale image data, the squaring algorithm is much more robust than the exclusive ORC technique.

A disadvantage of the square method is that it greatly increases the computational power required to achieve the required tracking cycle rate for approaching eye movements.

However, depending on the image quality, either the exclusive OR technique or the squaring technique is the optimal tracking algorithm. The exclusive OR technique is optimal if the image quality is high. Both of these methods are implemented in the preferred embodiment, and the main computer dynamically determines which method to use. The stability of the tracking system is approximately within the pixel resolution of the imaging CCD camera.

By adjusting the tracking servo mirror 22 to compensate for eye movements, the system maintains a steady image of the fundus for viewing and directing treatments to the fundus. In the example, the tracking system samples the current image and adjusts the tracking servo mirror 22 at a rate of 2000 times per second. If the field of view is 10 degrees, the maximum tracking speed is 117 degrees per second. When the field of view is 30 degrees, the maximum angular tracking speed is 351 degrees per second. field of view is 6
At 0 degrees, the maximum angular tracking speed is 707 degrees per second.

Referring again to FIG. 1, the physician may indicate a therapeutic target by keypad 102 or, alternatively, by keyboard 103. Referring again to FIG. The coordinates for the target are stored in the memory of the main computer 62. Main computer 62 sends target information on system bus 64 to image coprocessor 91.
Image coprocessor 91 generates a graphical image for the target marker and stores the marker on system bus 64 in graphical memory 8.
Send to 8. When the screen of the cathode ray tube 90 is made a new screen, the marker appears on it. Braun tube 90
The marker on the fundus image displayed above represents a point on the fundus at which the energy burst from laser 110 will be directed during treatment.

Laser treatment can begin immediately after selecting the laser target. The main computer 62 selects the target placed by the physician. The main computer 62 sends the laser mirror adjustment coordinates on the system bus 64 to the laser mirror control unit 180, and the laser beam 170 targets the retina 40 at a point indicated by a marker displayed on the cathode ray tube 90.
The laser beam is adjusted by the laser mirror 178 and the control unit 180 to direct the laser beam to the desired direction.

Laser mirror 178 and control unit 180 are essentially the same as the tracking servo mirror system shown in FIG.

The only difference is that the laser control unit does not use the X and Y mirror adjustment registers numbered 156 and 160, respectively, in FIG. This is because the laser mirror directs the laser according to the constant image of the base of the eye obtained from the adjustments made to the tracking mirror 26.

In the case of a single target, the target is highly illuminated on the cathode ray tube 90 and the target of the laser beam is verified by a low intensity prediction beam.

When the user confirms the treatment, the main computer 62 sends the ignition signal on the system bus 64 to the light/shutter controller 1.
Send to. Controller 1 sends an ignition signal on line 168 to laser 170 . The laser light travels on optical fiber 172 to collimator optical system 174 . Optical system 174 collimates the laser beam on optical path 176 with respect to laser servo mirror 178 . The settings of collimator optics 174 determine the spot size of the laser on the patient's retina. The laser beam is reflected from laser mirror 178 on optical path 182. The laser beam passes through transfer optics 184 and continues on optical path 186 to dichroic beam splitter 18 designed for the particular laser wavelength selected. From the beam splitter 18, the laser light hits the retina at one point corresponding to the indicated target displayed on the cathode ray tube 9o.
The light travels along the same optical path as other light.

The target may consist of a single line segment, in which case automated shock application will generally be effective due to its ability to achieve controlled and regular burns. It is possible to obtain superior results over the means impacting systems currently used in the art. In such cases, a set of coordinates is always sent to the laser mirror control unit 180 to form a straight line section from the collection of points.

In any case during laser treatment, if it is not possible to follow the eye movement, the tracking subsystem
2 transmits a tracking error signal on the system bus 64.

The illumination/shutter controller 1 reacts to the error signal by sending a cutoff signal on line 168 directly to the laser 170. Delivery of laser energy will not begin again until tracking is completed and the physician once again reviews the treatment plan displayed on the screen.

With the invention thus disclosed, if one skilled in the art encounters other embodiments of the invention including addition, subtraction, or variations of the disclosed embodiments, it is contemplated that the invention as defined in the claims may be modified. This should be done within the scope.

[Brief explanation of the drawing]

FIG. 1 is a block diagram illustrating a system according to an embodiment of the present invention, FIG. 2 is a block diagram illustrating an example of a subsystem for image generation in the system of FIG. 1, and FIG. FIG. 4 is a block diagram showing an example of the subsystem for controlling the servo reflector in the system shown in FIG. FIG. 6 is a diagram illustrating dimensions of an example of a tracking CCD that can be used with the system shown in FIG. 6. FIG. 6 is a diagram illustrating an image correction vector for correcting eye movement. (Explanation of symbols) 1... Illumination/shutter controller 2... Xenon arc lamp 6... Filter wheel 30... Scanning eye optical system 34... Contact lens 48... Zoom lens 60 ...Camera control unit 62...Workstation/main computer 68...
Frame grabber 2 Video storage device 90 Braun tubes 108, 110 Servo digital signal processor 140 Tracking CCD camera 174 Collimator optical system 180 Laser mirror control Unit 184... Transfer optical system 194... Tape 196... Winchiesta disk 198... Floppy procedural amendment (method) March 13, 1991 Mr. Satoshi Uematsu, Commissioner of the Japan Patent Office 1, of the case Display 1990 Patent Application No. 292184 2 Title of the invention Device for displaying electronic images of non-stationary objects 3 Address of the person making the amendment Address 5-8-10 Toranomon-chome, Minato-ku, Tokyo 105
Date of amendment order Kiitaroshima - 2812 days (unissued θ9 6, subject of amendment (1) “Representative of applicant” column of application (2) Power of attorney (3) Specification (4) Drawings 7 , Contents of the amendment (1) (2) As per attached sheet (3) Engraving of the description (no change in content) (4) Engraving of drawings (no change in content) List of attached documents (1) Application for correction 1 copy ( 2) Power of attorney and translation 1 copy each (3
) 1 copy of the engraving specification (4) 1 copy of the engraving drawing

Claims (10)

[Claims] 1. An apparatus for displaying a real-time, substantially motionless electronic image of a non-stationary object, the apparatus comprising: (A) object tracking means for tracking the movement of the non-stationary object; (B) A non-stationary object, comprising: inspection means for generating an optical image of the non-stationary object and forming an electronic representation of the optical image; and (C) display means for converting the electronic representation. A device that displays an electronic image of an object. 2. The object tracking means (A) generates optical energy having a first frequency band;
(B) means for generating and directing light energy having the first frequency band reflected from the non-stationary object through an adjustable optical path; (C) tracking receiving means for receiving the level sensing electronic signals and for converting the received optical energy into a set of level sensing electronic signals; Optical path correction means for generating a correction signal, comprising: (i) means for storing a test image representative of a desired image of the non-stationary object; (ii) storage of a current image of the non-stationary object; (iii) correlation means for comparing the current image and the test image to determine a displacement therebetween; and (iv) generating a first set of optical path adjustment signals in response to the displacement. (D) responsive to the optical path correction means to adjust the adjustable optical path in accordance with the adjustment signal to match the current image to the test image; 2. The apparatus of claim 1, comprising: means. 3. 3. The apparatus of claim 2, wherein said correlation means comprises means for performing statistical video correlation between said test video and said current video. 4. addressable storage means for the inspection means to identify and address any discrete point on the electronic display; and
2. The apparatus of claim 1, comprising means for retrieving said electronic representation from said addressable storage means. 5. 5. The apparatus of claim 4, wherein said inspection means further comprises marking means cooperating with said addressable storage means and said display means for indicating and recording a target mark within a field of said image. 6. 6. The apparatus of claim 5, further comprising processing means cooperating with said object tracking means and said inspection means for generating and directing a beam of optical energy toward a portion of said non-stationary object in accordance with said indicated target. Device. 7. The processing means comprises: (A) light energy emitting means for generating a processing beam of optical energy; (B) first path conditioning means cooperating with the marking means to direct the processing beam to the target; and 7. The apparatus of claim 6, further comprising: (C) second path adjustment means responsive to the target tracking means to adjust the path of the processing beam to compensate for motion of the non-stationary object. . 8. Apparatus for producing a real-time stabilized true color representation of an image of an object, the apparatus comprising a multi-wavelength light source emitting light in the red, green, and blue color wavelengths. , a color selection means for selectively emitting one color of light from the light source at one time, the color selection means being controlled so that each of the color of light is periodically and sequentially sent out; an adjustable light path for transmitting the emitted light to an object and returning from the object a portion of the emitted light that is reflected at the object; image detection means for generating an electronic display corresponding to a monochromatic image of the object; display means for displaying the monochromatic image in the colors of the emitted light, the image being different for each color of the emitted light; controlled to be displayed periodically and sequentially at a rate such as to provide a true color display in cooperation with the color selection means; and adjusting the light path. A device for generating a color display of an image of an object, comprising: tracking means for compensating the movement of the object. 9. The multi-wavelength light source comprises a broad spectrum arc lamp, and the color selection means is a filter wheel having segmentally mounted red, green, and blue filters; 9. The apparatus of claim 8, wherein the apparatus is adapted to rotate to transmit each of the red, green, and blue wavelengths of light in a periodic and sequential manner. 10. means for detecting the position of the filter wheel and generating a position signal in accordance with the transmitted color; and in response to the position signal, controlling the display means to display the monochromatic image in the color transmitted by the filter wheel. 10. The apparatus of claim 9, further comprising means for displaying.