Methods of Automated Determination of Parameters for Vision Correction

Abstract

A method for optimizing an ophthalmic treatment, comprising: measuring a patient's eye with an ophthalmic measurement instrument, fabricating a trial correction lens and testing it on the patient's eye, determining a score or success criteria for the trial correction, using the score or success criteria to provide training information to a machine-learning algorithm, and using the machine-learning algorithm to determine an optimal ophthalmic correction.

Claims

1. A method for optimizing an ophthalmic treatment of a patient's eye, comprising: a. measuring a patient's eye with an ophthalmic measurement instrument; b. choosing a trial correction contact and measuring the patient's eye using the trial correction fitted on the eye; c. determining a success criteria score for the trial correction; d. using the success criteria score to provide training information to a machine-learning algorithm; and e. using the machine-learning algorithm to determine an optimal ophthalmic correction.

2. The method of claim 1 where the ophthalmic instrument comprises a wavefront aberrometer, a corneal topographer, a profilometer, a tomographer, or combinations of these instruments.

3. The method of claim 1, wherein measurement information is recorded dynamically as a function of time.

4. The method of claim 1, wherein a fixation condition of the eye is controlled through active stimulus with astigmatism and defocus correction.

5. The method of claim 4, wherein target defocus is dynamically controlled.

6. The method of claim 2, wherein measurement information is synchronized with cloud-based or network-based storage to provide a central storage location.

7. The method of claim 1, wherein outcome information is recorded through an order process.

8. The method of claim 1, wherein the scoring information is recorded with an electronic medical records system.

9. The method of claim 1, wherein the score is selected from the group consisting of fit, lens stability, and level of correction, or combinations thereof.

10. The method of claim 1, wherein the ophthalmic correction comprises a soft contact lens.

11. The method of claim 1 wherein the ophthalmic correction comprises spectacle lenses.

12. The method of claim 1, wherein the ophthalmic correction comprises a gas permeable contact lens.

13. The method of claim 12, wherein the contact lens comprises a scleral contact lens.

14. The method of claim 1, wherein the ophthalmic correction comprises a result of laser refractive surgery.

15. The method of claim 1, wherein the ophthalmic correction comprises a phakic-IOL.

16. The method of claim 1, wherein the ophthalmic correction comprises a pseudo-phakic IOL.

17. The method of claim 15, further comprising using laser-induced refractive index change (LIRIC) for either multifocal correction or higher order aberration correction.

18. The method of claim 1, wherein the ophthalmic correction is implemented by changing an index of refraction of a native eye tissue.

19. A method for optimizing an ophthalmic treatment of a patient's eye, comprising: a. measuring a patient's eye with a wavefront aberrometer; b. choosing a trial correction contact and measuring the patient's eye using the trial correction fitted on the eye; c. determining a success criteria score for the trial correction; d. using the success criteria score to provide training information to a machine-learning algorithm; and e. using the machine-learning algorithm to determine an optimal ophthalmic correction.

20. A method for optimizing an ophthalmic treatment of a patient's eye, comprising: a. measuring a patient's eye with a wavefront aberrometer; b. choosing a trial correction contact and measuring the patient's eye using the trial correction fitted on the eye; c. determining a success criteria score for the trial correction; d. using the success criteria score to provide training information to a machine-learning algorithm; and e. using the machine-learning algorithm to determine an optimal ophthalmic correction; f. wherein measurement information from the wavefront aberrometer is synchronized with cloud-based or network-based storage to provide a central storage location.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 show a flow chart of a first example illustrating generalized architecture and steps for data storage and transfer for an ophthalmic instrument, according to the present invention.

[0012] FIG. 2 shows a flow chart of a second example illustrating more detail on the cloud-based architecture data collection, according to the present invention.

[0013] FIG. 3 shows a flow chart of a third example illustrating an alternative arrangement where more of the data collection features are cloud-based, according to the present invention.

[0014] FIG. 4 shows a flow chart of a fourth example illustrating depicts that steps that are typical in a measurement/fitting cycle, according to the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0015] Certain embodiments of an ophthalmic system can comprise (according to the present invention): [0016] An ophthalmic device for measuring wavefront aberrations of a patient's eye; [0017] An ophthalmic device for measuring cornea shape and elevation; [0018] An ophthalmic device for measuring iris and pupil dimensions; [0019] Means for dynamic capture of ocular measurements; [0020] Means for collecting and storing information; [0021] Means for measuring the eye in a controlled state of fixation (e.g., astigmatism, defocus); [0022] Means for dynamically measuring the eye; [0023] A method for generating correction prescriptions; and [0024] A processor (local or cloud-based) for implementing a Machine Learning (ML) calculation.

[0025] The ophthalmic instruments described in co-pending U.S. application Ser. No. 17/183,327 includes all of the measurement capabilities described above. It is capable of dynamic measurements, and stores information in rapid sequences to local computer storage. This same instrument includes controls for the fixation target which allow for control of the target astigmatism and defocus (or fogging) in a controlled manner. This provides all the information needed for determining optimal refraction, aberration correction, tear film assessment, contact lens fitting, and spectacle fitting.

[0026] The information is stored locally, and then synchronized to a cloud-based storage system for further analysis.

[0027] The system can include an on-line part ordering or tracking system that tracks the orders placed by the eye care practitioner. This is correlated with the measurement data to provide scoring for the predictions. This is used to train the AI algorithm. Both success and failure data are important to improve the efficiency of the AI algorithm training.

[0028] FIG. 1 depicts a flow chart of a first example illustrating generalized architecture and steps for data storage and transfer for an ophthalmic instrument, according to the present invention. Data is acquired by the instrument and is stored on local media. This data may consist of movies of aberrometry, images, corneal topography or other processed data. The acquired data is stored along with conditions of the measurement (target correction and fixation, illumination levels, etc).

[0029] After the measurement sequence is acquired, it is synchronized to a cloud storage system. This can either be a local intranet system with a dedicated server, or an Electronic Medical Records (EMR) system, or one of the many common cloud storage platforms (OneDrive, Google drive, Dropbox, etc). Preferably this data is secured with encryption that is medical device data compliant. Once the data is synchronized to the cloud-based storage, it can be viewed and analyzed by any number of separate data viewer/analysis programs. In addition, this data is available to the AI system, which is also cloud based. There are several pre-existing AI platforms, including Google AI Platform, Microsoft Azure, and Amazon AWS Machine Learning. There are other AI systems that may be built on local platforms, either with native or through the use of a standard platform (e.g., MATLAB).

[0030] FIG. 2 shows a flow chart of a second example illustrating more detail on the cloud-based architecture data collection, according to the present invention. As the measurement system is used to measure patients, some of this data is used by the eye care provider (ECP) to prescribe a prescription. This may be a refraction used for spectacles or contact lenses, or it may be more detailed, including: CL base curve, astigmatism, axis, higher order aberrations, or other features. This data is sent to the contact lens manufacturing laboratory using an order process which records the desired order parameters. Typically, the doctor uses an EMR system to track patient records. This will also record various outcome measures, including fit success and visual acuity (VA). This combination of order record with outcome success can be fed to the AI algorithm to provide training datasets.

[0031] FIG. 3 shows a flow chart of a third example illustrating an alternative embodiment where more of the data collection features are cloud-based, according to the present invention. Most EMR systems are now cloud-based to provide portability across platforms. Thus, the patient information, outcome measures, prescription history are already available in the cloud. The order process can also be cloud-based. The operations that take place outside the cloud are all physical interactions with the patient or correction (measuring the patient, manufacturing the contact lens or spectacles, and test fitting the correction on the patient). These can also be monitored with systems that could provide information to the cloud-based storage (re-measuring the patient plus correction with the ophthalmic instrument and/or correction metrology with cloud storage-linked equipment). The more the data collection can be automated, the more data will be available to the AI system.

[0032] FIG. 4 shows a flow chart of a fourth example illustrating depicts that steps that are typical in a measurement/fitting cycle, according to the present invention. The patient is measured with an ophthalmic device, which transmits its data to cloud-based storage (as described above). The appropriate contact lens is manufactured and then test-fit on the patient. The initial treatment design methodology can be anything that is familiar to the Eye Care Practitioner (ECP). The correction is then tested on the patient with any number of methods. This might be a slit lamp exam for contact lens fitting, or another measurement on the ophthalmic instrument. Outcome measures (lens comfort, visual acuity) may also be recorded (typically saved in the EMR). After test fitting, the ECP will make a decision about the correction. Either it doesn't work well enough, or it is acceptable and correct. It is important to record both of these outcomes so that the AI system will have both positive (success) and negative (poor outcome) measures to collect. The algorithm “learns” by correlating the score (outcomes) with the measurement data to provide an optimized treatment correction.

[0033] This methodology can be applied to several different ophthalmic corrections. This can include soft contact lenses, hard contact lenses, scleral lenses, spectacles, phakic, or pseudo-phakic Intraocular Lens (IOL), refractive surgery, cataract surgery, or LIRIC treatment. Improved measurement of the eye in a known state, and measurement of accommodative range, will provide improvements for presbyopia treatment, as well.