System and method for remote optical monitoring of intraocular pressure

Abstract

Provided herein are devices, systems, and methods for determining intraocular pressure (IOP) of an eye. In some embodiments, at least two markers are implanted on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof. The intraocular pressure (IOP) of the eye is determined by reading first image data, analyzing the first image data using a trained deep neural network, and annotating the first image data using the analysis produced by the trained deep neural network.

Description

Technical Field

[0001] (Cross - reference to Related Applications) This application claims the benefit of U.S. Provisional Patent Application No. 63 / 489,680, filed Mar. 10, 2023, which is hereby incorporated by reference in its entirety.

[0002] This application relates to U.S. Provisional Application No. 62 / 790,752, filed Jan. 10, 2019, titled "METHOD AND DEVICE FOR REMOTE OPTICAL MONITORING OF INTRAOCULAR PRESSURE", and U.S. Patent Application No. 16 / 124,630, filed Sep. 7, 2018, titled "CLOSED MICROFLUIDIC NETWORK FOR STRAIN SENSING EMBEDDED IN A CONTACT LENS TO MONITOR INTRAOCULAR PRESSURE", the contents of each of which are hereby incorporated by reference in their entirety.

[0003] The present disclosure relates to systems and methods using a wearable optical imaging sensor system for measuring intraocular pressure.

Background Art

[0004] Glaucoma is the second most common cause of blindness worldwide. It is a multifactorial disease with several risk factors, among which intraocular pressure (IOP) is the most important. IOP measurement is used for glaucoma diagnosis and patient monitoring. IOP has wide diurnal variations and depends on body posture, so that spot measurements taken by ophthalmic medical specialists in a clinic can be misleading.

Summary of the Invention

Means for Solving the Problems

[0005] Provided herein are devices, systems, and methods for determining intraocular pressure (IOP) of an eye. Aspects of this disclosure describe a method for determining the IOP of an eye, comprising the steps of: acquiring or receiving a first image of at least two markers implanted in the eye, detected using a first detector, and a second image of at least two markers, detected using a second detector, wherein the optical axis of the first detector is at an angle with respect to the optical axis of the second detector; and determining the IOP of the eye from the positions of one or more of the at least two markers from the first image or the at least two markers from the second image.

[0006] In some embodiments, at least two markers are embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof. In some embodiments, at least two markers include fluorescent markers. In some embodiments, the method further includes the step of providing a light source to at least two markers embedded in the eye and acquiring or detecting a first or second image. In some embodiments, the light source includes light-emitting diodes. In some embodiments, the light source, the first detector, the second detector, or a combination thereof is at least partially contained within a chassis. In some embodiments, the chassis includes eyewear, glasses, goggles, a head-up display, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

[0007] In some embodiments, a first image of at least two markers implanted in the eye is acquired or detected by a first detector through a first polarizing filter, and a second image of at least two markers implanted in the eye is acquired or detected by a second detector through a second polarizing filter.

[0008] In some embodiments, at least two markers include particles. In some embodiments, the at least two markers comprise a first pair of markers and a second pair of markers. In some embodiments, the eye's IOP is determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers. In some embodiments, the positions of at least two markers are provided as input to a machine learning algorithm or predictive model, which is trained to provide an output related to the eye's IOP.

[0009] Described herein is a system for determining the IOP of an eye, comprising: a detector optically coupled to the eye, configured to detect signals from at least two markers implanted in the eye; and one or more processors electrically coupled to the detector, configured to process signals from two or more markers implanted in the eye and to determine the IOP of the eye from the positions of the two or more markers. In some embodiments, the at least two markers are positioned at a distance of up to about 6 millimeters from each other.

[0010] In some embodiments, the system further comprises a light source optically coupled to the target eye. In some embodiments, the light source includes light-emitting diodes (LEDs). In some embodiments, the light source and detector are at least partially contained within a chassis, which is positioned at a distance of up to approximately 2 centimeters from the tangential surface of the eye. In some embodiments, the chassis includes eyewear, glasses, goggles, a head-up display, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

[0011] In some embodiments, the detector includes a first detector and a second detector. In some embodiments, the optical axis of the first detector is at an angle to the optical axis of the second detector. In some embodiments, the detector includes a camera, a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or any combination thereof.

[0012] In some embodiments, the system comprises one of the devices used in the methods described herein.

[0013] Provided herein is a system for determining the IOP of an eye, comprising one or more processors and memory for storing one or more programs for execution by one or more processors, the one or more programs including instructions for (i) acquiring or receiving images of one or more markers embedded in an eye, the one or more images being detected using a first detector and a second detector, the optical axis of the first detector being at an angle with respect to the optical axis of the second detector, and (ii) determining the IOP of an eye from the distance between the at least two markers in one or more images. In some embodiments, the at least two markers are positioned at a distance of up to about 6 millimeters from each other. In some embodiments, the one or more images are acquired or received from a server, cloud-based storage device, eyewear device, or any combination thereof. In some embodiments, the eyewear device includes virtual reality eyewear, augmented reality eyewear, or a combination thereof. In some embodiments, the one or more processors reside on the eyewear device, teleprocessing device, remote server, cloud server, or any combination thereof. In some embodiments, the system comprises one of the devices used in the methods described herein.

[0014] Described herein is a method for training a machine learning algorithm or predictive model to determine the IOP of an eye, comprising the steps of receiving or acquiring one or more images of at least two markers embedded in one or more eyes and corresponding IOPs of one or more eyes, and training an untrained or partially untrained machine learning algorithm or predictive model using one or more images of at least two markers embedded in one or more eyes and corresponding IOPs of one or more eyes, thereby producing a trained machine learning algorithm or trained predictive model. In some embodiments, the untrained or partially untrained machine learning algorithm or predictive model is further trained with respect to the distance between at least two markers in one or more images of at least two markers embedded in one or more eyes.

[0015] In some embodiments, images of one or more markers implanted in one or more eyes are detected using a first detector and a second detector, the optical axis of the first detector being at an angle to the optical axis of the second detector. In some embodiments, the at least two markers are positioned at a distance of up to approximately 6 millimeters from each other. In some embodiments, the at least two markers are located on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof.

[0016] In some embodiments, systems described elsewhere herein are used to train machine learning algorithms or predictive models to determine the eye's IOP.

[0017] Additional aspects and advantages of the present disclosure will become readily apparent to those skilled in the art from the following detailed description, which illustrates only exemplary embodiments of the present disclosure. As will be recognized, the present disclosure is capable of other and different embodiments, and some of the details thereof are capable of modification in various respects without departing from the present disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature and not restrictive. (Incorporation by reference)

[0018] All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

Brief Description of the Drawings

[0019] The novel features of the present invention are set forth specifically in the appended claims. A more thorough understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description and the accompanying drawings, which describe exemplary embodiments in which the principles of the present invention are utilized.

[0020] [Figure 1] FIG. 1 illustrates a schematic diagram showing lateral variations in corneal thickness according to an exemplary embodiment described herein.

[0021] [Figure 2] FIG. 2 shows a table of structural eye parameters of a model eye according to an exemplary embodiment described herein.

[0022] [Figure 3] FIG. 3 illustrates a side view of an instrument used within the devices, systems, and methods of the subject matter according to an exemplary embodiment described herein.

[0023] [Figure 4A]Figure 4A illustrates a schematic cross-section of a cornea with two implants positioned on opposite sides of the central axis of the cornea, according to an exemplary embodiment described herein.

[0024] [Figure 4B] Figure 4B illustrates IOP as a function of the calculated distance between corneal implants, according to an exemplary embodiment described herein.

[0025] [Figure 5] Figure 5 illustrates the calculated distance between corneal implants as a function of the positioning angle in degrees, according to an exemplary embodiment described herein.

[0026] [Figure 6] Figure 6 illustrates a front view of an eye with the positions of fluorescent implants on the cornea and sclera, according to an exemplary embodiment described herein.

[0027] [Figure 7] Figure 7 illustrates an imaging goggle in which two cameras (visible from the front and side) can be used to image the cornea and implants, according to an exemplary embodiment described herein.

[0028] [Figure 8] Figure 8 illustrates a schematic side view of an eye with the positions of two fluorescent implants on the cornea, according to an exemplary embodiment described herein.

[0029] [Figure 9] Figure 9 illustrates a schematic front view of an eye with the positions of four fluorescent implants on the cornea, according to an exemplary embodiment described herein.

[0030] [Figure 10] Figure 10 illustrates an example of the measured distance between two fluorescent particles on a model cornea when subjected to various internal pressures, according to an exemplary embodiment described herein.

[0031] [Figure 11] Figure 11 illustrates the positioning of the imaging illuminator, corneal implant, and a pair of observation cameras relative to the eye according to an exemplary embodiment described herein.

[0032] [Figure 12] Figure 12 illustrates an exemplary flowchart of the steps for training a deep neural network according to an exemplary embodiment described herein.

[0033] [Figure 13] Figure 13 illustrates an exemplary flowchart of the steps using a deep neural network according to the exemplary embodiments described herein. [Modes for carrying out the invention]

[0034] Detailed explanation Patients with severe glaucoma require intermittent monitoring of their intraocular pressure (IOP) to maintain the health of their optic nerve and prevent vision deterioration and loss. Conventional methods for measuring IOP require patients to visit their ophthalmologist and / or optometrist and have their IOP measured using in-office devices, which can be cumbersome and inconvenient for the patient, requiring frequent or intermittent monitoring of IOP. Therefore, there is an unmet need for devices, systems, and / or methods that can measure IOP conveniently and accurately.

[0035] An increase in the eye's IOP results in corneal bulging and, consequently, a change in the corneal and / or ocular radius of curvature. IOP changes affect the structural characteristics of the cornea, causing changes in corneal radius and apex height relative to the corneal margin. A change in IOP of approximately 1 mmHG can be determined and / or monitored by measuring the difference in corneal radius with a resolution and / or scaling of approximately 4 micrometers. Therefore, there is an unmet need for an IOP measuring device, system, and / or method that can take multiple measurements of a subject's eye as the subject goes through its normal routine, and / or continuously monitor its curvature, in order to determine changes in IOP, for example, throughout the entire day, which would otherwise require a visit to the ophthalmologist's and / or optometrist's office. There is also a need for a device with sufficient sensitivity and / or accuracy in measuring ocular curvature to produce reliable data for accurate determination of IOP. Furthermore, there is a need for such a device to operate in a manner that does not interfere with the patient's normal vision and activity. Furthermore, there is a need for a device that can reliably function while the patient continues their normal daily activities, and this device does not require any particular critical positioning or alignment with the patient's eye. There is still a need for the device to be user-friendly.

[0036] Devices, systems, and methods for measuring, acquiring, and / or determining the IOP of an eye using eyewear positioned at a certain distance from the eye are described herein. In some cases, the eyewear may comprise one or more illuminators, one or more image sensors, or a combination thereof. A combination of one or more illuminators and one or more image sensors may eliminate one or more ambient illumination effect variations and / or mismatch errors that would otherwise introduce noise and errors into the corneal radius measurement and / or the resulting IOP determination. Devices, systems, and / or methods described elsewhere herein may measure changes in the radius of curvature of the cornea (for example, in the case of an adult cornea, at about 4 micrometers per about 1 mmHg change in IOP). This change along the corneal plane normal may be difficult to measure using conventional visible light imaging systems. One or more particles and / or markers (e.g., fluorescent or dye-doped microparticles) may be embedded at a predetermined location on the cornea to convert the change in corneal curvature into vertical and / or horizontal displacements of one or more particles and / or markers. Changes in corneal shape in response to changes in IOP may result in changes in the position and / or distance between corneal particles and / or marker implants, which can be measured by imaging the positions of the particles and / or markers. In some cases, the vertical and / or horizontal displacements of one or more particles and / or one or more markers may be measured, detected, and / or determined by obtaining one or more images of one or more embedded particles using one or more image sensors. One or more images of one or more embedded particles may be processed using image processing methods, one or more machine learning algorithms, one or more predictive models, or any combination thereof, as described elsewhere herein, to measure the positions (e.g., relative positions and / or inter-particle distances) between one or more particles embedded in the cornea. The optical design may enable image processing and sensor fusion. In some cases, a marker may be a type of sensor. A marker may be a type of sensor that signals its own position.The position can be determined by combining image processing and sensor data. Markers can also be magnetic particles or similar entities. Measured changes in position and / or interparticle distance may be used in calculations using machine learning programs, learning neural networks, artificial intelligence programs, other analytical calculation programs, or any combination thereof, to correlate measured changes in the position of one or more particles with changes in corneal radius. In some cases, changes in corneal radius may be converted into changes in the eye's IOP. Methods described elsewhere herein may also use preliminary characterization of corneal thickness and structural features, where the radius of curvature at a known IOP is obtained by conventional ophthalmic methods (e.g., Goldmann applanation tonometry (GAT), Tonopen tonometry, Pneumo tonometry, non-contact or air-exhaust tonometry, dynamic contour tonometry (DCT), or any combination thereof). Corneal curvature data of an object obtained using eyewear, as described elsewhere herein, may then be compared with a dataset of known IOPs and corresponding corneal curvatures to calculate the IOP. In some cases, a dataset of known IOPs and corresponding corneal curvatures may be used to train one or more partially untrained and / or untrained machine learning algorithms and / or predictive models. The dataset may be processed and / or manipulated by a computing device such as a mobile phone, personal computer, laptop computer, server, cloud processing server, or any combination thereof. In some embodiments, this disclosure describes a wearable optical device that measures IOPs through image acquisition from one or more image sensors and uses image data in addition to reference data about a particular individual to accurately determine the IOPs.

[0037] These and other objectives can be satisfied using the methods, devices, and / or systems described herein. In various embodiments, this disclosure describes a method for remotely measuring IOP. The method may include the implantation of fluorescent beads at a specific location within the cornea and one or more devices for imaging the position of the fluorescent particles using a fluorescence imaging system comprising an excitation light source, an emission filter, a camera, or any combination thereof. The system may include a device for imaging the relative position of the fluorescent particles implanted in the cornea and a microcontroller configured to receive and / or acquire the measured position of the particles in order to measure the intraocular pressure (IOP) of the eye. In various embodiments, the IOP may be determined using positional measurements of fluorescent dye-doped microparticles. In some embodiments, one or more of these elements may be replaced with uniform elements. In some embodiments, the fluorescent dye-doped microparticles may be replaced with colored microbeads. The reading device may be a pair of goggles, glasses, or other eyewear.

[0038] In some embodiments, an eyewear device for measuring intraocular pressure (IOP) may exist. The device may have a frame and a first lens mounted on the frame such that the lens may be within the field of view of the person wearing the eyewear device. The eyewear device may have a first illumination source positioned to illuminate the user's eye with an excitation wavelength, typically green; an emission filter in front of a camera to eliminate transmission of excitation light; a first image sensor positioned to capture an image of the user's eye; a first communication portal communicating electronically or signal-wise with a calculation device; or any combination thereof.

[0039] In some embodiments, a method for training an image processing pipeline may exist. This method may involve the steps of: collecting personalized ophthalmic data regarding the user's anatomical structure and corneal properties at known IOPs; collecting personalized data from an eyewear device for measuring IOPs; and generating at least one set of training data for a neural network component pipeline using at least one computation mode and ray tracing under one or more geometric configurations. In various embodiments, the computation device may be a mobile phone, tablet, laptop computer, or any combination thereof. The computation device may be attached to a wearable eyewear device.

[0040] In some embodiments, this disclosure describes a method for determining intraocular pressure (IOP) of an eye. This method may involve the steps of reading out first image data, analyzing the first image data using a trained deep neural network, and annotating the first image data using the analysis produced by the trained deep neural network.

[0041] This disclosure describes wearable eyewear, systems, and methods for measuring the cornea of ​​the eye and determining the intraocular pressure of the eye measured based on changes in corneal curvature. This disclosure describes a calculation device that can calculate IOP values ​​based on corneal implants, eyewear, and / or corneal deformation data collected by the eyewear. This disclosure also describes a method that can calculate IOP. Where the terms “eyewear device” or “eyewear” may be used, they are intended to be used synonymously herein, and any reference to “eyewear device” or “eyewear” is understood to mean any of the wearable eyewear systems, apparatus, and devices described herein unless the context specifically indicates otherwise.

[0042] Eyewear as described herein can take various forms. The shape factor may be one of the options for the user, or one of the options for the user's optometrist or other medical professional concerned with the user's eye health. In some embodiments, the shape factor may include a frame and a lens. The frame may be one that the user can wear in front of their eyes (note that the use of male or female pronouns may be randomly distributed herein and is synonymous with respect to human subjects and / or patients). The disclosed technology is gender independent of the user. The synonymous use of user or other human gender described herein is solely for the convenience of the application. The frame may be any type of eyewear frame used for modern eyewear, including frames for sunglasses, corrective glasses, safety glasses, and all types of goggles (e.g., swimming, sports, safety, skiing, etc.). The frame may be suitable for a single lens for one eye, a lens for two eyes (e.g., a visor), or a single lens and eye cover (e.g., for people with amblyopia or who may suffer the loss of one eye). The lenses may be prescription lenses for visual correction, clear or tinted lenses for appearance, or opaque lenses that cover the eyes. In some embodiments, the lenses may have an area defined across the user's field of vision. The field of vision may be clear to avoid obstructing the user's vision. Various elements of the eyewear device may be mounted on the periphery of the lens, on the frame, or a combination thereof. The frame or lens may have flanges or other protrusions and / or tabs for mounting image sensors, light sources, batteries, computing devices, any other components, or any combination thereof, which are suitable for use with the present disclosure.

[0043] Wearable eyewear may have one or more image sensors positioned to face the user's eye so that the image sensors can capture an image of the eye. The image sensors may be a camera, a CCD (charge-coupled device), a CMOS (complementary metal-oxide-semiconductor) or other image acquisition technology. Wearable eyewear may have one or more light sources for projecting light onto the eye. In some embodiments, the light source may be some form of illumination that produces a specific wavelength of light. The light emission may be at a shallow angle to the curvature of the cornea and projected outside the lens portion of the eye so that the light does not interfere with the user's normal vision. In some embodiments, the light source may be an LED (light-emitting diode), and in other embodiments, the light source may be any light generation technology that is currently known or still undeveloped.

[0044] In various embodiments, the light source and image sensor may be positioned such that the image captured by the image sensor can ignore ambient light, glare, other optical artifacts, or any combination thereof, which may interfere with the accurate reading of changes in corneal curvature. The light source and image sensor may use one or more polarizing filters to sufficiently reduce and / or eliminate light of specific polarizations, wavelengths, intensities, or any combination thereof, so that the captured image may have better reliability and less signal noise. In some cases, the eyewear may have an optical sensor that helps adjust the ambient lighting effect conditions when they are appropriate for capturing a suitable image of the eye for determining corneal curvature. The image captured by the image sensor may be stored locally over a time period and / or transmitted to a computation device via a communication portal as described elsewhere herein.

[0045] In some embodiments, the communication portal may be an antenna for wireless transmission of data to the computing device. The communication portal may send and receive information, such as a step of transmitting image data and / or a step of receiving medication information for a drug delivery device. In various embodiments, the computing device may be a mobile phone, tablet computer, laptop computer, personal computer, any other computing device, or any combination thereof, which the user may choose to perform program (App) functions for the eyewear device. In some embodiments, the computing device may reside on the eyewear. In some embodiments, the communication portal may be a wired connection between the image sensor, light source, computing device, power source for all electrical components, or any combination thereof. In some cases, the communication portal may connect the eyewear to the cloud.

[0046] In some embodiments, this disclosure describes a method for determining the IOP of an eye. In some embodiments, the method may use a basic operation pipeline. The pipeline may receive image data from various sources. In some embodiments, the image data may arise as eyewear is fitted by a user and / or subject. In some embodiments, the image data may arise from a database having stored ophthalmic data of the user and / or subject at a fixed point in time. In some embodiments, the image may be anatomical data of the user from a fixed point in time. In some embodiments, some or all of the available image data may be used within a deep neural network with an image processing frontend. The image processing frontend may derive or calculate IOP readings. In some embodiments, the IOP readings may be updated at a video data rate that provides a near real-time output.

[0047] In some embodiments, the data pipeline may cause the image sensor to modify exposure level, gain, brightness, contrast, or any combination thereof to capture a non-saturated image. The image may be passed through a threshold filter to reduce or eliminate background noise. Several high-resolution images may be stored in temporary memory for rapid processing while blurred and / or low-resolution images are formed. The low-resolution images may then be passed through a matching filter and / or a feature detection filter to accurately indicate spots corresponding to one or more particles and / or markers illuminated by illumination / light sources in various captured images. The rough locations of one or more particles and / or markers may then be used to segment the high-resolution image and perform a peak fitting algorithm to individually determine the position and width of each peak in the image. The results of the peak locations and widths may then be used in conjunction with a pre-trained neural network, which may then be used to estimate the corneal coordinates and radius of curvature. A nonlinear equation solver may be used to convert the radius of curvature to IOP readings.

[0048] As described herein, a wearable eyewear device may be coupled to a calculation device to measure the IOP of the user's eye. The user and / or subject may be a person wearing the eyewear unless the context of use clearly indicates otherwise.

[0049] Various components and images are referenced herein. The use of names is intended to help the reader to further understand this disclosure. In particular, singular versions of nouns are used in many cases, but it should be understood that embodiments take into full consideration that multiple numbers of components and images are also within the scope of this disclosure.

[0050] Referring here to Figure 1, a cross-sectional view of human cornea model 102 is shown. The aspherical model parameters are given in Figure 2. The schematic diagram in Figure 1 shows distance in millimeters (mm) on the Y-axis and corneal curvature, i.e., the difference in surface deflection, on the X-axis. The cornea can be defined by an aspherical surface that deviates from the spherical surface, with parameters given in Figure 2. The corneal thickness may be approximately 0.91 mm at the edge and decreases to approximately 0.6 mm at the apex. Depending on the change in IOP, the corneal shape can be treated as fixed at the edge in contact with the sclera, which leads to a change in the radius of curvature of the cornea. The anterior surface is illustrated using solid lines, while the posterior surface is illustrated using dashed lines.

[0051] Figure 2 shows Table 202 of the structural parameters of a new schematic eye. The model eye shows the direction of angle alpha and pupillary eccentricity. The table gives the radius of curvature, as well as aspheric parameters and refractive index, across the anterior and posterior surfaces of the cornea.

[0052] Referring here to Figure 3, various instruments 302 for generating pockets for containing particles, markers, and / or fluorescent beads (304) as described elsewhere in this specification are illustrated. In some embodiments, the implantation area may be selected on the surface of the eye. The location of the implantation area may be on the cornea, lens, other surface area of ​​the eye, or any combination thereof. In some embodiments, one or more implantation areas may be located in tissue adjacent to the eye to serve as a reference point for measurement. In some embodiments, the implantation area in the cornea for housing one or more particles, markers, and / or fluorescent beads may be generated using a femtosecond laser. The laser may be used to penetrate the cornea to a desired corneal depth so that the generated pocket can be large enough to hold the fluorescent beads, particles, and / or markers. In some embodiments, the tunnel may also be fabricated using the laser according to guide dimensions that can be entered at the horizontal distance closest to the pocket generated in the implantation area. The incision created inside the eye using a laser may be opened using a spatula or other instrument, and fluorescent beads, particles, and / or markers may be placed in the pocket or tunnel with the help of a guide.

[0053] The corneal incision, which will be formed as a bead, particle, and / or marker pocket, may also be formed using a microsurgical knife 303, which can make an incision according to the dimensions of the beads, particles, and / or marker to be placed to the intended depth. In some embodiments, the depth may be about 5 to about 300 microns. In some embodiments, the depth may be about 50 to about 250 microns. In some other embodiments, the depth may be about 150 to about 200 microns. In some embodiments, the depth can be greater than 5 microns, greater than about 45 microns, greater than about 85 microns, greater than about 125 microns, greater than about 165 microns, greater than about 205 microns, greater than about 245 microns, greater than about 285 microns, or greater than about 300 microns. In some cases, the depth can be less than approximately 5 microns, less than approximately 45 microns, less than approximately 85 microns, less than approximately 125 microns, less than approximately 165 microns, less than approximately 205 microns, less than approximately 245 microns, less than approximately 285 microns, or less than approximately 300 microns.

[0054] The incision may be made using a microsurgical knife 303 in a horizontal setting, and beads, particles, and / or markers may be delivered to the implantation area via the guide 301. The implantation guide 301 may carry one or more fluorescent beads, one or more particles, and / or one or more markers 304, which can be delivered into the pocket. In some embodiments, the guide may have a silicone tip that can assist in performing a corneal incision.

[0055] Referring to Figure 4A, a cross-section 402 of a model cornea is shown with two fluorescent microparticles and / or marker implants 403. In some embodiments, each fluorescent particle and / or marker may be placed in a surgically generated pocket. Each pocket may be molded to hold the particle and / or marker in place and prevent the particle and / or marker from drifting, moving, or otherwise migrating. The pocket may be sealed or otherwise treated to reinforce the natural tissue and prevent particle and / or marker migration. This reinforcement may involve the addition of supplemental nutrients, growth factors, or several artificial materials that can strengthen the pocket to accelerate corneal healing.

[0056] In various embodiments, the observed distance between implants 401 may be measured in millimeters, or as the ratio of the lengths of the distances between two or more pairs of implants (e.g., particles and / or markers), or as a measurement from one particle and / or marker to an artificial reference point (fixed-position implant / particle) or a natural reference point (center of the eye, location of a known structure, etc.). In various embodiments, the (actual or observed) distance between implants (or between implants and reference points) may be measured to help determine intraocular pressure.

[0057] In some embodiments, the angle 405 of the implants along the visual axis (or cylindrical axis of symmetry, e.g., the optical axis of the eye), such as measured from the central axis of the pupil, may be used to help determine the optimal positioning of fluorescent beads, particles, and / or markers whose separation is most sensitive to changes in IOP. The determination of the angle is mathematically equivalent to the selection of the radius in which the fluorescent particles and / or markers are embedded.

[0058] Referring here to Figure 4B, the calculation of the distance between the implants 401 as a function of IOP404 is given. The corneal surface along the XZ plane, which can be described according to the relationship between the instantaneous tangential radius of curvature ρ(x) and the degree of asphericity Q, can be derived as follows. In the perpendicular (Y) meridian, the conical division is expressed as follows.

[0059] x 2 +(1+Q)z 2 -2zR=0(equation 1)

[0060] In the equation, Q and R are the asphericity and radius of curvature of the corneal surface, respectively. Based on Equation 1, assuming a displacement of approximately 4 micrometers at the corneal apex in response to an IOP change of approximately 1 mmHG, and assuming that the scleral / corneal interface is not affected by the IOP change, the relative distance between two implants can be calculated. The calculation assumes that the edge of the cornea in contact with the sclera is fixed, and that the cornea bulges in response to an increase in IOP. This perturbation is assumed to be a primary value that affects the radius of curvature, and not that it affects the asphericity parameter Q. (Assuming a total corneal size of 12 mm) To estimate the corneal surface position at a given IOP for a corneal opening of x = -6 to 6 mm, R(IOP) = R(IOP ref )-(IOP-IOP ref We assume a radius of curvature obtained by ) × 0.004 mm. The Z position of the corneal surface is calculated by numerically solving equation 1, and the minimum displacement amount at the edge of the cornea (z min ) was found for x=-6, and the overall Z(x) curve is the minimum value of the corneal position z minThe displacement is calculated by subtracting . This is to emulate the assumption that the scleral / corneal interface does not change its height in response to changes in IOP. The resulting corneal position is then used to calculate the X and Z displacements of the implant. For a first-order approximation, assuming that the angle between the implanted fluorescent particle and / or marker and the corneal origin (e.g., the intersection of the scleral / corneal interface plane and the central axis or cornea) does not change, the X and Z positions of the implant are calculated with respect to the implant, with respect to the reference IOP and the actual IOP. The change in distance between two implants can then be calculated by finding the difference in their X positions. The distance between two implants positioned symmetrically on either side of the corneal optical axis is calculated. The results of such calculations are plotted as a function of IOP with respect to a 45-degree implantation angle. The procedure, therefore, converts changes in the radius of curvature (which can be difficult to measure precisely using a simple frontal camera) into displacements of the imaging field, making it possible to precisely determine the relative positions of various implants on the cornea.

[0061] In some embodiments, the distance between two symmetrically positioned implants (symmetrical with respect to the optical axis of the cornea, i.e., the Z-axis) across the cornea is a function of the positioning angle 502. The change is shown in Figure 5 for an IOP change of approximately 40 mmHG. The distance change may depend on the implantation angle (i.e., the radius in which the fluorescent particles and / or markers are implanted), and can be maximized, for example, at approximately 50–60 degrees. In some cases, the distance change can be maximized over approximately 45–65 degrees. This angle may correspond to approximately 60–70 percent of the corneal radius, based on geometric calculations performed on a schematic eye model. As the angle increases, the displacement may decrease. For example, at an angle of approximately 85 degrees, the implant may be at approximately 90% of the total corneal radius, and the displacement may be bisected. By implanting four fluorescent particles and / or markers, i.e., one pair at ±90% radius and a second pair at ±60% radius, an imaging configuration can be achieved that may be unaffected by changes in camera distance and eye rotation (with respect to small deviations from a perfectly aligned eye). Alternatively, an imaging configuration that may be unaffected by changes in camera distance and eye rotation (with respect to small deviations from a perfectly aligned eye) can be achieved by implanting four fluorescent particles, i.e., one pair at ±30% radius and a second pair at ±60% radius (where they are closer to the apex of the cornea but sufficiently distal to the pupil so as not to obstruct vision). The measurements may be corrected for the human eye, taking into account ophthalmologically determined data measured at reference pressure, such as the radius of curvature and / or aspherical parameter Q. A normalized displacement can be calculated by determining the ratio of the distance between the outer implant (e.g., the implant closer to the sclera) and the inner implant (e.g., at approximately 60% of the corneal radius). The normalized displacement may not be sensitive to imaging magnification or slight eye rotation. In some embodiments, this ratio may be used to calculate the IOP using a reference initial measurement obtained at a known IOP. One or more cameras positioned to view the eye may be used to further refine this measurement. Measurements of corneal apex position and implant height for one or more cameras may be used to compensate for magnification errors and eye rotation.Off-angle views from one or more cameras may be used to correct the measured displacement values. Such off-angle views may be from the side, top, bottom, other angles, or any combination thereof, that are approximately 60 degrees off from the principal axis of the eye (the visual axis for a person looking directly straight ahead).

[0062] In some embodiments, an image of the eye 602 is present, referring to Figure 6. In some embodiments, a camera facing the front of the eye may use live images, captured images, or a combination thereof to determine the location of eye features (e.g., iris, pupil, cornea, or a combination thereof) as well as any particles and / or markers placed within the eye. In various embodiments as described herein, the particles and / or markers 403 may be placed in pockets at one or more depths within the cornea and / or in other surface structures of the eye, such as the sclera, the sclera / corneal interface, or a combination thereof. The particles and / or markers in these pockets are visible to a front-facing camera and may be seen in image 602. The stars in the image are representations of particles and / or markers placed within the surface of the eye. In some embodiments, the particles and / or markers may be placed along the X and Y axes, with their origin being a hypothetical point at the center of the pupil of the eye. Two particles and / or markers in the positive X direction and two particles and / or markers in the negative X direction may form a pair of inner and outer implants, as discussed elsewhere herein. Observed differences in position and / or calculated differences in some ratio between these points may be used to determine the IOP of the eye.

[0063] In some cases, the markers may be embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or in any combination thereof. The markers may be fluorescent, as described herein. The markers can be positioned at a maximum distance of about 10 mm from each other. The markers can be positioned at distances of about 0.5 mm, about 1 mm, about 1.5 mm, about 2 mm, about 2.5 mm, about 3 mm, about 3.5 mm, about 4 mm, about 4.5 mm, about 5 mm, about 5.5 mm, about 6 mm, about 6.5 mm, about 7 mm, about 7.5 mm, about 8 mm, about 8.5 mm, about 9 mm, about 9.5 mm, or about 10 mm from each other. The markers may be positioned at a distance of approximately 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, 3 mm, 3.5 mm, 4 mm, 4.5 mm, 5 mm, 5.5 mm, 6 mm, 6.5 mm, 7 mm, 7.5 mm, 8 mm, 8.5 mm, 9 mm, 9.5 mm, or 10 mm from each other. The markers may be positioned at a distance of at least approximately 0.5 mm, at least approximately 1 mm, at least approximately 1.5 mm, at least approximately 2 mm, at least approximately 2.5 mm, at least approximately 3 mm, at least approximately 3.5 mm, at least approximately 4 mm, at least approximately 4.5 mm, at least approximately 5 mm, at least approximately 5.5 mm, at least approximately 6 mm, at least approximately 6.5 mm, at least approximately 7 mm, at least approximately 7.5 mm, at least approximately 8 mm, at least approximately 8.5 mm, at least approximately 9 mm, at least approximately 9.5 mm, or at least approximately 10 mm from each other.

[0064] In some cases, there may be multiple pairs of markers. In some cases, there may be about 2, about 3, or about 4 pairs of markers. In some cases, there may be up to about 2, up to about 3, or up to about 4 pairs of markers. In some cases, there may be at least about 2, at least about 3, or at least about 4 markers. The IOP of the eye can be determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers. The IOP of the eye can be determined from the ratio between additional pairs of markers.

[0065] In some cases, the fluorescent beads, particles, and / or markers 403 may be annular, tubular, circular, elliptical, cylindrical, helical, hexagonal, triangular, square, rectangular, quadrilateral, or any other shape.

[0066] In some cases, the diameter of the fluorescent particles 403 can be approximately 10 microns to approximately 1,000 microns. The diameter of the fluorescent particles and / or markers can be approximately 10 to 100 microns, approximately 10 to 200 microns, approximately 10 to 300 microns, approximately 10 to 400 microns, approximately 10 to 500 microns, approximately 10 to 600 microns, approximately 10 to 700 microns, approximately 10 to 800 microns, approximately 10 to 900 microns, approximately 10 to 1,000 microns, approximately 100 to 200 microns, approximately 100 to 300 microns, approximately 100 to 4 00 microns, approximately 100 microns to approximately 500 microns, approximately 100 microns to approximately 600 microns, approximately 100 microns to approximately 700 microns, approximately 100 microns to approximately 800 microns, approximately 100 microns to approximately 900 microns, approximately 100 microns to approximately 1,000 microns, approximately 200 microns to approximately 300 microns, approximately 200 microns to approximately 400 microns, approximately 200 microns to approximately 500 microns, approximately 200 microns to approximately 600 microns, approximately 200 microns to approximately 700 microns, approximately 200 microns to approximately 800 microns, approximately 200 microns to approximately 900 Micron, approximately 200 microns to approximately 1,000 microns, approximately 300 microns to approximately 400 microns, approximately 300 microns to approximately 500 microns, approximately 300 microns to approximately 600 microns, approximately 300 microns to approximately 700 microns, approximately 300 microns to approximately 800 microns, approximately 300 microns to approximately 900 microns, approximately 300 microns to approximately 1,000 microns, approximately 400 microns to approximately 500 microns, approximately 400 microns to approximately 600 microns, approximately 400 microns to approximately 700 microns, approximately 400 microns to approximately 800 microns, approximately 400 microns to approximately 900 Micron, approximately 400 microns to approximately 1,000 microns, approximately 500 microns to approximately 600 microns, approximately 500 microns to approximately 700 microns, approximately 500 microns to approximately 800 microns, approximately 500 microns to approximately 900 microns, approximately 500 microns to approximately 1,000 microns, approximately 600 microns to approximately 700 microns, approximately 600 microns to approximately 800 microns, approximately 600 microns to approximately 900 microns, approximately 600 microns to approximately 1,000 microns, approximately 700 microns to approximately 800 microns, approximately 700 microns to approximately 900 microns, approximately 700 microns to approximately 1,It can be 000 microns, approximately 800 to 900 microns, approximately 800 to 1,000 microns, and approximately 900 to 1,000 microns.

[0067] In some cases, the diameter of the fluorescent particles and / or markers may be less than approximately 10 microns, less than approximately 100 microns, less than approximately 200 microns, less than approximately 300 microns, less than approximately 400 microns, less than approximately 500 microns, less than approximately 600 microns, less than approximately 700 microns, less than approximately 800 microns, less than approximately 900 microns, or less than approximately 1,000 microns. In some cases, the diameter of the fluorescent particles and / or markers may be greater than approximately 10 microns, greater than approximately 100 microns, greater than approximately 200 microns, greater than approximately 300 microns, greater than approximately 400 microns, greater than approximately 500 microns, greater than approximately 600 microns, greater than approximately 700 microns, greater than approximately 800 microns, greater than approximately 900 microns, or greater than approximately 1,000 microns.

[0068] Referring to Figure 7, exemplary eyewear is shown, for example, goggles 702 with a pair of cameras according to embodiments described herein. The front camera 701 and the side camera 703 may generally be aimed at the surface of the eye so that the cameras can view the eye and / or capture one or more images of the eye. In some embodiments, the front camera may generally be positioned on the visual axis of the eye and “looking down” the eye, capturing an image of the eye from a front oblique view of the eye. In some embodiments, there may be a side camera that can view the eye from an out-of-angle view. The out-of-angle view may be to the side as shown in Figure 7, or at some other angle from which an image can be captured or viewed. As shown in Figure 7, the arrows pointing from the box towards the eye represent the line of sight of the cameras.

[0069] The goggles may comprise a chassis and / or frame and one or more lenses. The chassis may include, at least partially, one or more detectors, light sources, or a combination thereof. The chassis may comprise eyewear, glasses, goggles, head-up displays, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

[0070] In some cases, the goggles can be full-coverage goggles 702, as shown in Figure 7. In some cases, the goggles can be thinner. In some cases, the thickness of the goggles can be approximately 10 mm to approximately 2 mm. In some cases, the thickness of the goggles can be approximately 2 mm to approximately 4 mm, approximately 2 mm to approximately 6 mm, approximately 2 mm to approximately 8 mm, approximately 2 mm to approximately 10 mm, approximately 4 mm to approximately 6 mm, approximately 4 mm to approximately 8 mm, approximately 4 mm to approximately 10 mm, approximately 6 mm to approximately 8 mm, approximately 6 mm to approximately 10 mm, or approximately 8 mm to approximately 10 mm. In some cases, the thickness of the goggles can be less than approximately 10 mm, less than approximately 8 mm, less than approximately 6 mm, less than approximately 4 mm, or less than approximately 2 mm.

[0071] In some cases, goggles may not have an upper, lower, or both portion that comes into contact with the subject's face. In some cases, goggles may include eyeglasses. In some cases, goggles may include laboratory eyeglasses. Goggles may include light-blocking eyeglasses (e.g., sunglasses). Goggles may have a tinting function similar to transition eyeglasses. Goggles may include indoor eyeglasses.

[0072] In some cases, the goggles may have a band 704 around the head, as shown in Figure 7. The goggles may have ear loops or temple tips for holding the weight of the goggles with the camera, for example, so that they can be stabilized when the camera is taking one or more images.

[0073] The goggles may be any type of eyewear goggles used for modern eyewear, including goggles for sunglasses, corrective glasses, safety glasses, and all types of goggles (e.g., for swimming, sports, safety, skiing, etc.). The goggles may be suitable for a single lens for one eye, a lens for two eyes (e.g., a visor), a single lens and eye cover (for people with amblyopia or who may suffer from the loss of one eye, etc.), or any combination thereof. The lenses may include one or more prescription lenses for visual correction, clear or tinted lenses for appearance, eye coverings, opaque lenses, or any combination thereof. In some embodiments, the lenses may have a defined area for the field of view of the user's eye. The field of view may be clear to avoid obstructing the user's vision. Various elements of the eyewear device may be mounted on the periphery of the lens, on the frame, or in combination thereof. The goggles may have flanges, other protrusions, tabs, or combinations thereof for mounting an image sensor, light source, battery, computing device, any other component suitable for use with the Disclosure, or any combination thereof.

[0074] The goggles can be positioned at a distance of up to approximately 2 centimeters (cm) from the tangential surface of the eye. The goggles can be positioned at a distance of at least approximately 0.5 cm, at least approximately 1 cm, at least approximately 1.5 cm, or at least approximately 2 cm. In some cases, the goggles can be positioned at a distance of less than approximately 0.5 cm, less than approximately 1 cm, less than approximately 1.5 cm, or less than approximately 2 cm.

[0075] Referring to Figure 8, a schematic diagram of the view of the side-view camera 802, as described elsewhere in this specification, is shown. Side-view images and / or out-of-angle images obtained, detected, and / or acquired by the side-view camera may show the position of one or more particles and / or markers 403 (e.g., implants as described elsewhere in this specification) and their relative angles off-center from the central axis of the eye. In some cases, the side-view images and / or out-of-angle images may show two or more particles 403. In some cases, the side-view images and / or out-of-angle images may show four particles 403, as shown in Figure 9. An artificial origin may be used for each particle to determine the angle from the principal axis of the eye, as described elsewhere in this specification. Images may show a stationary origin, which may be another implanted particle and / or marker, an anatomical feature, an artificial reference point (e.g., one that can be attached to goggles), or any combination thereof.

[0076] In some embodiments, the positions of the embedded fluorescent particles and / or markers may be displaced from the center 902 of the cornea, as shown in Figure 9. In some embodiments, the positions of the particles 403 may not be aligned with the artificial X or Y axis. The particles can be in any orientation or alignment, as long as their positions can be accurately measured, whether at actual distances and / or observed distances.

[0077] In some cases, the markers are embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or in any combination thereof. The markers may be fluorescent, as described herein. The markers can be positioned at a maximum distance of about 10 mm from each other. The markers can be positioned at distances of about 0.5 mm, about 1 mm, about 1.5 mm, about 2 mm, about 2.5 mm, about 3 mm, about 3.5 mm, about 4 mm, about 4.5 mm, about 5 mm, about 5.5 mm, about 6 mm, about 6.5 mm, about 7 mm, about 7.5 mm, about 8 mm, about 8.5 mm, about 9 mm, about 9.5 mm, or about 10 mm from each other. The markers may be positioned at a distance of approximately 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, 3 mm, 3.5 mm, 4 mm, 4.5 mm, 5 mm, 5.5 mm, 6 mm, 6.5 mm, 7 mm, 7.5 mm, 8 mm, 8.5 mm, 9 mm, 9.5 mm, or 10 mm from each other. The markers may be positioned at a distance of at least approximately 0.5 mm, at least approximately 1 mm, at least approximately 1.5 mm, at least approximately 2 mm, at least approximately 2.5 mm, at least approximately 3 mm, at least approximately 3.5 mm, at least approximately 4 mm, at least approximately 4.5 mm, at least approximately 5 mm, at least approximately 5.5 mm, at least approximately 6 mm, at least approximately 6.5 mm, at least approximately 7 mm, at least approximately 7.5 mm, at least approximately 8 mm, at least approximately 8.5 mm, at least approximately 9 mm, at least approximately 9.5 mm, or at least approximately 10 mm from each other.

[0078] In some cases, there are multiple pairs of markers. In some cases, there are approximately 2, 3, or 4 pairs of markers. In some cases, there are up to approximately 2, 3, or 4 pairs of markers. In some cases, there are at least approximately 2, 3, or 4 markers. The IOP of the eye can be determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers. The IOP of the eye can be determined from the ratio between additional pairs of markers.

[0079] In some embodiments, an artificial eye model fabricated from an elastomer material can be used to measure the distance between embedded fluorescent microparticles and / or markers as a function of the applied IOP. The data shown in Figure 10 shows a correlation between the microparticle displacement and the applied IOP. The particle positions may be measured on one or more captured images acquired and / or detected by a camera facing the front of the model eye, a side-view camera, or a combination thereof, and the central peak of the spot corresponding to the fluorescent beads and / or markers may be calculated using centroid calculation or peak fitting to the Gaussian intensity profile of the spot. Higher IOP values ​​may result in increased distances between the fluorescent beads and / or markers. The fluorescent implant may be assumed to have a narrowband excitation spectrum (e.g., green light may be used for excitation in a non-limiting embodiment) and a narrowband emission spectrum (e.g., red light may be emitted in a non-limiting embodiment). In practice, a fluorescence imaging camera may be used in which a fluorescence excitation light source (e.g., a green LED) and a normal illumination source (e.g., a red LED) sequentially illuminate one or more particles and / or one or more markers, so that one or more images may be collected, detected, and / or acquired for red and / or green illumination. One or more images captured after illumination with the red illuminator may provide a monochrome regular image of the eye, enabling a neural network to find and / or identify corneal positions, for example, based on pattern matching. The green illuminator may be blocked by an emission filter in front of the camera, so that only dye-doped fluorescent particles and / or markers are imaged as bright spots when the green LED is on, thus removing the background from the iris and / or the rest of the eye. When the eye is illuminated by the red illuminator, a monochrome regular image collected at an earlier point in time (e.g., 30 milliseconds before the image of one or more particles and / or markers illuminated by the green illuminator) may be used as a guide to estimate the neighborhood of the positions of the fluorescent beads and / or markers.In some cases, an area based on a monochrome regular image may be used to define a window for curve fitting or centroid calculation to precisely determine the peak positions of fluorescent beads and / or marker spots. In this way, undesirable residual fluorescence from the iris and other ocular tissue noise sources can be excluded, and the precise positions of particles and / or markers can be obtained.

[0080] In some cases, the undesirable background noise of residual fluorescence can be reduced by approximately 5% to approximately 95%. In other cases, the undesirable background noise of residual fluorescence can be reduced by at least approximately 5%, at least approximately 15%, at least approximately 25%, at least approximately 35%, at least approximately 45%, at least approximately 55%, at least approximately 65%, at least approximately 75%, at least approximately 85%, or at least approximately 95%. In other cases, the undesirable background noise of residual fluorescence can be reduced by less than approximately 5%, less than approximately 15%, less than approximately 25%, less than approximately 35%, less than approximately 45%, less than approximately 55%, less than approximately 65%, less than approximately 75%, less than approximately 85%, or less than approximately 95%.

[0081] In some embodiments, a schematic diagram 1102 showing a sample cross-section of the imaging system is shown in Figure 11. The system and the devices described therein can be used in methods described elsewhere herein to measure the IOP of the eye.

[0082] Using one or more cameras, a user can acquire and / or receive a first image. The image may be of at least two particles and / or markers implanted in the eye. The image can be detected using a first detector. The first detector may be a camera, a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or any combination thereof. The detector may be configured to detect signals from at least two particles and / or markers implanted in the eye. The image may be captured by a camera placed on goggles, glasses, other eyewear, or any combination thereof. Using one or more cameras, a user can acquire or receive a second image. The second image may be of at least two particles and / or markers implanted in the eye. The image can be detected using a second detector. The second detector may be a camera, a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or any combination thereof. The second detector may be a different camera from the camera of the first detector. One or more detectors may be incorporated, at least partially, within the goggles, as described elsewhere herein. The optical axis of the first detector may be at an angle to the optical axis of the second detector. A user can determine the IOP of a target using the embedded particles and / or markers from the positions of one or more of at least two particles and / or markers from the first image, or one or more of at least two markers from the second image. In some cases, the positions of at least two markers and / or particles may be provided as input to a machine learning algorithm or predictive model that is trained to provide an output related to the IOP of the eye.

[0083] In some cases, the markers are embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or in any combination thereof. The markers may be fluorescent, as described herein. The markers can be positioned at a maximum distance of about 10 mm from each other. The markers can be positioned at distances of about 0.5 mm, about 1 mm, about 1.5 mm, about 2 mm, about 2.5 mm, about 3 mm, about 3.5 mm, about 4 mm, about 4.5 mm, about 5 mm, about 5.5 mm, about 6 mm, about 6.5 mm, about 7 mm, about 7.5 mm, about 8 mm, about 8.5 mm, about 9 mm, about 9.5 mm, or about 10 mm from each other. The markers may be positioned at a distance of approximately 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, 3 mm, 3.5 mm, 4 mm, 4.5 mm, 5 mm, 5.5 mm, 6 mm, 6.5 mm, 7 mm, 7.5 mm, 8 mm, 8.5 mm, 9 mm, 9.5 mm, or 10 mm from each other. The markers may be positioned at a distance of at least approximately 0.5 mm, at least approximately 1 mm, at least approximately 1.5 mm, at least approximately 2 mm, at least approximately 2.5 mm, at least approximately 3 mm, at least approximately 3.5 mm, at least approximately 4 mm, at least approximately 4.5 mm, at least approximately 5 mm, at least approximately 5.5 mm, at least approximately 6 mm, at least approximately 6.5 mm, at least approximately 7 mm, at least approximately 7.5 mm, at least approximately 8 mm, at least approximately 8.5 mm, at least approximately 9 mm, at least approximately 9.5 mm, or at least approximately 10 mm from each other.

[0084] In some cases, there may be multiple pairs of markers. In some cases, there may be about 2, about 3, or about 4 pairs of markers. In some cases, there may be up to about 2, up to about 3, or up to about 4 pairs of markers. In some cases, there may be at least about 2, at least about 3, or at least about 4 markers. The IOP of the eye can be determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers. The IOP of the eye can be determined from the ratio between additional pairs of markers.

[0085] In some cases, a light source is provided to a marker implanted in the eye to acquire and / or detect a first or second image, as described elsewhere herein. One or more light sources may be present. The light sources may include light-emitting diodes (LEDs), lasers, broadband superluminescent diodes, coherent light sources, or any combination thereof. The light sources may include LEDs. Light-emitting diodes can emit light in various bands of wavelengths of the visual spectrum, for example, from green to red wavelengths. The light sources may be positioned on goggles as described above. The light sources may be optically coupled to one or more eyes of the subject.

[0086] In some cases, one or more polarizing filters may be used to acquire one or more images using a first detector and / or a second detector. A first image of at least two particles and / or markers implanted in the eye can be acquired or detected by the first detector through the first polarizing filter. A second image of at least two particles and / or markers implanted in the eye can be acquired or detected by the second detector through the second polarizing filter.

[0087] In some cases, one or more processors may be electrically coupled to one or more detectors. The processor may be configured to process signals from two or more particles and / or markers implanted in the eye and to determine the eye's IOP from the positions of the two or more particles and / or markers. The processor may include memory. The memory may be separate from the processor. In some cases, the processor may be outside the system shown in Figure 11. The processor may be located on an eyewear device, a teleprocessing device, a remote server, a cloud server, or any combination thereof. The processor may acquire or receive one or more images from the server, cloud-based storage device, eyewear device, or any combination thereof.

[0088] The memory can store one or more programs for execution by one or more processors. One or more programs may include instructions for acquiring and / or receiving results from an imaging system, as described elsewhere herein. One or more programs may include instructions for acquiring and / or receiving one or more images of at least two particles and / or markers implanted in the eye. One or more images may be detected using a first detector and / or a second detector. The optical axis of the first detector may be at an angle to the optical axis of the second detector, as described elsewhere herein. The program may include instructions for determining the IOP of the eye from the distance between at least two particles and / or markers in one or more images.

[0089] Specifically referring to an exemplary system as shown in Figure 11, a front-facing camera 1106 (e.g., a first detector as described elsewhere herein) may be positioned behind a long-pass optical filter 1105 with a positioned yellow (e.g., a cutoff wavelength of about 580 nm) to obtain and / or acquire one or more planar images of the eye with the embedded particles 403. A side-facing camera 1101 (e.g., a second detector as described elsewhere herein) may also be positioned behind a long-pass optical filter 1104 with a yellow (e.g., a cutoff wavelength of about 580 nm). One or more fluorescence excitation light sources 1103 (e.g., green LEDs, i.e., with emission wavelengths of about 520 nm to 540 nm) may provide excitation light to one or more microparticles and / or marker embeddings. A regular imaging illuminator 1107 (e.g., a red LED, i.e., emitting at a central wavelength of approximately 650 nm) may be turned off when the fluorescence excitation light source 1103 is turned on and / or emitting light. After obtaining a fluorescence image, the green LED 1103 may be turned off, and one or more red LEDs 1107 may be turned on to capture, obtain, detect, and / or acquire a regular image of the eye that identifies one or more ocular anatomical features (e.g., iris, sclera, sclerocorneal junction, or any combination thereof). Bright-field images taken with red illumination may be used to position the pupil, iris, cornea, or any combination thereof, which may provide additional information about the positioning of the cornea relative to the first detector, the second detector, the regular imaging illuminator, the fluorescence imaging illuminator, or any combination thereof. The fluorescence image may then be used to extract the locations of microparticles, particles, and / or markers embedded in the eye by fitting a Gaussian function to the particle and / or marker image. Precise determination of the peak positions of particle and / or marker blob representations corresponding to the microparticles and / or markers can be achieved using Gaussian fitting. The sub-pixel accuracy provided by Gaussian fitting can enable precise determination of the distances between particles and / or markers.

[0090] The front camera 1106 and / or the side-view camera 1101 may be directed and / or aligned with the surface of the eye so that the front camera 1106 and / or the side-view camera 1101 can view the eye and / or capture images of particles and / or markers embedded and / or placed within the eye. In some embodiments, the front camera 1106 may be positioned on the visual axis and, for example, “looking down” on the eye, viewing the eye from directly in front of it. In some embodiments, a side-view camera 1101 may be present that can view the eye from an out-of-angle. The out-of-angle view may be to the side, as shown in Figure 11, or at some other angle from which one or more images can be captured, acquired, and / or viewed. The arrows pointing from the box toward the eye represent the camera’s line of sight, as shown in Figure 11.

[0091] In some embodiments, image data 1202 may be used to train a deep neural network 1210 in the process illustrated in Figure 12.

[0092] In some cases, one or more processors may be electrically coupled to one or more detectors (e.g., first and / or second detectors as described elsewhere herein). In some cases, one or more processors may be electrically coupled to a system as described herein. The processors may be configured to process signals from two or more particles and / or markers implanted in the eye and to determine the IOP of the eye from the positions of the two or more particles and / or markers. The processors may be located on an eyewear device, a teleprocessing device, a remote server, a cloud server, or any combination thereof. The processors may acquire and / or receive one or more images from the server, cloud-based storage device, eyewear device, or any combination thereof. The processors may include memory. The memory may be separate from the processors.

[0093] The memory can store one or more programs for execution by one or more processors. One or more programs may include instructions for acquiring and / or receiving results from the imaging system as described above. One or more programs may include instructions for acquiring and / or receiving one or more images of at least two particles and / or markers implanted in the eye. One or more images may be detected using a first detector and / or a second detector as described elsewhere herein. The optical axis of the first detector may be at an angle to the optical axis of the second detector as described elsewhere herein. The program may include instructions for determining the IOP of the eye from the distance between at least two markers in one or more images.

[0094] Referring to Figure 12, the positions of at least two markers, including image data 1202, can be provided as input to a machine learning algorithm or predictive model that is trained to provide an output related to the IOP of an eye. The image data 1202 may be one or more still images and / or one or more video frames, segments, and / or clips from one or more cameras and / or any suitable source such as a data storage medium (e.g., memory). The machine learning algorithm and / or predictive model can receive and / or acquire one or more images of at least two particles and / or markers embedded in one or more eyes, and the corresponding IOPs of one or more eyes, through the system described herein. An untrained and / or partially untrained machine learning algorithm and / or predictive model can be trained using one or more images of at least two particles and / or markers embedded in one or more eyes and the corresponding IOPs of one or more eyes to produce a trained machine learning algorithm and / or trained predictive model. Untrained or partially untrained machine learning algorithms and / or predictive models can also be trained on the distance between at least two particles and / or markers in one or more images of at least two particles and / or markers implanted in one or more eyes.

[0095] The training process may include step 1204 of extracting one or more frames from image data 1202 in which various regions can be identified and / or labeled to produce one or more focus-marked regions 1206. The focus-marked regions 1206 may be tens, hundreds, or thousands in number and may be drawn from a similar number of image data 1202 sources. The focus-marked regions 1206 of the image data 1202 may be considered training data for training a deep neural network 1208. Once the deep neural network 1208 is considered sufficiently trained, based on an acceptable rate of error of new data inputs compared to analysis by the trained individual, the deep neural network 1210 may be trained with target data not used to train machine learning algorithms and / or predictive models and may be ready for diagnostic use.

[0096] The acceptable error rate may be less than approximately 10%. The acceptable error rate may be less than approximately 10%, less than approximately 8%, less than approximately 6%, less than approximately 4%, less than approximately 2%, less than approximately 1%, less than approximately 0.5%, or less than approximately 0.1%. The acceptable error rate for the calculated positions of one or more particles and / or one or more markers and / or the distance between one or more particles and / or one or more markers is about 0.1mm to about 0.5mm, about 0.1mm to about 1mm, about 0.1mm to about 2mm, about 0.1mm to about 4mm, about 0.1mm to about 6mm, about 0.1mm to about 8mm, about 0.1mm to about 10mm, about 0.5mm to about 1mm, about 0.5mm to about 2mm, about 0.5mm to about It may be 4mm, approximately 0.5mm to 6mm, approximately 0.5mm to 8mm, approximately 0.5mm to 10mm, approximately 1mm to 2mm, approximately 1mm to 4mm, approximately 1mm to 6mm, approximately 1mm to 8mm, approximately 1mm to 10mm, approximately 2mm to 4mm, approximately 2mm to 6mm, approximately 2mm to 8mm, approximately 2mm to 10mm, approximately 4mm to 6mm, approximately 4mm to 8mm, approximately 4mm to 10mm, approximately 6mm to 8mm, approximately 6mm to 10mm, and approximately 8mm to 10mm.

[0097] In some embodiments, the operation of using a deep neural network in combination with image data from patients is illustrated in Figure 13.

[0098] In some cases, one or more processors may be electrically coupled to one or more detectors. In some cases, one or more processors may be electrically coupled to a system as described herein. A processor may be configured to process signals from two or more particles and / or markers implanted in the eye and to determine the eye's IOP from the positions of the two or more markers. A processor may be located on an eyewear device, a teleprocessing device, a remote server, a cloud server, or any combination thereof. A processor may acquire and / or receive one or more images from a server, cloud-based storage device, eyewear device, or any combination thereof. A processor may include memory. The memory may be separate from the processor.

[0099] The memory can store one or more programs for execution by one or more processors. One or more programs may include instructions for acquiring and / or receiving results from the imaging system as described above. One or more programs may include instructions for acquiring and / or receiving one or more images of at least two particles and / or markers implanted in the eye. One or more images may be detected using a first detector and / or a second detector. The optical axis of the first detector may be at an angle to the optical axis of the second detector, as described elsewhere herein. The program may include instructions for determining the IOP of the eye from the distance between at least two particles and / or markers in one or more images.

[0100] Referring to Figure 13, the image data 1302 may be drawn from one or more cameras, memory devices, intermediate data sources such as the cloud, data bus, processor, camera memory, computing device memory, or any combination thereof. The received image data 1302 may be input into a trained deep neural network 1304, and the trained deep neural network machine learning algorithm and / or trained predictive model may produce an annotated eye image dataset 1306, which can be displayed on a screen (mobile phone, tablet, computer screen, or any combination thereof) and / or stored in computing device memory for later retrieval. The deep neural network 1304 may be trained as described elsewhere in this specification or by additional methods.

[0101] By using the provided process, objects whose IOP can be monitored can have rapid and reliable results using the systems, devices, and methods described herein, in combination with equipment for measuring IOP from the installation location of fluorescent beads, particles, and / or markers. Examples of electronic components

[0102] Embodiments of subject matter and operation described herein may be implemented in digital electronic networks or in computer software, firmware, and / or hardware, including structures disclosed herein and their structural equivalents, or any combination thereof. Embodiments of subject matter described herein may be implemented as one or more modules of computer program instructions encoded on one or more computer storage media for execution by one or more computer programs, for example, by data processing devices such as processing circuits, or for controlling their operation. Processing circuits such as controllers and / or CPUs may comprise any digital and / or analog circuit components configured to perform the functions described herein, such as microprocessors, microcontrollers, application-specific integrated circuits, programmable logic, or any combination thereof. In some cases, program instructions may be encoded on mechanically generated electrical, optical, or electromagnetic signals, which may be generated to encode information for transmission to a receiver device suitable for execution by a data processing device.

[0103] A computer storage medium may be a computer-readable storage device, a computer-readable storage board, a random or serial access memory array or device, or any combination thereof, or contained within one. Furthermore, a computer storage medium may not be a propagating signal, but may be the source or destination of computer program instructions encoded within artificially generated propagating signals. A computer storage medium may also be one or more distinct components or media (e.g., multiple CDs, disks, other storage devices, or any combination thereof), or contained within one. Therefore, a computer storage medium can be both tangible and non-transient.

[0104] The operations described herein may be implemented as operations performed by a data processing apparatus with respect to data stored on and / or received from one or more computer-readable storage devices. The terms “data processing apparatus” or “computing device” encompass, in embodiments, any kind of apparatus, device, and / or machine for processing data, including, but not limited to, a programmable processor, a computer, a system on a chip, or more, or a combination thereof, may also include special-purpose logic networks, such as FPGAs (field-programmable gate arrays) and / or ASICs (application-specific integrated circuits). In addition to hardware, the apparatus may also include code, such as processor firmware, a protocol stack, a database management system, an operating system, a cross-platform runtime environment, a virtual machine, or any combination thereof, that generates an execution environment for the computer program. The apparatus and / or execution environment can realize various different computing model infrastructures, such as web services, distributed computing, and grid computing infrastructures.

[0105] A computer program (i.e., a program, software, software application, script, code, or any combination thereof) may be written in any form of programming language, including compiled or interpreted languages, declarative or procedural languages, and may be deployed in any form, including as a standalone program or as modules, components, subroutines, objects, or other units suitable for use in a computing environment, or any combination thereof. A computer program may correspond to a file in a file system, although this is not required. A program may store parts of files that hold other programs or data (e.g., one or more scripts stored within a markup language document) in a single file dedicated to that program, or in multiple collaborative files (e.g., files storing one or more modules, subprograms, or parts of code). A computer program may be deployed to run on one computer, or on multiple computers located on one site, or distributed across multiple sites and interconnected by a communication network.

[0106] The processes, operations, and / or logical flows described herein may be implemented by one or more programmable processors that execute one or more computer programs to perform actions by operating with respect to input data and generating outputs. The processes, operations, and / or logical flows may also be implemented by special-purpose logic networks, such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the devices may also be implemented as such.

[0107] Processors suitable for executing computer programs may, in examples, include both general-purpose microprocessors and specialized microprocessors, and any one or more processors of any type of digital computer. Generally, a processor can receive instructions and data from read-only memory and / or random-access memory or both. A computer comprises a processor for performing actions according to instructions and one or more memory devices for storing instructions and data. Generally, a computer may also include, or be operably coupled to, one or more mass storage devices for storing data, such as magnetic, magneto-optical disks, optical disks, or any combination thereof, for receiving data from, transferring data to, or both. A computer may also be embedded in another device, such as a mobile phone, personal digital assistant (PDA), mobile audio or video player, game console, Global Positioning System (GPS) receiver, portable storage device (e.g., Universal Serial Bus (USB) flash drive), or any combination thereof. Devices suitable for storing computer program instructions and / or data may, in embodiment, include any form of non-volatile memory, medium, and memory device, including semiconductor memory devices such as EPROM, EEPROM, and flash memory devices, magnetic disks such as internal hard disks or removable disks, magneto-optical disks, CD-ROMs and DVD-ROM disks, or any combination thereof. The processor and memory may be complemented by or incorporated within special-purpose logic networks.

[0108] To provide user interaction, embodiments of the subject matter described herein may be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), an LCD (liquid crystal display) monitor, an OLED (organic light-emitting diode) monitor, other forms of displays for displaying information to a user, or any combination thereof, and a keyboard and / or pointing device, e.g., a mouse or trackball, by which the user can provide input to the computer. Other types of devices may also be used to provide user interaction, for example, the feedback provided to the user may be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback, and input from the user may be received in any form, including acoustic, verbal, tactile input, or any combination thereof. In addition, the computer may interact with the user by sending documents to and / or receiving documents from devices used by the user, for example, by sending a web page to a web browser on the user's client device in response to a request received from a web browser.

[0109] While one embodiment of the method and system has been described, it will be apparent to those skilled in the art that other embodiments incorporating the concept may also be used. It should be understood that the systems described above may provide any or more of their components, and these components may be provided on a standalone machine or, in some embodiments, on multiple machines within a distributed system. Systems, devices, and / or methods described elsewhere in this specification may be implemented as methods, apparatus, and / or articles using programming and / or engineering design techniques to produce software, firmware, hardware, or any combination thereof. In addition, systems and methods described elsewhere in this specification may be provided as one or more computer-readable programs embodied on and / or within one or more articles. As used herein, the term “article of manufacture” is intended to encompass code and / or logic accessible from and embedded within one or more computer-readable devices, firmware, programmable logic, memory devices (e.g., EEPROM, ROM, PROM, RAM, SRAM, or any combination thereof), hardware (e.g., integrated circuit chips, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), or any combination thereof), electronic devices, computer-readable non-volatile storage units (e.g., CD-ROMs, floppy disks, hard disk drives, or any combination thereof), or any combination thereof. An article of manufacture may be accessible from a file server that provides access to a computer-readable program via a network transmission line, a wireless transmission medium, a signal propagating through space, radio waves, infrared signals, or any combination thereof. An article of manufacture may be a flash memory card and / or magnetic tape. An article of manufacture may include hardware logic and software and / or programmable code embedded within a computer-readable medium, executed by a processor.Generally, computer-readable programs may be implemented in any programming language such as LISP, PERL, C, C++, C#, PROLOG, any bytecode language such as JAVA®, or any combination thereof. Software programs may be stored as object code on or within one or more products. Examples of machine learning methodologies

[0110] As used in this disclosure and the appended claims, the terms “artificial intelligence,” “artificial intelligence technique,” ​​“artificial intelligence operation,” and “artificial intelligence algorithm” generally refer to any system and / or computational procedure that can take one or more actions to simulate human intelligence processes in order to improve or maximize the likelihood of achieving a goal. The term “artificial intelligence” may include “generative modeling,” “machine learning” (ML), or “reinforcement learning” (RL).

[0111] As used in this disclosure and the appended claims, the terms “machine learning,” “machine learning technique,” ​​“machine learning operation,” and “machine learning model” generally refer to any system or analytical or statistical procedure that can progressively improve the computer performance of a task. In some cases, ML may generally involve steps of identifying and recognizing patterns in existing data to facilitate making predictions about subsequent data. ML may include ML models (which may include, for example, ML algorithms). Machine learning can provide deductive or hypothetical inferences based on real or simulated data, whether they are inherently analytical and / or statistical. ML models may be trained models. ML techniques may include one or more supervised, semi-supervised, self-supervised, or unsupervised ML techniques. For example, an ML model may be a trained model that is trained through supervised learning (for example, various parameters are determined as weights or scaling factors). ML may include one or more of the following: regression analysis, regularization, classification, dimensionality reduction, ensemble learning, meta-learning, association rule learning, cluster analysis, anomaly detection, deep learning, ultra-deep learning, or any combination thereof.ML is not limited to k-means clustering, k-nearest neighbors, learned vector quantization, linear regression, nonlinear regression, least squares regression, partial least squares regression, logistic regression, stepwise regression, multivariate adaptive regression splines, ridge regression, principal component regression, least absolute contraction and selection operation (LASSO), least angular regression, canonical correlation analysis, factor analysis, independent component analysis, linear discriminant analysis, multidimensional scaling, non-negative matrix factorization, principal component analysis, principal coordinate analysis, projection tracking, Sammon mapping, t-distribution stochastic nearest neighbor embedding, AdaBoosting, boosting, gradient boosting, bootstrap aggregation, ensemble averaging, decision trees, conditional decision trees, boosted decision trees, gradient boosted decision trees, and random This may include forests, stacked generalizations, Bayesian networks, Bayesian belief networks, naive Bayes, Gaussian naive Bayes, polynomial naive Bayes, hidden Markov models, hierarchical hidden Markov models, support vector machines, encoders, decoders, autoencoders, stacked autoencoders, perceptrons, multilayer perceptrons, artificial neural networks, feedforward neural networks, convolutional neural networks, recurrent neural networks, long- and short-term memory, deep belief networks, deep Boltzmann machines, deep convolutional neural networks, deep recurrent neural networks, generative adversarial networks, or any combination thereof.

[0112] The methods and / or systems of this disclosure can process and / or analyze one or more corneal images, for example, to determine the IOP of an eye to monitor for glaucoma, as described elsewhere in this specification. In some cases, the processing and / or analysis of one or more corneal images may be performed using one or more machine learning algorithms and / or one or more predictive models with instructions provided using one or more processors, as described elsewhere in this specification. For example, one or more machine learning algorithms and / or predictive models may process one or more, two or more features of the corneal images, as described elsewhere in this specification.

[0113] In some cases, the target IOP can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with a sensitivity of at least approximately 70%, at least approximately 75%, at least approximately 80%, at least approximately 85%, or at least approximately 90%.

[0114] In some cases, IOP can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with a sensitivity of up to approximately 70%, 75%, 80%, 85%, or 90%.

[0115] In some cases, IOPs can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with specificities of at least approximately 70%, at least approximately 75%, at least approximately 80%, at least approximately 85%, or at least approximately 90%.

[0116] In some cases, IOPs can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with specificities of up to approximately 70%, 75%, 80%, 85%, or 90%.

[0117] In some cases, IOPs may be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with positive predictive values ​​of at least approximately 70%, at least approximately 75%, at least approximately 80%, at least approximately 85%, or at least approximately 90%.

[0118] In some cases, IOPs can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with positive predictive values ​​of up to approximately 70%, 75%, 80%, 85%, or 90%.

[0119] In some cases, IOP may be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with negative predictive values ​​of at least approximately 70%, at least approximately 75%, at least approximately 80%, at least approximately 85%, or at least approximately 90%.

[0120] In some cases, IOP can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with negative predictive values ​​of up to approximately 70%, 75%, 80%, 85%, or 90%.

[0121] In some cases, IOP can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with an area under the receiver operating characteristic curve (AUROC) of at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.82, at least about 0.84, at least about 0.86, at least about 0.88, or at least about 0.90.

[0122] In some cases, IOP can be determined and / or predicted using one or more machine learning algorithms and / or one or more predictive models with area under receiver operating characteristic curves of up to approximately 0.65, 0.70, 0.75, 0.80, 0.82, 0.84, 0.86, 0.88, or 0.90.

[0123] Algorithms and / or predictive models can be implemented using software, depending on their execution by one or more central processing units. In some cases, the predictive model may include a machine learning predictive model. In other cases, the machine learning predictive model may include one or more statistical, machine learning, artificial intelligence algorithms, or any combination thereof. Examples of algorithms, machine learning algorithms, and / or predictive models used may include support vector machines (SVMs), naive Bayesian classification, random forests, neural networks (such as deep neural networks (DNNs)), recurrent neural networks (RNNs), deep RNNs, long-shorter-time memory (LSTM) recurrent neural networks (RNNs), decision tree algorithms, unsupervised clustering algorithms, supervised clustering algorithms, unsupervised clustering algorithms, regression algorithms, gradient boosting algorithms (e.g., gradient boosted decision trees, gradient boosting implementations of machine learning algorithms and / or predictive models), gated regressive units (GRUs), supervised learning algorithms, unsupervised learning algorithms, statistical, deep learning algorithms for classification and / or regression, or any combination thereof. In some cases, a recurrent neural network may include units that are LSTM units or GRUs. In some cases, a predictive model and / or machine learning algorithm may include an ensemble of one or more predictive models and / or machine learning algorithms.

[0124] The machine learning prediction model may also involve estimation of an ensemble model consisting of multiple machine learning algorithms and / or prediction models, and may utilize techniques such as gradient boosting in the construction of a gradient boosting decision tree. The machine learning prediction model may be trained using one or more training datasets corresponding to a model cornea. In some embodiments, one or more training datasets may include distances in millimeters measured between embedded fluorescent particles and / or markers and the corresponding IOPs of the eye.

[0125] Training records may be constructed from a sequence of observations. Such sequences may have a fixed length to facilitate data processing. For example, sequences may be zero-padded or selected as independent subsets of records of a single subject.

[0126] One or more predictive models and / or one or more machine learning algorithms may process one or more input features and produce one or more output values, including an IOP of an eye. For example, such an IOP may include a binary classification of healthy / normal health status (e.g., absence of disease or disorder) or adverse health status (e.g., presence of disease or disorder), a group classification of category labels (e.g., "no disease or disorder", "apparent disease or disorder", and "likely disease or disorder"), the likelihood of developing a particular disease or disorder (e.g., relative likelihood or probability), a score indicating the presence of disease or disorder, a score indicating the level of systemic inflammation suffered by the patient, a "risk factor" regarding the likelihood of the patient's mortality, a prediction of the time at which the patient is expected to develop disease or disorder, a confidence interval for any numerical prediction, or any combination thereof. Various predictive models and / or machine learning algorithms may be cascaded such that the output of one or more predictive models and / or one or more machine learning algorithms can be used as one or more input features to subsequent layers or subsections of one or more predictive models and / or one or more machine learning algorithms.

[0127] To train one or more predictive models and / or machine learning algorithms to generate real-time classification and / or predictions (for example, by determining the weights and correlations of the predictive models and / or machine learning algorithms), the models may be trained using datasets described elsewhere herein (e.g., training datasets). Such datasets may be large enough to generate statistically significant classification and / or predictions. For example, the dataset may include a database of anonymized data, including one or more distances between one or more particles and / or markers and associated IOPs of eyes with one or more embedded particles and / or markers.

[0128] A dataset, as described elsewhere in this specification, may be divided into subsets (e.g., discrete or overlapping) such as a training dataset, a development dataset, and a test dataset. For example, a dataset may be divided into a training dataset comprising 80% of the dataset, a development dataset comprising 10% of the dataset, and a test dataset comprising 10% of the dataset. The training dataset may comprise about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90% of the dataset. The development dataset may comprise about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90% of the dataset. The test dataset may comprise about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90% of the dataset. The training set (e.g., the training dataset) may be selected by random sampling of the data set corresponding to one or more target cohorts to ensure sampling independence. In some cases, the training set (e.g., the training dataset) may be selected by proportional sampling of the data set corresponding to one or more target cohorts to ensure sampling independence.

[0129] To improve the accuracy of predictions by predictive models and / or machine learning algorithms and to reduce overfitting of predictive models and / or machine learning algorithms, datasets may be augmented to increase the number of samples in the training set. For example, data augmentation may include the step of rearranging the order of observations in the training records. To adapt datasets with missing observations, methods for filling in missing data may be used, such as forward filling, backward filling, linear interpolation, multitask Gaussian processes, or any combination thereof. Datasets may be filtered to remove confounding factors. For example, a subset of the target dataset may be excluded from the database.

[0130] Neural network techniques such as dropout or regularization may be used while training one or more predictive models and / or one or more machine learning algorithms to prevent overfitting. The neural network may comprise multiple subnetworks, each configured to generate classification and / or prediction of different types of output information (for example, they may be combined to form the overall output of the neural network). The one or more predictive models and / or one or more machine learning algorithms may, alternatively, utilize statistical or related algorithms, including random forests, classification and regression trees, support vector machines, discriminant analysis, regression techniques, ensembles and gradient-boosted variance, or any combination thereof.

[0131] When one or more predictive models and / or one or more machine learning algorithms generate an IOP classification or prediction, a notification (e.g., an alert or alarm) may be transmitted to a healthcare provider, such as a physician, nurse, managing healthcare worker, or any combination thereof, treating the subject within a hospital. The notification may be transmitted via automated phone calls, short message service (SMS), multimedia messaging service (MMS) messages, email, alerts in a dashboard, or any combination thereof. The notification may include output information such as the IOP prediction.

[0132] Different performance metrics may arise to demonstrate the performance of one or more predictive models and / or one or more machine learning algorithms. For example, the area under the receiver operating characteristic curve (AUROC) may be used to determine the diagnostic and / or classification capabilities of one or more predictive models and / or one or more machine learning algorithms. For example, one or more predictive models and / or one or more machine learning algorithms may use adjustable classification thresholds such that specificity and sensitivity are adjustable, and the receiver operating characteristic curve (ROC) may be used to identify different operating points corresponding to different values ​​of specificity and sensitivity for one or more predictive models and / or one or more machine learning algorithms.

[0133] In cases where the dataset is not sufficiently large, cross-validation may be performed to assess the robustness of one or more predictive models and / or one or more machine learning algorithms across different training and test datasets.

[0134] The following definitions may be used to calculate performance metrics such as sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), AUPRC, AUROC, any combination thereof, or analogues: "false positive" may refer to a positive outcome or an outcome where the result is inaccurate or premature. "true positive" may refer to a positive outcome or an outcome where the result occurs accurately. "false negative" may refer to a negative outcome or an outcome where the result occurs. "true negative" may refer to a negative outcome or an outcome where the result occurs.

[0135] One or more predictive models and / or one or more machine learning algorithms may be trained until certain conditions regarding accuracy and / or performance are met, such as having a minimum desired value corresponding to a classification and / or diagnostic accuracy measure. Examples of diagnostic accuracy measures may include susceptibility, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, area under the precision recall curve (AUPRC), and area under the receiver operating characteristic (ROC) curve (AUROC), corresponding to the diagnostic accuracy of detecting or predicting IOPs.

[0136] For example, such a predetermined condition may be that the predicted susceptibility to IOP includes values ​​such as at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0137] In another embodiment, such predetermined conditions may include, for example, a specificity predicting IOP that includes values ​​of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0138] In another embodiment, such predetermined conditions may include a positive predictive value (PPV) predicting IOP that includes, for example, values ​​of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0139] In another embodiment, such predetermined conditions may include a negative predictive value (NPV) predicting IOP that includes, for example, values ​​of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0140] In another embodiment, such a predetermined condition may be that the area under the curve (AUC) (AUROC) of the receiver operating characteristic (ROC) curve predicting the IOP includes values ​​of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0141] In another embodiment, such predetermined conditions may include an Area Under the Accuracy Reproducibility Curve (AUPRC) predicted by the IOP having a value of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0142] In some embodiments, the trained model may be trained or configured to predict IOPs with a sensitivity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0143] In some embodiments, the trained model may be trained or configured to predict IOPs using specificities of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0144] In some embodiments, the trained model may be trained or configured to predict IOPs using positive predictive values ​​(PPVs) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0145] In some embodiments, the trained model may be trained or configured to predict IOP using a negative predictive value (NPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0146] In some embodiments, the trained model may be trained or configured to predict IOP using an Area Under Curve (AUC) (AUROC) of the receiver operating characteristic (ROC) curve of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0147] In some embodiments, the trained model may be trained or configured to predict IOP using an Area Under the Accuracy Reproducibility Curve (AUPRC) of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0148] The training dataset may be collected from the step of training subjects (e.g., humans). Each subject to be trained has a diagnostic status indicating whether they are diagnosed and / or classified as high IOP or not. The training procedure, as described elsewhere herein, may be performed for each subject within a group of subjects.

[0149] In some embodiments, machine learning analysis is performed by a device that runs one or more programs (e.g., one or more programs stored in non-persistent memory or persistent memory) that include instructions for performing data analysis. In some embodiments, data analysis is performed by a system comprising at least one processor (e.g., a processing core) and memory (e.g., one or more programs stored in non-persistent memory or persistent memory) that include instructions for performing data analysis.

[0150] The steps for training an ML model may, in some cases, include selecting one or more untrained data models to train using a training dataset. The selected untrained data models may include any type of untrained ML model for supervised, semi-supervised, self-supervised, unsupervised machine learning, or any combination thereof. The selected untrained data models may be defined based on inputs (e.g., user inputs) that define relevant parameters to be used as predicted variables, or other variables to be used as latent explanatory variables. For example, the selected untrained data models may be defined to produce outputs (e.g., predictions) based on inputs. Conditions for training an ML model from the selected untrained data models may be selected, such as limits on ML model complexity and / or limits on ML model refinement beyond a certain point. The ML model may be trained using the training dataset (e.g., via a computer system such as a server). In some cases, a first subset of the training dataset may be selected for training an ML model. The selected untrained data models may then be trained with respect to the first subset of the training dataset using appropriate ML techniques based on the type of ML model selected and any conditions defined for training the ML model. In some cases, due to the processing power requirements of the steps for training the ML model, the selected untrained data model may be trained using additional computing resources (e.g., cloud computing resources). Such training may continue in some cases until at least one aspect of the ML model is demonstrated and satisfies the selection criteria for use as a predictive model as described elsewhere herein.

[0151] In some cases, one or more aspects of an ML model may be demonstrated using a second subset of the training dataset (e.g., distinctly different from a first subset of the training dataset) to determine the accuracy and robustness of the ML model. Such demonstration may include the step of applying the ML model to the second subset of the training dataset and making predictions derived from the second subset of training data. The ML model may then be evaluated to determine whether its performance is sufficient based on the derived predictions. The sufficiency criteria applied to the ML model may vary depending on the size of the training dataset available for training, the performance of previous iterations of the trained model, user-defined performance requirements, or any combination thereof. If the ML model does not achieve sufficient performance, additional training may be performed. Additional training may include the step of refining the ML model or retraining it with respect to a different first subset of the training dataset, after which the new ML model may be demonstrated and evaluated again. When the ML model achieves sufficient performance, in some cases, the ML model may be stored for current and / or future use. The ML model may also be stored as a set of parameter values ​​or weights for the analysis of further inputs (e.g., further predicted variables, further explanatory variables, further user interaction data, or further relevant parameters to be used as any combination thereof), which may include analytical logic or indications of model validity in several cases. In some cases, multiple ML models may be stored to generate predictions under different sets of input data conditions. In some embodiments, the ML model may be stored in a database (e.g., associated with a server).

[0152] This disclosure includes many details of embodiments, but these should not be construed as limitations on the scope of any embodiment or claimed, but rather as descriptions of features specific to particular embodiments. Certain features described in this disclosure in the context of a separate embodiment may also be implemented in combination to a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple separate embodiments or in any preferred secondary combination. Furthermore, features may be described above as acting in a combination, but one or more features from a claimed combination may, in some cases, be removed from that combination, and the claimed combination may be a secondary combination or a variation of a secondary combination.

[0153] Similarly, while operations are depicted in a specific order in these drawings and / or disclosures, this should not be understood as requiring that such operations be performed in a specific indicated order and / or sequential order, or that all illustrated operations be performed, in order to achieve a desired result. Some operations or parts of an operation of a method may be repeated multiple times. In some situations, multitasking and / or parallel processing may be advantageous. Furthermore, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems may, in general, be integrated into a single software product and / or packaged into multiple software products.

[0154] Accordingly, specific embodiments of this subject matter have been described. Other embodiments are also within the scope of the following claims. In some cases, the actions enumerated in the claims may be performed in a different order and still achieve the desired results. In addition, the processes depicted in the accompanying figures do not necessarily require a specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing may be advantageous.

[0155] Preferred embodiments of the present invention are shown and described herein, but it will be apparent to those skilled in the art that such embodiments are provided only as examples. Numerous modifications, alterations, and substitutions will be conjured upon those skilled in the art without departing from the present invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the present invention. The following claims define the scope of the present invention, and it is intended that methods and structures within the scope of these claims and their equivalents are thereby covered. Embodiment

[0156] Embodiment 1 is a method for determining the intraocular pressure (IOP) of an eye, comprising the steps of: acquiring or receiving a first image of at least two markers implanted in the eye, detected using a first detector, and a second image of at least two markers, detected using a second detector, wherein the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector; and determining the IOP of the eye from the positions of one or more of the at least two markers from the first image or the at least two markers from the second image.

[0157] Numbering Embodiment 2 includes the method of Embodiment 1, wherein at least two markers are embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof.

[0158] Numbering embodiment 3 includes the method according to embodiment 1 or 2, wherein at least two markers include fluorescent markers.

[0159] Numbering embodiment 4 includes the method of any one of embodiments 1-3, further comprising the steps of providing a light source to at least two markers implanted in the eye and acquiring or detecting a first image or a second image.

[0160] Numbered Embodiment 5 includes the method according to Embodiment 4, wherein the light source includes a light-emitting diode (LED).

[0161] Numbered embodiment 6 includes the method of embodiment 4 or 5, wherein the light source, the first detector, the second detector, or a combination thereof is at least partially contained within the chassis.

[0162] Numbered Embodiment 7 includes the method according to Embodiment 6, wherein the chassis includes eyewear, glasses, goggles, head-up displays, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

[0163] Embodiment 8 includes the method according to any one of Embodiments 1-7, wherein a first image of at least two markers implanted in the eye is acquired or detected by a first detector through a first polarizing filter, and a second image of at least two markers implanted in the eye is acquired or detected by a second detector through a second polarizing filter.

[0164] Numbered embodiment 9 includes the method according to any one of embodiments 1-8, wherein at least two markers include particles.

[0165] Numbering embodiment 10 includes the method according to any one of embodiments 1-9, wherein at least two markers comprise a first pair of markers and a second pair of markers.

[0166] Numbering embodiment 11 includes the method of embodiment 10, wherein the IOP of the eye is determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers.

[0167] Numbering embodiment 12 includes the method of any one of embodiments 1-11, wherein the positions of at least two markers are provided as input to a machine learning algorithm or predictive model, and the machine learning algorithm or predictive model is trained to provide an output related to the eye's IOP.

[0168] The numbered embodiment 13 is a system for determining the intraocular pressure (IOP) of an eye, comprising: a detector optically coupled to the eye, configured to detect signals from at least two markers implanted in the eye; and one or more processors electrically coupled to the detector, configured to process signals from two or more markers implanted in the eye and to determine the IOP of the eye from the positions of the two or more markers.

[0169] Numbered embodiment 14 comprises the system described in embodiment 13, further comprising a light source optically coupled to the target eye.

[0170] Embodiment 15 comprises the system described in Embodiment 14, wherein the light source includes a light-emitting diode (LED).

[0171] Numbered embodiment 16 comprises the system described in embodiment 14 or 15, wherein the light source and detector are at least partially contained within the chassis, and the chassis is positioned at a distance of up to approximately 2 centimeters from the tangential surface of the eye.

[0172] Numbered embodiment 17 comprises the system described in embodiment 16, wherein the chassis includes eyewear, glasses, goggles, a head-up display, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

[0173] Embodiment 18 comprises the system described in any one of embodiments 13-17, wherein the detector includes a first detector and a second detector.

[0174] Embodiment 19 comprises the system described in Embodiment 18, wherein the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector.

[0175] The numbering embodiment 20 comprises the system described in any one of embodiments 13-19, wherein at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

[0176] Numbered embodiment 21 comprises a system according to any one of embodiments 13-20, wherein the detector includes a camera, a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or any combination thereof.

[0177] The numbered embodiment 22 is a system for determining the intraocular pressure (IOP) of an eye, comprising one or more processors and a memory for storing one or more programs for execution by one or more processors, wherein one or more programs include instructions for performing the steps of (i) acquiring or receiving one or more images of at least two markers implanted in an eye, wherein one or more images are detected using a first detector and a second detector, and the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector, and (ii) determining the IOP of the eye from the distance between at least two markers in one or more images.

[0178] The numbering embodiment 23 comprises the system described in embodiment 22, wherein one or more images are acquired or received from a server, a cloud-based storage device, an eyewear device, or any combination thereof.

[0179] Numbered embodiment 24 comprises the system described in embodiment 23, wherein the eyewear device includes virtual reality eyewear, augmented reality eyewear, or a combination thereof.

[0180] The numbered embodiment 25 comprises the system described in embodiment 23 or 24, wherein one or more processors reside on an eyewear device, a remote processing device, a remote server, a cloud server, or any combination thereof.

[0181] The numbering embodiment 26 comprises the system described in any one of embodiments 22-25, wherein at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

[0182] Embodiment 27 is a method for training a machine learning algorithm or predictive model to determine intraocular pressure (IOP) of an eye, comprising the steps of receiving or acquiring one or more images of at least two markers implanted in one or more eyes and the corresponding IOPs of one or more eyes, and training an untrained or partially untrained machine learning algorithm or predictive model using one or more images of at least two markers implanted in one or more eyes and the corresponding IOPs of one or more eyes, thereby producing a trained machine learning algorithm or trained predictive model.

[0183] Numbered embodiment 28 includes the method of embodiment 27, wherein an untrained or partially untrained machine learning algorithm or predictive model is further trained with respect to the distance between at least two markers in one or more images of at least two markers implanted in one or more eyes.

[0184] Numbering embodiment 29 includes the method according to embodiment 27 or 28, wherein one or more images of at least two markers implanted in one or more eyes are detected using a first detector and a second detector, and the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector.

[0185] Numbering embodiment 30 includes the method according to any one of embodiments 27-29, wherein at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

[0186] Numbered embodiment 31 includes the method of any one of embodiments 27-30, wherein at least two markers are located on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof. definition

[0187] Unless otherwise defined, all technical terms, expressions, and other technical and scientific terms or jargon used herein are intended to have the same meaning as those generally understood by those skilled in the art to which the claimed subject matter pertains. Where there are terms that have a generally understood meaning, terms are defined herein for clarity and / or for ease of reference, and the inclusion of such definitions herein should not necessarily be construed as representing a substantial difference across the generally understood meanings in the art.

[0188] Throughout this application, various embodiments may be presented in scope form. It should be understood that the scope descriptions are for convenience and simplification only and should not be interpreted as inflexible limitations on the scope of this disclosure. Therefore, scope descriptions should be considered to have all conceivable sub-scopes and individual numerical values ​​within those scopes, as specifically disclosed. For example, a scope description such as 1-6 should be considered to have specifically disclosed sub-scopes such as 1-3, 1-4, 1-5, 2-4, 2-6, 3-6, and individual numerical values ​​within those scopes, e.g., 1, 2, 3, 4, 5, and 6. This applies regardless of the scope.

[0189] The reference to "or" can be interpreted as comprehensive, meaning that any term described using "or" may refer to a single term, more than one, or all of the terms described.

[0190] As used herein and in the claims, the singular forms “a,” “an,” and “the” include plural nouns unless the context clearly indicates otherwise. For example, the term “a sample” includes multiple samples, including a mixture thereof.

[0191] The terms "microparticle," "fluorescent particle," "bead," "fluorescent bead," "particle," and "marker" are used synonymously throughout this application.

[0192] The terms “subject,” “individual,” and “patient” are often used synonymously in this specification. “Subject” can be a biological entity containing expressed genetic material. A biological entity can be a plant, animal, or microorganism, including, for example, bacteria, viruses, fungi, and / or protozoa. A subject can be a tissue, cell, and / or offspring of a biological entity, obtained in vivo or cultured in a test tube. A subject can be a mammal. A mammal can be a human. A subject may be diagnosed with a disease, such as glaucoma, or suspected to be at high risk thereof. In some cases, a subject is not necessarily diagnosed with a disease or suspected to be at high risk thereof.

[0193] As used herein, the term “about” refers to a number within ±10% of that number. The term “about” range refers to a range from -10% of its lowest value to +10% of its highest value.

[0194] As used herein, the terms “treatment” or “treating” are used to refer to a pharmaceutical or other interventional administration plan for obtaining beneficial or desired outcomes in a recipient. Beneficial or desired outcomes include, but are not limited to, therapeutic benefits and / or preventive medical benefits. Therapeutic benefits may mean the elimination or improvement of symptoms or the underlying ailment being treated, e.g., a reduction in IOP for one or more subjects. Therapeutic benefits may also be achieved with the elimination or improvement of one or more physiological symptoms associated with the underlying ailment, such that the improvement is observed in the IOP of one or more subjects, even though the subject may still be suffering from the underlying ailment. Preventive medical effects include delaying, preventing, and / or eliminating the onset of disease and / or condition; delaying and / or eliminating the onset of symptoms of disease or condition; slowing, interrupting, and / or reversing the progression of disease and / or condition; or any combination thereof. Regarding preventive medical benefits, individuals at risk of developing a particular disease, and / or those experiencing one or more of the physiological symptoms of the disease, may receive treatment even if they have not been diagnosed with the disease.

Claims

1. A method for determining the intraocular pressure (IOP) of the eye, wherein the method is: The method involves acquiring or receiving a first image of at least two markers embedded in the eye detected using a first detector, and a second image of the at least two markers detected using a second detector, wherein the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector. Determining the IOP of the eye from the positions of at least two markers from the first image or one or more of the at least two markers from the second image. Methods that include...

2. The method according to claim 1, wherein the at least two markers are embedded on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or in any combination thereof.

3. The method according to claim 1, wherein the at least two markers include fluorescent markers.

4. The method according to claim 1, further comprising providing a light source to the at least two markers embedded in the eye, and acquiring or detecting the first image or the second image.

5. The method according to claim 4, wherein the light source includes a light-emitting diode (LED).

6. The method according to claim 4, wherein the light source, the first detector, the second detector, or a combination thereof is at least partially contained within a chassis.

7. The method according to claim 6, wherein the chassis includes eyewear, glasses, goggles, a head-up display, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

8. The method according to claim 1, wherein the first image of the at least two markers implanted in the eye is acquired or detected by the first detector through a first polarizing filter, and the second image of the at least two markers implanted in the eye is acquired or detected by the second detector through a second polarizing filter.

9. The method according to claim 1, wherein the at least two markers include particles.

10. The method according to claim 1, wherein the at least two markers comprise a first pair of markers and a second pair of markers.

11. The method according to claim 10, wherein the IOP of the eye is determined from the ratio of the distance between the first and second markers of the first pair of markers to the distance between the third and fourth markers of the second pair of markers.

12. The method according to claim 1, wherein the positions of the at least two markers are provided as input to a machine learning algorithm or predictive model, the machine learning algorithm or predictive model is trained to provide an output related to the IOP of the eye.

13. A system for determining the intraocular pressure (IOP) of the eye, wherein the system is A detector optically coupled to the eye, wherein the detector is configured to detect signals from at least two markers embedded in the eye, One or more processors electrically coupled to the detector, wherein the one or more processors are configured to process the signals of the two or more markers implanted in the eye and to determine the IOP of the eye from the positions of the two or more markers. A system that includes these features.

14. The system according to claim 13, further comprising a light source optically coupled to the eye of the subject.

15. The system according to claim 14, wherein the light source includes a light-emitting diode (LED).

16. The system according to claim 14, wherein the light source and detector are at least partially contained within a chassis, and the chassis is positioned at a distance of up to about 2 centimeters from the tangential surface of the eye.

17. The system according to claim 16, wherein the chassis includes eyewear, glasses, goggles, a head-up display, virtual reality eyewear, augmented reality eyewear, or any combination thereof.

18. The system according to claim 13, wherein the detector includes a first detector and a second detector.

19. The system according to claim 18, wherein the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector.

20. The system according to claim 13, wherein the at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

21. The system according to claim 13, wherein the detector includes a camera, a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or any combination thereof.

22. A system for determining the intraocular pressure (IOP) of the eye, wherein the system is One or more processors and memory, wherein the memory stores one or more programs for execution by the one or more processors, and the one or more programs are (i) Acquiring or receiving one or more images of at least two markers implanted in the eye, wherein the one or more images are detected using a first detector and a second detector, and the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector, (ii) Determining the IOP of the eye from the distance between at least two markers in one or more images. One or more processors and memory containing instructions for performing the following actions A system that includes these features.

23. The system according to claim 22, wherein one or more images are acquired or received from a server, a cloud-based storage device, an eyewear device, or any combination thereof.

24. The eyewear device includes virtual reality eyewear, augmented reality eyewear, or a combination thereof, according to claim 23.

25. The system according to claim 23, wherein the one or more processors are located on the eyewear device, the remote processing device, the remote server, the cloud server, or any combination thereof.

26. The system according to claim 22, wherein the at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

27. A method for training a machine learning algorithm or predictive model to determine the intraocular pressure (IOP) of an eye, wherein the method is: Receiving or acquiring images of one or more markers implanted in one or more eyes and corresponding IOPs of the one or more eyes, Using the images of the at least two markers implanted in one or more eyes and the corresponding IOPs of the one or more eyes, an untrained or partially untrained machine learning algorithm or predictive model is trained, thereby producing a trained machine learning algorithm or trained predictive model. Methods that include...

28. The method according to claim 27, wherein the untrained or partially untrained machine learning algorithm or predictive model is further trained with respect to the distance between the at least two markers in the one or more images of the at least two markers implanted in the one or more eyes.

29. The method according to claim 27, wherein the images of the at least two markers implanted in the one or more eyes are detected using a first detector and a second detector, and the optical axis of the first detector is at a certain angle with respect to the optical axis of the second detector.

30. The method according to claim 27, wherein the at least two markers are positioned at a distance of up to approximately 6 millimeters from each other.

31. The method according to claim 27, wherein the at least two markers are located on the surface of the eye, within the cornea of ​​the eye, within the lens of the eye, within tissue adjacent to the eye, or any combination thereof.

Images