Apparatus and method for estimating refractive error from measured visual acuity

Abstract

The refractive error of a user's eye is estimated by receiving a measure (21) of visual acuity VA of the user's eye, referred to as a reference VA measure, and data (22) relating to different, respectively associated blur levels, referred to as reference blur data, referred to as a reference blur data, referred to as a reference blur data, referred to as a reference blur data, of the user's eye. Reference VA measurements are performed within a reference measurement period by applying respective levels of blur on the user's eyes. A generic model (23) relating to the theoretical VA, the theoretical blur data and the theoretical refractive error is embodied as a specific model (24) of the eye of the user from the reference VA measurement and the reference blur data, and information (25) relating to the refractive error of the eye of the user is generated and provided from the specific model. The method is applied to vision follow-up visit and myopia, hyperopia and astigmatism control.

Description

Technical Field

[0001] This invention relates to the field of visual acuity testing, and more specifically to the estimation of refractive errors. Background Technology

[0002] Refractive errors are related to an impaired ability of the eye to focus light accurately on the retina. They are also known as refractive inaccuracies. They are closely related to the shape of the eye, especially the cornea and the length of the eyeball.

[0003] Refractive errors can be associated with myopia (nearsightedness) or nearsightedness, which corresponds to an excessively long eyeball causing light to focus in front of the retina. Distant objects appear blurry, while near objects are clearly visible. Alternatively, refractive errors can be associated with hyperopia (farsightedness) or farsightedness, which corresponds to an eyeball that is too short, causing light to focus behind the retina. Near objects also appear blurry, while distant objects are clearly visible. Astigmatism is a type of refractive error in which the eye's refractive power is affected by rotational asymmetry, which is associated with an incorrect shape of the cornea causing blurred vision at any distance. As for presbyopia, it is related to a decline in accommodative ability with age, making it difficult to focus clearly on near objects.

[0004] The ability to diagnose the presence, nature, and severity of refractive errors is crucial for eye care professionals (hereinafter referred to as "ECPs") such as orthodontists, optometrists, or ophthalmologists. This can rely on objective refraction, which utilizes objective measures such as using specialized equipment (e.g., a retinal speculum) (allowing the examiner to shine light into the patient's eye and observe the reflection from the retina) or an automated refractometer (a computer-controlled machine that measures how light is affected as it is reflected by the eyeball), or subjective refraction, which gathers subjective feedback from the patient regarding the appropriateness of lenses with different powers placed in front of the tested eye to correct vision. Subjective refraction is typically performed after objective refraction to refine the obtained measurements.

[0005] A diagnosis of refractive error allows the ECP to determine the appropriate treatment, which may include correction via eyeglasses or contact lenses or refractive surgery. Other settings may include, for example, encouraging children to spend more time outdoors to prevent or slow the progression of myopia.

[0006] While measurements taken by well-equipped professionals in a medical setting are essential for obtaining a relevant and accurate diagnosis, autonomous mobility resources, which are suitable for other settings and may be used by laypeople to estimate refractive errors, can be helpful in a variety of situations.

[0007] For example, having parents monitor their child's eyes at home between two medical appointments with an ophthalmologist can allow for alerting in cases of significant worsening myopia and allows for anticipated appointment scheduling when needed.

[0008] Moreover, regularly collected refractive error data (even rough approximations) may help improve ECP diagnosis by providing a better understanding of its past evolutionary history.

[0009] Additionally, in places or periods where it is practically difficult to meet with professionals in a healthcare setting for any reason (e.g., absence of professionals, distance, reduced mobility, or obstruction), remote online medical consultations based in part on user-estimated refractive errors may be helpful for routine follow-ups, frontline examinations, or patient orientation.

[0010] Several portable solutions have been developed for estimating refractive errors. In fact, they are typically based on visual acuity, or VA, which is the ability to accurately identify small details, expressed as the spatial resolution of the visual processing system. Since refractive errors are a common cause of low visual acuity, they can often be reflected in visual acuity, although neural factors may alternatively or additionally involve the retina (e.g., macular degeneration) or the brain (e.g., damage following traumatic injury).

[0011] Visual acuity (VA) is typically measured subjectively by relying on optometric fonts or visual acuity chart symbols placed at a standardized distance from the patient for identification. Therefore, the smallest reliably identifiable symbol is considered to represent the patient's VA. In Europe, the standardized distance used to simulate optical infinity is typically 6 meters (especially as specified in the European standard ENISO 8596), and in the United States, this standardized distance is typically 20 feet.

[0012] Refractive errors and VA are particularly related because in subjective refraction, the appropriate lens or lens combination is the lens or lens combination that provides the best correction for VA or BCVA.

[0013] Portable vision testing solutions are described, for example, in Eyeque Inc.'s patent application WO2020 / 176784. The disclosed device includes a binocular observer attached to a smartphone and can implement a combination of lens anterior and posterior surfaces, a reduction system, and other systems to replicate the user-perceived line of sight in a conventional VA test, but in a fairly compact manner. Additionally, an adjustable lens system providing variable power can be inserted into the device. Thus, the user can set the power of the adjustable lens system to achieve BCVA, thereby inferring refractive errors (see §17-18).

[0014] While potentially providing accurate refractive error estimation, this solution requires specialized equipment for the user, including specific devices and associated adjustable lenses.

[0015] In patent application WO 2019 / 089600 granted to Welch Allyn Inc., a vision screening device performs a photographic refractive eye screening test to generate refractive errors. The latter is used to determine the VA test equivalent, which involves adjusting the refraction font size in a Snellen chart test performed by the device in use. This solution could be used to identify neurological or other problems causing discrepancies between refractive error measurements and VA examinations.

[0016] Besides requiring specialized equipment, this solution may also be costly and difficult for users to operate reliably.

[0017] Patent application WO 2014 / 164020 granted to Spleee and A. Dallek proposes computerized refractive and astigmatic determination that does not require any lenses and can be performed at home, depending on the patient’s proper compliance with automated instructions, including the distance relative to the computerized screen (see, for example, §81-82)).

[0018] In fact, no effective refractive error estimation is performed in this disclosure; this disclosure focuses entirely on VA measurement.

[0019] Another solution for estimating refractive errors was developed in patent application WO 2014 / 195951 granted to O. Limon, OKAncri, and A. Zlotnik. The refractive error of the subject's eye is derived from the maximum distance of optimal visual acuity (or MDBA), which is the maximum distance at which the subject can identify a symbol displayed in a target image. Once the MDBA is properly estimated, the relevant refractive error of the eye is calculated based on the MDBA and the characteristics of the target image, without requiring the subject to try different corrective lenses.

[0020] This solution offers the opportunity to infer refractive errors based on VA measurements across a wide range of eye conditions, including myopia, hyperopia, presbyopia, and astigmatism. However, positioning the subject at the precise desired distance, while crucial in the described method, is highly sensitive to approximations and errors arising from the subject's subjective judgment (see §49, 107, and 211 for details).

[0021] In fact, although the relationship between refractive error and VA has been carefully studied for a long time, its complexity makes the transition from one user application to another very subtle and error-prone.

[0022] In this regard, see the groundbreaking article “Phenomenological model of visual acuity” by J. Gómez-Pedrero and J. Alonso in the Journal of Biomedical Optics, Vol. 125005, No. 21(12), 2016 (hereinafter referred to as “Gómez-Pedrero’s article”), which describes VA as a function of defocus and pupil diameter, taking into account inter-individual variability in astigmatism and BCVA. See also the introduction to related past works in that publication. Additional material can be found, for example, in “Influence of combined power error and astigmatism on visual acuity” by C. Fauquier et al. in Vision Science and its Applications, Technical Abstracts Series (Optics Publishing Group), 1995.

[0023] Therefore, estimating refractive errors using portable devices in a sufficiently reliable manner without requiring specialized equipment remains quite challenging. Similarly, safely deriving refractive errors from VA measurements is not an easy task. Summary of the Invention

[0024] The purpose of this disclosure is to provide a potentially reliable method for estimating refractive errors based on visual acuity measurements without the need for specialized equipment.

[0025] Another purpose of this disclosure is to enable users without medical status or training to make such an estimate.

[0026] An additional objective of this disclosure is to enable the estimation of refractive errors using common mobile devices such as smartphones, tablets, or laptops.

[0027] The specific applications related to refractive errors disclosed herein include intermediate monitoring between medical appointments, historical data collection and recording, with the aim of enhancing medical diagnosis, and remote online medical consultations.

[0028] Preliminary definition

[0029] The following definition supplements the above content.

[0030] "User" is currently defined as a person or animal, patient or subject who has visual ability and uses the disclosed device or method. The user may or may not be a person or device operating the device or performing the method.

[0031] "Blur" refers to defocus aberration, i.e., an image that is out of focus. In this disclosure relating to visual acuity, blur refers to the subjective perception of refractive error, encompassing any of the foregoing forms. In this context, "blurred data" refers to data relating to the type, level, and / or source of visual blur.

[0032] "Optical power" refers to the degree to which an optical system converges or diverges light. For an optical system with a focal length, the optical power is the reciprocal of that focal length, and is given in SI units (International System of Units) in diopters (the reciprocal of the meter). Conventionally, converging and diverging optical systems have positive and negative optical powers, respectively. Therefore, myopia is usually corrected with negative power lenses, while hyperopia or presbyopia is usually corrected with positive power lenses.

[0033] "Minimum Resolved Angle" or "MAR" is a commonly used scale. When expressed in arcminutes, it corresponds to the reciprocal of VA in decimal form.

[0034] The user's "field of vision" is the entire area that the eye can detect visual stimuli, and the "fixation point" is the specific point in the user's field of vision that the eye is looking at.

[0035] In the perimeter method (i.e., the measurement of visual function at a morphologically defined point in the visual field), the "meridian" is used to specify a point in the user's eye's visual field. It corresponds to the polar axis in a polar coordinate system with the fixation point as its origin, and therefore corresponds to the polar angle. Consistently, the meridian also corresponds to the plane that includes the corresponding polar axis and the optical axis that connects the user's eye to the fixation point or to a plane parallel to the fixation point.

[0036] By convention, in this respect, the user has a sagittal plane (i.e., the anatomical plane that divides the user's body into right and left segments), and the "horizontal" line is defined as a line perpendicular to this sagittal plane. Thus, the horizontal meridian is associated with 0° or 180°, the latter being used in eyeglass prescriptions, and the vertical meridian is associated with 90°.

[0037] Refractive error correction in an eyewear prescription can be expressed as "spherical-cylindrical correction." The latter includes the "spherical component" (or "spherical element"), which has the same power in every direction perpendicular to the optical axis of the eyewear. This can be achieved by using spherical lenses that meet this specificity. In the absence of astigmatism correction, eyewear correction can be limited to the spherical component, and the relevant prescription typically includes "DS" (representing "spherical power") after the spherical power.

[0038] In the case of astigmatism, correction further includes an "axial component," which is given by a meridian associated with the power of the spherical component. This meridian corresponds to either the highest or lowest corrected power, depending on the sign, and is expressed in degrees (polar angle). The "cylindrical component" (or "cylindrical part") is given by the offset between the lowest and highest corrected powers. The corrected power furthest from the spherical component power is further associated with an offset of +90° from the meridian of the axial component, and is designated hereinafter as the "vertical axis."

[0039] More precisely, in the "positive cylindrical notation," the cylindrical component is positive, causing the axial component and vertical axis to give the lowest (most divergent meridian) and highest (most convergent meridian) corrective powers, respectively. In contrast, in the "negative cylindrical notation," the cylindrical component is negative, causing the axial component and vertical axis to give the highest and lowest corrective powers, respectively.

[0040] For astigmatic users, refractive error correction can also be expressed as "equivalent spherical power," which is the average focal power value that takes into account both the spherical and cylindrical components, and can, for example, be the intermediate value between the highest and lowest optical power. It corresponds to the average curvature of the wavefront.

[0041] By extension, refractive errors are usually specified by the required refractive error correction, and this convention will be adopted below.

[0042] As described above, optometric fonts are visual acuity chart symbols placed at a standardized distance from the patient for identification. In the following text, the term will be used more generally to refer to a representation displayed to the patient on a chart at a given effective or analog distance (e.g., via virtual reality) for identification or perception, which may include one or more symbols and / or gratings having a given shape, size, and orientation.

[0043] In the following text, a “directional test target” is an optometric font or set of optometric fonts displayed on a visual acuity chart and distinguished relative to at least one specific direction corresponding to at least one specific meridian (i.e., polar angle), suitable for determining the axial and / or cylindrical components of the refractive error in the eye of an astigmatic user. A directional test target may, for example, include parallel lines suitable for rotation about the optical axis, or lines evenly spaced at regular angular intervals. In contrast, a “non-directional test target” refers to an optometric font or set of optometric fonts displayed on a visual acuity chart and not specifically distinguished relative to a specific direction for determining the axial and / or cylindrical components. Such a non-directional test target may be associated with determining the spherical power of a non-astigmatic user or the equivalent spherical power of an astigmatic user.

[0044] "Accommodation" is a process by which a user's eye changes its optical power as the distance to an object changes in order to maintain a clear image or focus on the object. A subject's accommodative ability (i.e., accommodative amplitude) is quantified in diopters by the difference between the optical power of the nearest clear visual acuity (i.e., representing "near point or PP") and the optical power of the farthest clear visual acuity (i.e., representing "far point or PR"), with PR and PP defining the accommodative range.

[0045] The pupil is the opening through which light passes into the retina. This opening is surrounded by the iris, a constrictible structure that adjusts the size of the pupil and thus the amount of light received by the retina. Therefore, the pupil functions similarly to the aperture of an optical system, limiting the size of the light beam entering the eye.

[0046] The terms “suitable” and “configured” are used broadly in this disclosure to cover the initial configuration, subsequent adaptation or supplementation, or any combination thereof, of the device, whether achieved by material or software means (including firmware).

[0047] The term "processor" should not be interpreted as hardware capable of executing software, but rather refers generally to a processing device, which may include, for example, a computer, microprocessor, integrated circuit, or programmable logic device (PLD). Furthermore, instructions and / or data capable of performing associated and / or resulting functions can be stored on any processor-readable medium, such as an integrated circuit, hard disk, CD (optical disc), optical disc (e.g., DVD (Digital Multifunction Disc), RAM (Random Access Memory), or ROM memory). Instructions may be specifically stored in hardware, software, firmware, or any combination thereof.

[0048] The purpose of this disclosure

[0049] This disclosure relates to a device for estimating refractive errors in a user's eye based on the user's visual acuity. The device includes:

[0050] - At least one input terminal, said at least one input terminal being adapted to receive at least two measurements of the visual acuity of a user's eye, referred to as reference VA measurements, and data associated with at least two different blur levels respectively associated with the reference VA measurements, referred to as reference blur data, wherein the reference VA measurements are performed by applying the corresponding blur levels to the user's eye during the reference measurement period.

[0051] - At least one processor configured to concretize a general model involving theoretical visual acuity, theoretical ambiguity data, and theoretical refractive error into a specific model for a user's eye based on reference VA measurements and reference ambiguity data, and to generate information related to the refractive error of the user's eye based on the specific model.

[0052] - At least one output terminal, which is adapted to provide information related to the refractive error of the user's eye.

[0053] Therefore, reference fuzzy data is fuzzy data with specificity associated with two or more different fuzzy levels, each of which is associated with one or more VA measurements of the user's eye.

[0054] The reference measurement period is advantageously chosen to be short enough to avoid physiological evolution of refractive errors in the user's eye during that period, and in some embodiments, the reference measurement period is as short as possible to favor uniform conditions rather than VA measurements. Thus, in certain modes, the reference measurement cycle is less than a day, advantageously less than an hour, and even more advantageously less than 10 minutes.

[0055] On the other hand, the reference measurement period can be chosen to be long enough to facilitate user testing by distributing the testing work over a period of time and / or to enable the repetition of similar tests for proper averaging. This averaging can be used to more accurately reflect the user's current effective refractive error and avoid distortions in results caused by factors such as user fatigue, user distraction, or environmental interference. Averaged VA measurements can be performed under similar environments, ambient lighting, daily hourly intervals, and / or user physical conditions. Alternatively, it can be performed intentionally under different conditions for the same or similar blur levels to mitigate the influence of relevant factors.

[0056] Accordingly, in a specific model, the reference measurement period is between one day and one week.

[0057] A specific model is an expression of one or more relationships between a user's visual acuity and refractive error, conforming to a general model with one or more parameters, wherein at least some of these parameters are defined. Parameters are defined as variables or multivalued elements associated with the general model that can be determined or evaluated. Parameters can, in particular, be terms in a function representing the general model, whose values ​​are selected from two or more forms to form a specific form of the function, or independent variables in a set of parametric equations. Parameters can also be indicators that determine the contents of the general model among two or more possible contents. For clarity, observed relevant parameters do not include measured visual acuity or expected refractive error. Reference fuzzy data may include parameter values ​​directly related to one or more parameters in the general model. A specific model is derived from a general model by reducing the number of one or more parameters or by extending the domain.

[0058] In some implementations, the resulting information relating to the user's eye's refractive error corresponds to an estimate of the refractive error during a reference measurement period. Accordingly, unknown values ​​(including parameters and refractive errors) in a general model are explicitly or implicitly acquired or rejected, whether through direct input, determination, estimation, elimination (e.g., in a system of equations), realistic guessing, or random selection within a given range. In other implementations, the resulting information corresponds to one or more parameter values ​​for a particular model, which can then be used to estimate the refractive error based on visual acuity measurements following the reference measurement period.

[0059] By utilizing a universal model and based on those reference fuzzy data and associated VA measurements, it is possible to produce an estimate of refractive error, rather than just a hint of refractive error.

[0060] Furthermore, compared to the solution developed in WO 2014 / 195951 that derives the calculation of refractive error from MDBA, the estimation of refractive error using the apparatus disclosed herein relies particularly on at least two VA measurements, each associated with a different level of ambiguity.

[0061] In some implementations, the disclosed apparatus can provide accurate and precise estimates of refractive errors by taking into account two or more levels of ambiguity, even when individual VA measurements may be limited by the finite level of available accuracy.

[0062] As technicians will notice, the solution unexpectedly deviates from the known processes and theoretical approaches proposed to date, which establish a one-to-one correspondence between VA and refractive errors affected by specific conditions (such as optimal visual acuity or given distance and pupil size) in the closest configuration.

[0063] Furthermore, the device disclosed herein can be used for follow-up of users with various vision conditions, including myopia, hyperopia, astigmatism, or presbyopia.

[0064] By enabling the estimation of refractive errors via VA measurement, the device disclosed herein allows users to continue refractive error estimation outside of medical settings (including possibly at home) in some modes, and / or can be associated with proper operation by users without medical status or training. Such VA measurements can be performed, for example, using portable or handheld devices (such as smartphones, tablets, or laptops).

[0065] Human visual acuity can be assessed, for example, as described in patent application WO 2022 / 013410 granted to Essilor International. According to that disclosure, a user is positioned at a predefined distance in front of a reflector, a mobile device including a camera and a screen is arranged with its front-facing camera facing the reflector, the distance between the front-facing camera and a virtual image of the mobile device in the reflector is measured, optometry text is displayed on the screen of the mobile device, and the user's visual acuity (VA) is assessed while the user views the optometry text.

[0066] As described above, the device disclosed herein can therefore be associated with intermediate monitoring between medical appointments, data history collection and recording (with the aim of enhancing medical diagnosis), or remote online medical consultations. It can also prompt the user to schedule a new appointment with the ECP when a significant change in VA is identified as requiring new refractive error correction. In cases of myopia in children, it can further enable the triggering of myopia control actions. The latter may include regularly spending more time outdoors and / or wearing specialized glasses designed to slow myopia progression, such as those manufactured by the applicant under the name "Stellest". TM "Those specialized eyeglasses with commercially available lenses (which rely on a continuous distribution of highly aspherical small lenses). The device also makes it possible to monitor the efficiency of this action."

[0067] In some implementations, the processor is configured to determine additional items from the reference ambiguity data, which may be used to obtain additional related items from the reference VA measurement. The reference ambiguity data and those additional items from the reference VA measurement are then input into the processor to generate refined and / or completed information items related to the refractive error. This feedback mechanism may iterate once or multiple times, taking into account the general model used and / or the target accuracy or precision level.

[0068] In the first (above) category of modes, the resulting information related to refractive errors includes an estimate of the user's eye refractive error during the reference measurement period.

[0069] In a second (as described above) category of modes that can be combined with modes of the first category in the same device, the resulting information related to refractive error includes at least one parameter value associated with a particular model. That parameter value enables the generation of an estimate of the user's eye's refractive error, referred to as a later VA measurement, based on at least one measurement of the user's eye's visual acuity after the reference measurement period.

[0070] In this respect, a particular model can be established, either fully or partially, during the reference measurement period and later used to estimate the evolution of refractive error over time.

[0071] Parameter values ​​can be further determined at a later time after the reference measurement period to update the specific model to account for physiological evolution. This update can be repeated whenever it is deemed useful or convenient, such as at regular time intervals or during ECP consultations.

[0072] In a specific implementation of the second category of modes, the input is adapted to receive a later VA measurement, the processor is configured to generate a later refractive error estimate based on a specific model and the later VA measurement, and the output is adapted to provide the later refractive error estimate.

[0073] That is, the device is then configured to perform the generation of a later refractive error estimate on its own.

[0074] In some variations, the device is adapted to make parameter values ​​associated with a particular model available to an external module responsible for generating subsequent refractive error estimates (directly or, for example, via a storage unit). In some modes, the device itself may not be configured for such estimation. For example, two different components are used, one for determining the particular model during a reference measurement period and the other for utilizing that particular model during a later period.

[0075] In some implementations of the second category of modes, the reference fuzzy data includes at least one measurement of the user's eye refractive error during the reference measurement period, referred to as the reference refractive error measurement.

[0076] Therefore, this measurement of refractive error is input into a device for materializing a general model so as to obtain a later estimate of the refractive error after a reference measurement period. In a more specific implementation, the estimate of the refractive error during the reference measurement period is looped back to the reference refractive error measurement in order to adjust or improve a particular model or to continue model calibration, which can be done, for example, iteratively.

[0077] The device can generate a later refractive error estimate by calculating the refractive error deviation relative to a reference refractive error measurement, which is a function of the deviation of the later VA measurement relative to the reference VA measurement.

[0078] In some other modes, the blur level is applied at least in part by performing reference VA measurements at two or more reference distances from the user's eye to the test target, and the theoretical blur data includes the theoretical distance from the user's eye.

[0079] In a more specific implementation, the general model successively involves three regions, starting from the user's eye and moving away from the user's eye. These three regions are associated with the user's theoretical visual acuity (VA) and include a nearest, accommodation, and farthest region. The accommodation region is associated with the flat range of the theoretical VA, and the nearest and farthest regions are associated with variable values ​​of the theoretical VA. The ambiguity level is then applied, at least in part, by performing reference VA measurements at reference distances that include at least one distance corresponding to the accommodation region and at least one distance corresponding to the nearest and / or farthest regions, respectively.

[0080] Conventionally, the farthest zone can be extended to infinity or even virtual (i.e., optically behind the user's eye), meaning that without additional lenses, only the nearest and accommodative zones are effectively accessible. This can reflect whether the user's eyes are emmetropic (normal vision) or hyperopic (farsighted), or whether the user has presbyopia (presbyopia) but no myopia (nearsightedness).

[0081] For example, the device is particularly suitable for myopia follow-up, performing reference VA measurements at one distance in the accommodation region and two distances in the farthest region. In another example where the device is particularly suitable for presbyopia or hyperopia follow-up, reference VA measurements are performed at one distance in the accommodation region and two distances in the nearest region. In an additional example focused on hyperopia, reference VA measurements are performed at at least one distance in the accommodation region while applying positive power correction to the user's eye, and it is also possible to perform reference VA measurements at at least one distance in the nearest region, as developed below.

[0082] In some other modes that can be combined with the previous modes, at least one of the ambiguity levels is applied, at least in part, by performing at least one of the reference VA measurements at at least one reference orientation angle of the oriented test target, the theoretical ambiguity data including the theoretical orientation angle relative to a predetermined axis, and the processor is configured to also concretize the general model according to the reference orientation angle.

[0083] The device is then configured to match one or more reference orientation angles with theoretical orientation angles when materializing a general model into a specific model.

[0084] For example, reference VA measurements are performed at two different reference orientation angles of the orientation test target, those orientation angles may be orthogonal, i.e., separated by ±90°.

[0085] In a specific implementation of the mode involving the orientation test target, the user's eye refractive error has a spherical component, a cylindrical component, and a cylindrical axis, while the theoretical refractive error has a theoretical spherical component, a theoretical cylindrical component, and a theoretical cylindrical axis. Therefore, the device enables:

[0086] - The input is adapted to receive at least one of reference VA measurements relative to a non-directional test target, reference VA measurements at a reference orientation angle of the directional test target, and associated reference ambiguity data, the reference ambiguity data further including information about at least one offset of the cylindrical lens axis relative to a predetermined axis.

[0087] The processor is configured to concretize a general model into a specific model based on reference VA measurements for both non-directional and directional test targets and reference fuzzy data, and to generate information related to the spherical and cylindrical components of the user's eye's refractive error.

[0088] - The output is adapted to provide information related to the spherical and cylindrical components.

[0089] Therefore, in some implementations, combining the use of non-directional and directional test targets can prove to be quite efficient. It is noteworthy that non-directional test targets can be used to derive the equivalent spherical power of the user's eye's refractive error, while directional test targets can be used to derive the cylindrical component, thereby deriving the spherical component—which can be given, for example, by the difference between the equivalent spherical power component and half of the cylindrical component.

[0090] In a particular mode, at least one of the blur levels is applied, at least in part, by performing at least one of the reference VA measurements with at least one lens arranged in front of the user's eye, so as to cause at least one predetermined optical power correction, and the theoretical blur data includes the theoretical optical power correction applied to the user's eye.

[0091] The lens can have positive power, in which case the PP and PR images passing through the lens move closer to the user's eye, or it can have negative power, in which case the PP and PR images passing through the lens move further away from the user's eye.

[0092] In a more specific embodiment, in the general model, the change in theoretical refractive error is equal to the theoretical optical power correction.

[0093] This is equivalent to predicting the potential evolution of refractive errors by equating, on the one hand, the relationship between the refractive error deviation over time and the VA deviation over time, and on the other hand, the relationship between the applied refractive error offset and the resulting change in VA during the reference measurement period.

[0094] In some modes:

[0095] - The input is adapted to receive pupil size data about the user's eye during a reference VA measurement, referred to as reference pupil data.

[0096] - The general model involves theoretical pupil size, and the processor is configured to further refine the general model into specific models based on reference pupil data.

[0097] Pupil size data can be used as at least part of reference fuzzy data in other modes (possibly in combination with previous modes), i.e., including different values ​​associated with two or more reference VA measurements respectively, and relying on these different values ​​to concretize a particular model.

[0098] Additionally, this disclosure relates to an automated module for myopia control, which includes means for estimating the refractive error of a user's eye according to any mode of this disclosure.

[0099] This disclosure also relates to a device selected from smartphones, tablets, laptops, and virtual reality head-mounted display (HMD) systems, which includes means for estimating refractive errors according to any mode of this disclosure.

[0100] This disclosure further relates to a method for estimating refractive errors in a user's eye by at least one processor based on the visual acuity of the user's eye. According to this disclosure, the method includes:

[0101] - Receive at least two measurements of the visual acuity of the user's eye, referred to as reference VA measurements, and data associated with at least two different blur levels respectively associated with the reference VA measurements, referred to as reference blur data, wherein the reference VA measurements are performed by applying the corresponding blur levels to the user's eye during the reference measurement period.

[0102] - For the user's eyes, based on reference VA measurements and reference fuzzy data, a general model involving theoretical visual acuity, theoretical fuzzy data, and theoretical refractive error is concretized into a specific model, and information related to the user's eye's refractive error is generated based on the specific model.

[0103] - Provides information related to the user's refractive errors.

[0104] The method for estimating refractive errors is advantageously performed by an apparatus for estimating refractive errors conforming to any mode of this disclosure.

[0105] Additionally, this disclosure relates to a computer program comprising software code that, when executed by a processor, is adapted to perform a method for estimating refractive errors conforming to any of the above-described execution modes.

[0106] This disclosure further relates to a computer-readable non-transitory program storage device that tangibly implements an instruction program that can be executed by a computer to perform a method for estimating refractive errors consistent with this disclosure.

[0107] Such non-transitory program storage devices can be, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any suitable combination of the foregoing. It should be understood that the following, while providing more specific examples, is merely illustrative and not exhaustive, and includes, as will be readily understood by one of ordinary skill in the art: portable computer floppy disks, hard disks, ROM (Read-Only Memory), EPROM (Erasable Programmable ROM), or flash memory, portable CD-ROM (Compact Optical Disc ROM). Attached Figure Description

[0108] This disclosure will be better understood by reading the following description of specific and non-limiting illustrative embodiments, and other specific features and advantages will become apparent, which is made with reference to the accompanying drawings, in which:

[0109] Figure 1 This is a block diagram schematically representing a specific pattern of a device for estimating refractive errors in a user's eye, consistent with this disclosure.

[0110] Figure 2 The implementation of the first category is shown. Figure 1 The device, wherein an estimate of refractive error can be obtained directly during a reference measurement period;

[0111] Figure 3 The second category of implementation methods is shown. Figure 1 A block diagram of an apparatus in which a particular model can be obtained during a reference measurement period, and an additional apparatus configured to utilize the particular model at a later time to obtain an estimate of the refractive error at a later time;

[0112] Figure 4 It is a schematic representation of... Figure 3 The second category of implementations also includes a block diagram of a device that covers the functions of other devices;

[0113] Figure 5 This demonstrates the effect of a reference measurement period and multiple later measurement periods over time. Figure 3 or Figure 4 The use of the device;

[0114] Figure 6 An exemplary curve showing the VA change of a myopic user's eye as a function of distance from the user's eye is shown, and the effect of utilizing... Figure 1 The possible positions selected on the curve for a specific embodiment of the device;

[0115] Figure 7The diagram shows a typical curve of the MAR change in the eyes of myopic users as a function of the inverse of the distance from the user's eyes, depending on the underlying relationship with... Figure 1 The device utilizes the parameters of defocus tolerance in the general model;

[0116] Figure 8 An exemplary curve showing the change in visual acuity (VA) of a farsighted user's eye as a function of the reciprocal of the distance from the user's eye is shown, and the effect of utilizing... Figure 1 The possible positions selected on the curve for a specific embodiment of the device;

[0117] Figure 9 The diagram illustrates the variation in visual acuity (VA) of the eyes of astigmatic and myopic users as exemplary curves representing distance from the user's eye, corresponding to the mean VA, VA along the astigmatic axis, and VA along an axis perpendicular to the astigmatic axis, as may be shown. Figure 1 Utilized in a specific embodiment of the device;

[0118] Figure 10A , Figure 10B , Figure 10C , Figure 10D , Figure 10E and Figure 10F It indicates the use of Figure 1 The embodiment of the device uses an exemplary optometric font when measuring astigmatism, and it is intended to estimate the average VA value corresponding to the equivalent spherical power of the refractive error. Figure 10A ), astigmatic axis and axial VA value;

[0119] Figure 11 The diagram shows a typical curve of the change in a user's eye's MAR as a function of the user's pupil diameter, as shown in... Figure 1 This may be utilized in specific embodiments of the device;

[0120] Figure 12A and Figure 12B This indicates the use of Figure 1 The device consists of two corresponding steps performed using a handheld device and a reflector to monitor the pupil diameter of the user's eye;

[0121] Figure 13 It shows the use of Figure 1 A flowchart of the method steps performed by the device;

[0122] Figure 14 It schematically shows the combination of Figure 1 The device that represents the apparatus.

[0123] In the accompanying drawings, the same or similar elements are indicated by the same reference numerals. Detailed Implementation

[0124] This specification illustrates the principles of this disclosure. Therefore, it will be understood that those skilled in the art will be able to design various arrangements that, although not expressly described or shown herein, implement the principles of this disclosure and are included within its spirit and scope.

[0125] All examples and conditional language described herein are intended for educational purposes to help readers understand the principles of this disclosure and the inventors’ ideas for advancing contributions in the field, and are to be interpreted as not being limited to such specific examples and conditions.

[0126] Furthermore, all statements and specific examples of the principles, aspects, and embodiments described herein are intended to cover both their structural and functional equivalents. Additionally, such equivalents are intended to include both currently known equivalents and future-developed equivalents; that is, any element developed to perform the same function, regardless of its structure.

[0127] Therefore, for example, those skilled in the art will understand that the block diagrams presented herein represent conceptual views of illustrative circuits implementing the principles of this disclosure. Similarly, it should be understood that any flowchart, block diagram, etc., represents various processes that can be substantially represented in a computer-readable medium and thus executed by a computer or processor, whether or not such computer or processor is explicitly shown.

[0128] The functionality of the various components shown in the accompanying drawings can be provided using dedicated hardware and hardware capable of executing software in association with appropriate software. When provided by a processor, the functionality can be provided by a single dedicated processor, a single shared processor, or multiple separate processors, some of which may be shared.

[0129] It should be understood that the elements shown in the accompanying drawings can be implemented in various forms of hardware, software, or a combination thereof. Preferably, these elements are implemented in a combination of hardware and software on one or more appropriately programmed general-purpose devices, which may include processors, memory, and input / output interfaces.

[0130] This disclosure will be described with reference to a specific functional embodiment of a device 1 used to estimate the refractive error of a user's eye based on the visual acuity (VA) of that eye, such as... Figure 1 What is shown.

[0131] Although device 1 is described with reference to a user’s specific eye, it can be adapted to each of the user’s two eyes and / or to multiple users.

[0132] The refractive error of a user's eye (denoted as "Rx" for brevity) can be given by one or more values. For example, an eye with myopia or hyperopia but no astigmatism can have refractive error, represented by a single value usually expressed in diopters, which corresponds to the spherical power error (positive astigmatism for myopia, negative astigmatism for hyperopia, and the required correction is the opposite). In the case of further regular astigmatism, three values ​​may be appropriate, giving the spherical, cylindrical, and axial components respectively (the spherical and cylindrical components are usually expressed in diopters, and the axial component in degrees). In the case of higher-order aberrations, additional values ​​may be needed, such as coefficients of Zernike polynomials, as described, for example, in the article "Eye aberrations analysis with Zernike polynomials" by VVMolebny et al., Proceedings of the IEEE Optical Engineering Society, 3246, June 1998.

[0133] Device 1 is advantageously a device or a physical part of a device designed, configured, and / or adapted to perform the mentioned functions and produce the mentioned effects or results. In alternative embodiments, device 1 is implemented as a group of devices or physical parts of a device, whether grouped in the same machine or grouped in different, possibly remote, machines.

[0134] The device 1 is adapted to receive a measurement 21 of the visual acuity of the user's eye during a reference measurement period T0, referred to as the reference VA measurement.

[0135] In a favorable execution mode, the time period T0 is less than 10 minutes, allowing user and environmental conditions to remain stable during measurement. However, in some implementations, the time period T0 can be extended, for example, between 10 and 30 minutes, or between 30 minutes and 1 hour, to allow time between monitoring operations to check and adjust some measurement aspects before resuming the operation. This extended time may be useful in complex measurements (e.g., in the case of astigmatism) to prevent the user from overexerting themselves within a short timeframe. Alternatively or additionally, to improve accuracy, the extended time can be used to average the results, for example, to mitigate interference from uncontrolled user-related factors or external factors. In some embodiments, the time period T0 is extended to between 1 hour and 1 day, or even between 1 day and 1 week. In the latter case, it may be appropriate to perform similar measurements over approximately the same hourly period over two or more consecutive days to achieve similar corresponding measurement conditions and continue to appropriately average the results obtained over those days. In any case, the time period T0 is advantageously chosen to be short enough to avoid physiological evolution of refractive errors in the user's eye during that time period, which may depend on the specific characteristics of the user's eye.

[0136] In addition to the reference VA measurement 21, the device 1 is adapted to receive data 22 associated with at least two different blur levels respectively associated with the VA measurement 21 during the time period T0, and these data are applied to the user's eye to obtain the respective associated VA measurement 21. These data 22 (referred to as reference blur data) can take various forms, in particular including distance from the user's eye, azimuth angle of orientation test target, pupil size data, refractive power of added lens, and measured refractive error, as will be developed below.

[0137] Reference VA measurement 21 and reference fuzzy data 22 can be input by the user, streamed, transmitted to device 1 via a telecommunications network (such as a wireless LAN, cell network, or wired network), or retrieved from a database.

[0138] The device 1 is further adapted to obtain a general model 23, which may in particular be stored in one or more local or remote databases 10 and retrieved from the one or more local or remote databases as appropriate. The database 10 may take the form of storage resources available from any kind of suitable storage device, which may in particular be RAM or EEPROM (Electrically Erasable Programmable Read-Only Memory), such as flash memory, possibly within an SSD (Solid State Drive).

[0139] The general model 23 involves theoretical VA, theoretical fuzzy data, and theoretical refractive error. This general model is provided as the basis for generating information 25 related to the user's eye's refractive error—by mapping the input reference VA measurement 21 and reference fuzzy data 22 to the theoretical VA and theoretical fuzzy data, respectively, and by inferring one or more estimates of the user's eye's refractive error based on the theoretical refractive error. Accordingly, the reference VA measurement 21 and reference fuzzy data 22 are consistent with and prepared correspondingly to the general model 23 used. Moreover, although only one general model has been developed in this disclosure, two or more general models can be provided to device 1 and used sequentially or in parallel.

[0140] The general model 23 can be implicitly provided to device 1, rather than explicitly obtained through appropriately provided parameter inputs and related operations. Such operations can, for example, include addition or subtraction, multiplication or division, exponentiation with integer or rational exponents (i.e., with roots), trigonometric transformations, logarithms, extraction from lookup tables, inference by artificial intelligence (whether via expert systems or machine learning), etc., of any type, quantity, and combination. In some specific modes, operations can, for example, include single multiplication or division based on a single parameter, while in other modes, operations can involve complex combinations of rational functions, circular functions, and exponentiation functions applied to more than 3, 5, or even 7 unknowns (including parameters and components of the theoretical Rx component, but without calculating the theoretical VA).

[0141] In some implementations, such as Figure 2 The information 25 related to the refractive error of the user's eye, as shown in the device 1 denoted as 1A, corresponds to multiple pieces of information 25A related to the refractive error during the reference measurement period T0. In particular, information 25 may provide one or more estimates of the refractive error during the period T0.

[0142] In other embodiments, information 25 includes parameter values ​​that enable the specification or improvement of the general model 23 to make it more relevant to the user's eye, with the aim of providing an Rx estimate during time period T0 by another entity rather than device 1, or a future Rx estimate after the end of time period T0. These values ​​relate to all the partial parameters of the general model 23, and once fully specified, the model can be used, in particular, by device 1 or another device, to infer future Rx estimates based on later visual acuity measurements after time period T0.

[0143] This two-step process facilitates the follow-up of a user's eye physiological characteristics over time based on potentially accurate initial monitoring. That is, future Rx estimates can be performed at various time points after time period T0. Additionally, the general model 23 can be re-specified or improved after time period T0 to account for the evolution of the user's eye capabilities. This can be performed any number of times and may result in periodic model updates corresponding to successive reference measurement periods, between which any number of Rx estimates can be performed based on the most recent model update.

[0144] For example, the reference measurement period T0 is less than one hour, and subsequent Rx estimates based on parameter values ​​established during period T0 are performed weekly, bi-weekly, or monthly, depending on the type of eye condition (e.g., myopia, presbyopia, or astigmatism) and its rate of evolution of the follow-up user. The frequency of subsequent Rx estimates can be adapted to previously monitored evolution, with a higher frequency when evolution is faster and a lower frequency when evolution is slower. The reference measurement can be updated, for example, monthly, while subsequent Rx estimates are performed weekly.

[0145] In yet another embodiment, device 1 combines the prior capabilities and is therefore adapted to provide information related to the refractive error of the user's eye during the reference measurement period T0, and is also used to specify or improve the general model 23 for future estimations after period T0.

[0146] In some patterns ( Figure 1 In this embodiment, device 1 is configured to output not only information 25, but also inferred reference fuzzy data 22'. Then, by measuring the visual acuity of the user's eye corresponding to those inferred reference fuzzy data 22', and thus obtaining additional terms of the reference fuzzy data 22, the reference fuzzy data itself can be utilized within the reference fuzzy data 22. Therefore, a feedback loop is formed, wherein these operations can be performed only once or repeated any number of times. Thus, it is feasible to progressively and interactively improve the obtained results during time period T0 in a potentially iterative manner, involving successive VA measurements.

[0147] In this process, the reference fuzzy data 22 and the associated reference VA measurement 21 can be initially set to have a finite number of items (e.g., corresponding to a single fuzzy level) and are completed or gradually increased with the inferred reference fuzzy data 22' and the associated VA measurement. Alternatively, the reference fuzzy data 22 and the associated reference VA measurement 21 can be initially set as a whole and gradually adjusted based on the inferred reference fuzzy data 22'.

[0148] In a specific implementation, device 1 is configured to continuously average two or more values ​​of at least the same item in information 25 generated during time period T0, which may be an estimate of refractive error. This averaging allows for the consideration of potential differences, fluctuations, and errors related to the user and / or environment, with the aim of obtaining more reliable values. Therefore, successive VA measurements are performed based on the same reference fuzzy data 22 (or, if a feedback loop is used, the same initial reference fuzzy data 22), thereby generating corresponding information items 25. For example, those measurements and information generation are performed consecutively within less than 30 minutes, and / or, if time period T0 is several days long, are performed within approximately the same hourly intervals over consecutive days. Averaging can include, but is not limited to, arithmetic mean, geometric mean, harmonic mean, median, or quadratic mean (i.e., root mean square). Furthermore, the values ​​to be averaged can be weighted, for example, to give the first measurement a greater weight compared to subsequent measurements with a correction function.

[0149] Device 1 interacts with user interface 16, through which a user can input and retrieve information. User interface 16 includes any device suitable for inputting or retrieving data, information, or instructions, particularly visual, tactile, and / or audio capabilities. These devices may encompass any one or more of the following devices well known to those skilled in the art: screen, keyboard, trackball, touchpad, touchscreen, speaker, and voice recognition system. Interaction with user interface 16 can be indirect, for example, where device 1 takes the form of a hardware component, such as all or part of an integrated circuit embedded in a device.

[0150] In the following text, modules should be understood as functional entities, rather than physically distinct material components. Therefore, they can be implemented as combined within the same tangible component, or distributed among several such components. Furthermore, each of these modules may be shared between at least two physical components. Additionally, these modules can be implemented in hardware, software, firmware, or any hybrid form thereof. Modules belonging to a device are preferably implemented within at least one processor of the device.

[0151] The device 1 includes an input module 11 adapted to receive a reference VA measurement 21 and reference fuzzy data 22. As described above, these can be derived from the inferred reference fuzzy data 22'. The device 1 also includes an output module 15 adapted to provide information 25 and, if relevant, the inferred reference fuzzy data 22', depending on the implementation.

[0152] Furthermore, the task of the embodiment module 12 of device 1 is to embody the general model 23 of the user's eye into a specific model 24 suitable for generating information 25 (and possibly inferred reference fuzzy data 22'). The specific model 24 can be considered as specifying or improving the results of the general model 23 fed by the reference VA measurement 21 and the reference fuzzy data 22.

[0153] In some implementations, a particular model 24 is fully set by the embodiment module 12 during time period T0, possibly after one or more iterations involving the inferred reference fuzzy data 22' and / or averaging steps. The fully set model 24 can then be used to generate estimates of the user's eye refractive error during and / or after time period T0. The particular model is fully set as long as the parameters of the particular model 24 have known values ​​that allow for the estimation of refractive error based on VA measurements associated with the fuzzy data.

[0154] While fully usable for estimation, a complete specific model24 can be updated over time to take into account the physiological evolution of the user's eye.

[0155] Furthermore, the specific model 24 can be intentionally loosened to allow adjustment of one or more parameters, thus enabling their use as variables. For example, the general model depends on the distance to the user's eyes, the azimuth angle relative to the orientation test, and / or the pupil diameter. While the latter have specified values ​​in the full specific model 24, at least some of them are flexible to reflect valid later measurement conditions.

[0156] In addition, a partial implementation of the specific model 24 can be performed by another entity instead of device 1 during the reference time period T0.

[0157] Downstream of the materialization module 12, the information generation module 13 is responsible for generating information 25, and the fuzzy data determination module 14 (if provided) is responsible for generating inferred reference fuzzy data 22'. Although presented as separate functional entities, modules 12, 13, and 14 can in fact involve the same set of operations applied to the reference VA measurement 21 and the reference fuzzy data 22 to generate information 25 (and possibly inferred reference fuzzy data 22').

[0158] In an implementation where information 25 includes parameter value 25B, such as Figure 3 As illustrated by device 1, denoted as 1B, another device 3 is responsible for inferring a later Rx estimate 28 based on a general model 23 and those parameter values ​​25B, according to a later visual acuity measurement 26 and associated later blur data 27. In a particular embodiment, device 1B is also configured to provide an Rx estimate 25A during time period T0, further providing parameter values ​​25B.

[0159] The parameter value 25B can be obtained or retrieved by the device 3 in any way, such as by extracting the parameter value from a storage source (e.g., a database 10 that previously recorded parameter values ​​from the device 1), or by receiving the parameter value from the device 1 on demand.

[0160] Later fuzzy data 27 is adapted to a particular model 24 and enables the direct inference of Rx estimates from the associated later VA measurements 26.

[0161] Device 3 includes an input module 31 for receiving later VA measurements 26 and later ambiguity data 27, and an output module 35 for providing a later estimate 28 of the user's eye's refractive error. It further incorporates a module 32 responsible for generating a later Rx estimate 28 based on the later VA measurements 26 and later ambiguity data 27, using a specific model 24 derived from parameter value 25B and a general model 23. Device 1B may be particularly interested in obtaining an accurate version of the specific model 24, for example, based on the substantial computing power of specialized equipment and / or workstations in ECP-managed medical settings. Device 3 may simply be a mobile device enabling lay personnel to obtain a relatively high level of accuracy in refractive error estimates based on simple operations dependent on the available specific model 24.

[0162] In an alternative embodiment where a specific model 24 is fully configured, the device 3 is not provided with later fuzzy data 27, but the VA measurement conditions conform to reference fuzzy data 22 associated with the model 24 (e.g., distance, orientation angle, pupil diameter).

[0163] In denoted as 1C and Figure 4 In another embodiment of the apparatus 1 shown above, wherein the information 25 includes parameter values ​​25B for determining a particular model 24 during time period T0, the apparatus 1C is adapted to continue generating parameter values ​​25B for determining a particular model 24 during time period T0, and subsequently using those parameter values ​​25B to generate a later Rx estimate 28 based on a later visual acuity measurement 26 and associated later fuzzy data 27.

[0164] Device 1C can be implemented, for example, in a mobile device that potentially provides the independent capability of relatively accurate Rx estimation based on time-varying, more in-depth measurements using simple VA measurements.

[0165] In operation, whether in combination with devices 1B and 3 or with device 1C (such as...) Figure 5 The combination of the above-mentioned data (as seen above) generates parameter values ​​25B and possible Rx estimates 25A based on reference VA measurements 21 and reference fuzzy data 22 during the reference measurement period T0. When deemed appropriate, an estimate 28-1 of the later refractive error is then inferred during a first later period T1 based on later visual acuity measurements 26-1 and associated later fuzzy data 27-1. A similar process is repeated during the subsequent period T2, thereby inferring the later Rx estimate 28-2 based on later visual acuity measurements 26-2 and associated later fuzzy data 27-2.

[0166] VA measurements can be based on any of a variety of optometric fonts, as are well known to those skilled in the art. Depending on the implementation, they may, for example, use Snellen charts, log-EMDRS (LogMAR-ETDRS) charts, Landau C charts, rolling E charts, or Golovin-Sivtsev tables. In some models, the optometric fonts utilized include at least one of circular, square, triangular, and vanishing optometric fonts with circles and / or lines, and gray and / or checkerboard backgrounds.

[0167] Visual acuity (VA) measurements can rely on optical instruments. They can also rely on computerized tests that enable self-managed VA measurements. In the latter case, VA can be evaluated using adaptive laddering methods, such as the optimal PEST procedure (meaning "optimal parameter estimation by sequential testing"), and more specifically, the Friiburg Visual Acuity and Contrast Test (meaning "Friburg Visual Acuity and Contrast Test"), as described by M. Bach in "The Friiburg Visual Acuity Test - Automatic Measurement of Visual Acuity," Optometry and Vision Science, Vol. 73, pp. 49-53, 1996.

[0168] In a specific mode, the VA test involves roughly determining the user's eye's VA using a scale, followed by the use of a shorter step target.

[0169] In some patterns that can be combined with previous patterns, the visual acuity (VA) is determined at a level more accurate than the row levels in the visual acuity chart (where a binary test is performed on each row) by considering the percentage of good answers at a given level (e.g., 60%). For example, a VA of 1.25 corresponds to the ability to give 60% good answers in an optometric font, where details are seen at an angle at 1 / 1.25 arcminute.

[0170] In some implementations, VA is measured by a parameterized probabilistic model and artificial intelligence that relies on the VA response, as described, for example, in “The Stanford Acuity Test: A Precise Vision Test Using Bayesian Techniques and a Discovery in Human Visual Response” by C. Piech et al., AAAI Artificial Intelligence Conference Proceedings, 34(01), pp. 471-479, 2020.

[0171] Since two or more VA measurements (corresponding to different fuzziness levels) are performed in association with the corresponding reference fuzzy data 22 when a specific model 24 is set by device 1, VA testing is selected to achieve a good balance between test duration and accuracy.

[0172] For devices 1B or 1C, the type of VA test performed during the reference period T0 for setting a specific model 24 (or a subsequent period for updating or refining the specific model 24) may differ from the type of VA test performed by devices 1C or 3 after period T0 for inferring the Rx estimate 28. For example, the VA test during T0 may be less complex than during a later period to reduce the time required for the user to perform multiple VA evaluations corresponding to different fuzziness levels. Conversely, the VA test during T0 may be more complex than during a later period to accurately set the parameter value 25B.

[0173] The wide range of operational capabilities of device 1 will be illustrated below in various embodiments.

[0174] Example 1 - Estimation of refractive error based on VA at multiple distances

[0175] In the current mode corresponding to a specific version of device 1A, VA is estimated at two or more distances from the user's eyes (thus constructing a reference VA measurement 21 associated with the selected distance as reference fuzzy data 22), a separate model is constructed (constructing a complete specific model 24), and an Rx estimate 25A is derived based on the collected data.

[0176] Distance monitoring and VA testing can be achieved by any available method. They can be performed entirely by means of a mobile device (including a front-facing camera and a screen) and a reflector, as described in patent application WO 2022 / 013410 granted to Essilor International, cited above. According to this implementation, the user positions himself at a distance d / 2 in front of the reflector, the mobile device is positioned such that the front-facing camera faces the reflector, is as close as possible to the user's head and covers the eyes of the user in question, the distance d between the front-facing camera and the virtual image of the mobile device in the reflector is measured, the screen of the mobile device displays optometric font of appropriate shape, position, and size, and the VA is evaluated. As developed in that document, the distance d can be measured by displaying one or more elements of known size (e.g., optometric font) on the screen of the mobile device and by estimating the distance d based on the angular resolution of the front-facing camera.

[0177] Alternatively, distance monitoring can be performed by a time-of-flight camera (e.g., an infrared camera) available in the mobile device. Monitoring can be obtained by a 3D sensor involving depth mapping and may utilize images from a front-facing camera in combination.

[0178] In this variant, distance monitoring of the mobile device is based on accelerometers and gyroscopes, for example, by utilizing an IMU (Inertial Measurement Unit) integrated into the device. Appropriate initial calibration is then expected, which can be accomplished, for example, by adapting a distant object of known size (e.g., a card) viewed on the screen to a specific size corresponding to a predefined distance (e.g., 1 meter).

[0179] In other distance monitoring implementations where user location is assessed by a remote device, distance is estimated by utilizing visible physical characteristics of the user. For example, after appropriate preliminary calibration steps, one or both of the user's eye width or interpupillary distance can be used as a reference unit of measurement.

[0180] In some modes that combine two types of VA monitoring performed by a mobile device, VA testing is conducted directly with the mobile device at distances less than 50cm from the user's eyes, and for distances greater than 50cm, a reflector as described above is used. In an advantageous embodiment, this allows for measurements between 0.2 meters and 4 meters with an accuracy of a few percent.

[0181] In the alternative execution mode, the VA test is performed on the user's eyes with the support of another person, without the use of a mirror. This other person could, for example, hold the mobile device at an appropriate distance from the user and display appropriate optometry text on a screen.

[0182] In other modes that omit the reflector and another person, the mobile device, which can display optometry text on the screen, is equipped with remote control capabilities (e.g., based on voice or gesture recognition) and recording capabilities. The user can then position the mobile phone appropriately, ideally vertically, and stand in front of the screen at a desired distance, while giving instructions, such as requesting an indication of the distance from the user's eyes, displaying the selected optometry text (shape, size, orientation, etc.), and considering and recording user feedback associated with the corresponding test items.

[0183] In the absence of astigmatism, or if astigmatism can be ignored, the refractive error of the user's eye can be represented by spherical defocus S (expressed in diopters, as required for correction). This requires defocus blur δ, which can be adjusted according to accommodation A (expressed in diopters) (taken as a value between 0 and A). max The values ​​between the two distances, the distance from the user's eye Dist (positive, expressed in meters), and S are formulated as follows:

[0184] δ=|S-A+(1 / Dist)| (1)

[0185] Adjustment A depends on the spherical defocus S and the distance Dist. Note:

[0186] S' = S + 1 / Dist (2)

[0187] get:

[0188] δ=|S'-A| (3)

[0189]

[0190] In other words, the value of A is its maximum value A below the nearest point (PP). max It decreases linearly to 0 as the distance between PP and the far point (PR) increases, and remains constant as a null value beyond PR. Furthermore, the values ​​corresponding to the distances to PR and PP are -1 / S and -1 / (SA), respectively. max ).

[0191] Visual acuity VA can be represented as a function of defocus blur δ, that is, a function of spherical defocus S, distance Dist, and accommodation A, and therefore S, Dist, and A are used to express this. max The functions are used to represent (equations (2), (3), and (4)).

[0192] This is Figure 6 The above demonstration addresses a specific case of myopia, where the user's eye has a defocus error of -2D (spherical astigmatism S in myopia) and a maximum accommodation of 4D. max The theoretical decimal VA (in arcmin) is given. -1 The curve 41, which is a function of distance Dist (in meters, axis 411) with axis 412 as the unit, has three regions: the distance viewing region A1-1 beyond PR, denoted as PR1, where A is null or assumed to be null (it can be stimulated by other factors); the accommodation region A1-2 between PR1 and PP, denoted as PP1, where the defocus blur δ should be null; and the near viewing region A1-3 from the user's eye to PP1, where the value of A is A max In the exemplary case considered, the values ​​of PR1 and PP1 are 0.5m (i.e., -1 / S) and 0.17m (i.e., -1 / (SA)), respectively. max From the maximum indicated distance to the user's eye: in the distance viewing area A1-1, as Dist changes from 4m to 0.5m, VA increases as defocus decreases; in the accommodation area A1-2, VA reaches its maximum value. max (And correspondingly, MAR reaches its minimum value MAR) min ), while adjusting A from 0 to A max Below 0.17m, VA decreases because adjusting A cannot compensate for the defocus caused by short distances.

[0193] General model 23 can be based on the previously cited article by Gómez-Pedrero. In this respect, VA can be expressed as:

[0194]

[0195] E1 and E2 are parameters that depend on the pupil diameter D of the user's eye and are specific to each subject.

[0196] Parameter E1 can be expressed as VA by the following formula. max Functions:

[0197] E1 = VA max –q (6)

[0198] This is derived from equation (5) regarding the adjustment region A1-2, where the defocus blur δ is null and the value of VA is constant. max .

[0199] Therefore, for the viewing distance region A1-1 where adjustment A is null:

[0200] VA = [VA max –q +E2|S+1 / Dist| q ] -1 / q When Dist ≥ PR1 (7)

[0201] This can be expressed using the MAR formula:

[0202] MAR = [MAR min q +E2|S+1 / Dist| q ] 1 / q When Dist≥PR1 (8)

[0203] MAR = MAR min When PP1 <Dist<PR1 (9)

[0204] In this theoretical model, the same formulas (7) and (8) also hold true in the near vision region A1-3, where S is replaced by (SA). max However, in terms of myopia, device 1A can focus on operation in the distance vision region A1-1 and the accommodation region A1-2.

[0205] Therefore, the general model 23 depends on five unknown variables: spherical defocus S (in diopters), Dist (in meters), and VA. max (in arcmin) -1 (in units), q (unitless) and E2 (in meters) q.arcmin q (Unit: )

[0206] As developed in Gómez-Pedrero's paper, the parameter q corresponds to the defocus tolerance. It typically lies between 1 and 3, and generally between 1.8 and 2.

[0207] The function of parameter q is, for example, in Figure 7 The exemplary curve 42 is shown, on which MAR (in arc minutes, axis 4202) is expressed as 1 / Dist (in meters, the reciprocal of the distance in meters). -1 Using units of q and axis 4201, for -2D myopia and different q values, the MAR values ​​of curves 421, 422, and 423 are 1.5, 2, and 2.5, respectively. It is evident that for these three curves 421, 422, and 423, the corresponding values ​​for 2m are obtained. -1 The same PR (denoted as PR2) and corresponding to approximately 4.5m -1 The same PP (denoted as PP2), these three curves are constantly at MAR in the adjustment region A2-2 between PR2 and PP2. min Horizontal. However, in the distance vision region A2-1 beyond PR2 and the near vision region A2-3 below PP2, the slope of the MAR change increases as the q value decreases.

[0208] In this embodiment, device 1A is configured to evaluate spherical defocus S and other desired parameters based on VA measurements at different appropriately selected distances. For simplicity, parameter q can be approximated as a fixed value, such as 2. Then, based on equation (7), S, E2, and VA can be calculated from the three pairs of distances Dist and the associated measurement VA. max .

[0209] In myopia-related patterns, one pair (Dist, VA) corresponds to the flat accommodation region, and the other two pairs correspond to the distance vision region. For example, as... Figure 6 As shown, points P1-1 and P1-2 are located in region A1-1, while point P1-3 is located in region A1-2. Device 1A can then extract VA from point P1-3. max And from this, S and E2 can be inferred from P1-1 and P1-2.

[0210] The maximum VA in the adjustment region A1-2 can be directly measured, so that S and E2 can be obtained only from equation (7), which can be done via P1-1 and P1-2.

[0211] In the variant, three pairs (Dist,VA) are selected in the distance vision region A1-1.

[0212] In any case, the distance Dist used to specify the general model 23 via equation (7) deserves special attention. In fact, the accuracy of the obtained estimate depends on the accuracy of each pair of (Dist,VA), but also on their positioning on the VA curve.

[0213] When evaluating the three unknown parameters of equation (7), more than three distances (Dist), such as four, can be used. While these operations are redundant, they can improve the accuracy and reliability of the results.

[0214] In the variant, device 1A is configured to further determine, in addition to VA, based on at least four pairs (Dist, VA). max The parameter q is in addition to S and E2. Although the requirements are higher, this implementation can improve the accuracy obtained.

[0215] In a specific implementation that facilitates the selection of the computational distance Dist, device 1A iteratively performs a backward mechanism, enabling the obtained results to be utilized for further distance selection (as inferred reference fuzzy data 22') by re-injecting the obtained results.

[0216] This implementation can be initialized by a rough estimate of the refractive error. For example, the initial estimate can be based on the Swaine rule, according to which the spherical astigmatism of a myopic eye can be evaluated in a certain VA range (typically between 1 / 10 and 5 / 10) by dividing 0.25 by the distance VA (expressed in decimals).

[0217] The following is an illustrative estimation scheme for low to moderate myopia (-2D to -0.5D), where q is set to 2, for example, such that, in the case of distance vision, according to equation (8):

[0218] MAR 2 =MAR min 2 +E2|S+1 / Dist| 2 (10)

[0219] In the initial steps, MAR min E2 is initialized to the corresponding value, MAR of 1 arcminute. min1 and 1.5m 2 .arcmin 2 E 2-1 (For a 3mm pupil), and the initial distance of vision Dist1 is chosen to be, for example, 4m. At distance Dist1, VA is measured to be equal to VA1 (which can correspond to...). Figure 6 P1-1), thus obtaining the first approximation S of the spherical defocus S. est1 :

[0220]

[0221] Estimated spherical defocus S est1 This also allows us to obtain the first estimate D of the PR distance. estPR1 , equal to -1 / S est1 Other starting points can be used instead of the current starting point, and these starting points can be tailored to the specific eye condition of a user, such as the level of myopia.

[0222] In the second step, at D est1 Choose a second distance, Dist2, between Dist1 and Dist2. For example, it is given by the arithmetic mean of the inverse of the distance:

[0223] 1 / Dist2 = 0.5(1 / D estPR1 +1 / Dist1) (12) This is equivalent to:

[0224] Dist2 = 2Dist1 / (1-S) est1 .Dist1) (13)

[0225] Using this second position, visual acuity VA2 can be measured (which can correspond to P1-2). This pair (Dist2,VA2), together with the previously obtained pair (Dist1,VA1), enables a better evaluation of S, representing spherical defocus S and parameter E2, respectively, according to equation (10). est2 and E 2-2 According to S est2 The enhanced estimate D of the PR distance can be inferred. estPR2 .

[0226] In the third step, a third position, Dist3 (sufficiently lower than D), is selected within the estimated adjustment region. estPR2 This can correspond to P1-3) so that VA can be directly measured respectively. max Or MAR min =1 / VA max The estimated value VA max3 Or MAR min3 Once MAR has been evaluated min Then the previously obtained pairs (Dist1,VA1) and (Dist2,VA2) can be introduced into equation (10), such as using MAR min3 The values ​​are specified so that the enhanced evaluation S of spherical defocus S and parameter E2 can be derived respectively. est3 and E 2-3 The enhancement value D of PR distance estPR3 It can also be derived that it equals -1 / S est3 .

[0227] The process can stop at this third step, with S est3 The resulting estimate is the spherical defocus S (i.e., Rx estimate 25A).

[0228] However, this process can alternatively continue with a fourth step, which is adapted to obtain an enhanced estimate of parameter q instead of 2. For example, this can be done as follows: in D estPR3 With D estPR2 Choose the fourth point (Dist4,VA4) and use equation (8) and three pairs (Dist1,VA1), (Dist2,VA2) and (Dist4,VA4) to extract the values ​​of the three unknowns S, E2 and q respectively. est4 E 2-4 And q4. This extraction can be performed, in particular, by numerical optimization via iterative methods, as is known to those skilled in the art, for example, by gradient descent applied to the cost function. The process may further include examining the inferred value D of the PR distance. estPR4 =-1 / S est4 Whether it is higher than Dist3, used to measure VA max .

[0229] In the variant, the estimated maximum VA value can be checked. max3 Is it located in the accommodation zone rather than the near vision zone? In this regard, a distance value lower than but sufficiently close to Dist3 can be selected, such as 0.90Dist3, and the relevant VA measurement can be verified to provide a close VA value. max3 The value is used to confirm that the location is in a flat area corresponding to the adjustment area.

[0230] In the improved version of the general model 23, the defocus blur δ is released in the accommodation region A1-2 instead of being set to a constant null value to represent accommodation lag. This allows the user's eye's accommodation deficit to be integrated relative to the requested stimulus, which is known to occur even if the stimulus is below the accommodation capacity, thus slightly increasing the VA loss.

[0231] In a relevant exemplary implementation, a “hysteresis term” inversely proportional to the distance Dist is added to the defocus blur δ (which may be inserted into Equation (3) for example, in addition to the absolute value). This term relates to a multiplication factor AG corresponding to the adjustment gain, for example, set to be between 0 and 1.

[0232] In another implementation, a constant “dark focus” term dkF is added to the defocus blur δ (which can be inserted into Equation (3) for example, in addition to the absolute value), which reflects the residual modulation in the absence of any stimulus (i.e. in darkness).

[0233] In a more improved implementation, the general model 23 considers both aspects mentioned above and thus involves the hysteresis term AG / Dist and the dark focus term dkF. Furthermore, considering that the adjustment gain AG is between 0 and 1, the dark focus term dkF is weighted by a multiplication factor (1-AG), thus representing the neutral point without hysteresis, denoted as (1-AG)dkF. Therefore, if AG is 0, there is no hysteresis term, and the dark focus term is added accordingly (the pure dark focus term implementation above). In contrast, if AG is 1, there is no dark focus term, and the hysteresis term equals 1 / Dist (the maximum hysteresis representation). By choosing a value for the adjustment gain AG that is between 0 and 1 and sufficiently far from 0 and 1, the general model 23 can be more realistic, generally resulting in the addition of the term: AG / Dist + (1-AG)dkF. In a favorable mode, the adjustment gain AG and / or the dark focus term dkF are treated as additional parameters that can be estimated a priori and / or derived from appropriate measurements and calculations in a manner similar to other parameters.

[0234] The above example is for myopia and focuses on the distance vision and accommodation areas. It will be obvious to the technician that a similar operation can also be performed in the near vision area, which may be particularly relevant to eye conditions such as hyperopia or presbyopia.

[0235] Specifically, each pair (Dist,VA) can be selected in the near vision region and the accommodation region, rather than in the far vision region and the accommodation region.

[0236] In a variant, device 1A is configured to select each pair (Dist,VA) by jointly using at least one point in the far-viewing region and at least one point in the near-viewing region to infer unknown parameters according to equation (7).

[0237] Furthermore, device 1A can be adapted to perform two or more of the above embodiments in any possible combination.

[0238] Parameter initialization and / or partial parameter setting during the iteration process (which allows for the upstream setting of one or more model parameters) can be based on previously estimated or monitored values. In this regard, a preliminary version of a specific model 24 (partial or complete) linking VA to refractive error can be constructed, in particular, based on machine learning and / or data collection. This preliminary model can be fed with personalized parameters, including at least one of age, refractive error, sex, and ethnicity. The preliminary model can be further or alternatively adjusted based on the user's group as retrieved from database information. This group can, for example, be described as a subgroup of refractive error sufferers with similar current refractive error and / or similar current visual acuity, or a subgroup of presbyopia sufferers with similar age and / or similar corrected refractive error. The preliminary model can be further or alternatively based on and / or improved upon data collected about a particular individual user under consideration, thereby guiding and / or improving predictions about that user.

[0239] Additionally, parameter initialization can be based, at least in part, on values ​​from prescriptions held by users who already wear glasses. It is worth noting that the VA under distance vision can be VA. max Provide a good starting point.

[0240] Users who wear glasses can further maintain their glasses during the measurement, allowing the refractive error shift to be assessed instead of the full value. This may be relevant for myopia if the residual defect is sufficiently large (typically above 0.25D at 4m). However, for presbyopia, even smaller values ​​may be suitable for difference estimation.

[0241] Example 2 - Based on one or more model parameters of VA at multiple distances

[0242] In the current mode, instead of providing Rx estimates 25A, device 1 (device 1B or 1C) is configured to provide parameter values ​​25B that partially or completely determine a particular model 24 during the reference period T0. This can then be used to derive updated estimates of the refractive error during a later period.

[0243] The specific model 24 established can be complete, in which case the specific values ​​of distance Dist and pupil diameter D are particularly fixed (e.g., 6 meters and 3 mm, respectively), or it can be partial, leaving room for at least some of the parameters involved (e.g., Dist and D, therefore VA). max And E2, but not q).

[0244] For example, returning to the case of Example 1, not only is the spherical defocus S, but also VA maxq and E2 are all evaluated together with S. In a specific implementation, at least one of the evaluated parameter values ​​is retained for later estimation of the refractive error. This can then be reused in later periods without recalculation, thus reducing associated operations. For example, over time, q and the possible E2 may be considered more significant than S and VA. max It is significantly more stable and therefore remains unchanged for a period of time significantly longer than T0. This duration may further depend on the nature of the parameters under consideration, for example, it is different for q and E2.

[0245] Distance monitoring and VA assessment can be performed using a mobile device as mentioned above in Example 1, to set a specific model 24 via parameter value 25B during time period T0, and to utilize those values ​​25B in the later Rx estimation.

[0246] In other embodiments, particularly when parameter value 25B (excluding distance Dist) is assessed by the ECP in a dedicated location during time period T0, the distance is measured, for example, by a dedicated remote monitoring device, or is available directly on the floor or wall in the form of a scale bar. This is equivalent to an upstream calibration. Subsequent assessments of the refractive error based on those parameter values ​​25B can then be performed using only a mobile device for distance monitoring and VA assessment as described above.

[0247] In some embodiments designed to evaluate parameter value 25B, the ECP provides the user with a test frame lens with the target correction and also provides him / her with an application suitable for performing the first VA test in calibration mode (e.g., on a mobile phone). In other embodiments, the user is provided with a fitted test lens for such evaluation, which may be performed under the remote supervision of the ECP.

[0248] In other embodiments, the user's eye refractive error is measured during a reference time period T0 and provided as input to device 1B or 1C (thereby becoming part of the reference fuzzy data 22). The device is accordingly configured to provide parameter value 25B, where the refractive error is known.

[0249] For example, during time period T0, a triplet including distance Dist, VA, and spherical defocus S is obtained from measurements in the far-view region. VA can then be estimated using equation (7). max The unknowns are E2 and q. If, for example, q is set to 2, then only VA needs to be evaluated. max And E2, which makes it possible that measurements at two distances in the field of view suffice. If q is also the evaluation target, then measurements at three or more distances are taken. The feedback loop mechanism described in Example 1 can be used.

[0250] Once all desired parameters have been assessed during the reference period T0, they can be relied upon to estimate refractive errors at later stages. For example, q and E2 remain constant over a pre-specified duration (e.g., between 1 and 3 months) that is significantly longer than T0 (e.g., 1 hour), during which VA only needs to be estimated along with S. max This reduces the number of unknowns from four to two, which are solved based on two or more pairs (Dist,VA) obtained from later VA measurements 26 and later fuzzy data 27. This operation can be repeated periodically for a preset duration (e.g., weekly).

[0251] Alternatively or in combination, the estimated VA during the reference period T0 max The value remains constant over a duration longer than the reference period T0 but shorter than the preset duration used for E2 and q. For example, if T0 is 1 hour and weekly follow-up measurements are performed, VA max It remains unchanged over 1 month, and q and E2 remain unchanged over 3 months.

[0252] In this variant, the adjustment region is periodically updated based on estimates made after time period T0. Therefore, when the latest batch of measurements is performed, the adjustment region can be considered sufficiently approximated to allow for direct measurement of one or more VA values ​​at relevant distances within it, thereby deriving the VA. max Therefore, the overall required measurements and / or calculations can be significantly reduced.

[0253] Similar to Example 1, while measurements in the distance vision region are particularly suitable for myopia, measurements can also be performed in the near vision region, especially in cases of other eye conditions, such as hyperopia or presbyopia.

[0254] Example 3 - Based on model parameters of a VA with one or more power corrections

[0255] In the current mode, similarly, device 1 (device 1B or 1C) is configured to provide parameter values ​​25B that partially or completely determine a particular model 24 during the reference time period T0. This can then be used to derive an updated estimate of the refractive error during a later time period.

[0256] However, instead of obtaining VA measurements at different distances, measurements can be performed for different optical power offsets, and 0 optical power offset (i.e., no optical power correction) may be included in the measurement.

[0257] For example, the general model 23 is represented by a simplified formula:

[0258] S=k*VA+l (14)

[0259] Where VA is the visual acuity measured at a predetermined distance from the user's eye (e.g., 6 meters or 20 feet), and k and l are parameters. According to formula (14), the resulting differential evolution model is given by:

[0260] ΔS=k*ΔVA (15)

[0261] The "Δ" represents the change over time and is applied to spherical defocus S or visual acuity VA.

[0262] In a specific implementation, the evolution of spherical defocus S is equivalent to the change in Rx applied by optical power correction during the reference period T0, and the resulting change in VA during T0 is equivalent to the change in VA over time. In this regard, the lens is positioned in front of the user's eye to induce the desired optical power correction corresponding to the change in S, DiffS1, for example, +0.25D or +0.5D for myopia. The corresponding change in VA, DiffA1, is then measured. According to equation (15), the parameter k can therefore be estimated as the ratio DiffS1 / DiffA1.

[0263] After the reference period T0, device 1C or 3 is configured to infer the change ΔS based on the measured change VA ΔVA and formula (15), and thus infer an estimate of the refractive error S.

[0264] Although this estimation only provides a rough approximation of Rx evolution, it can offer a first assessment of interest over a considerable period following the reference period T0. In particular, it can enable systems or individuals to identify urgent situations requiring recommendations (e.g., scheduling an appointment with an ECP). This can be especially useful in the follow-up of myopia progression.

[0265] In fact, as long as there is no substantial evolution of Rx (which can be quantified by a relative threshold (e.g., a 10% or 20% change)), the simplified general model 23 expressed by formula (15) can be considered satisfactory.

[0266] This approximation can also be used to initialize iterative processes that conform to the feedback methods described in Example 1 or Example 2 in order to obtain more accurate and relevant estimates.

[0267] In addition, refractive error and VA measurements can be performed by ECP using specialized equipment to provide an accurate estimate of the k-ratio parameter. After the reference period T0, refractive error can likely be derived directly from VA measurements using simple mobile devices.

[0268] While using a single corrective lens may be sufficient to specify the model associated with equation (14) (based on VA measurements with and without the lens, respectively), in advantageous embodiments, two or more lenses, such as 0.25D and 0.5D, are used. In this way, parameter k can be estimated with better accuracy, for example, by linear regression (because it depends on three or more points rather than two).

[0269] In those implementations, such calibration measurements can be performed with particular care in similar or identical environments at very close times in order to reduce the adverse effects of daytime or changes in sensitivity to light conditions.

[0270] Although refractive errors have been represented above by spherical defocus S, alternative or additional Rx components can be handled in a similar manner.

[0271] Moreover, more complex models than linear models can be used, such as those conforming to equation (7), by applying an appropriate number of optical power corrections to determine the unknowns.

[0272] This implementation can be combined with consideration of multiple values ​​for distance Dist as disclosed in Examples 1 and 2.

[0273] The use of one or more corrective lenses may also be particularly attractive in measuring conditions other than myopia. Notably, in a specific embodiment, device 1 is used to extract information 25 (either the estimate itself 25A or model parameters 25B) related to refractive errors in hyperopic eyes. This may be particularly relevant for users under 35 years of age, for whom presbyopia is generally not the primary cause of short-distance blur. In the relevant mode, at least a portion of the reference VA measurement 21 is performed outside the PP while the corrective lens is applied to the user's eye. This method allows for the simulation of myopic behavior, although the distance and maximum VA typically associated with corrected visual acuity are taken into account, the latter being intentionally degraded by artificially altered optical power. This is equivalent to generating a PR and a distance region beyond the PR (which does not previously exist for hyperopic eyes), except that it brings the PP closer to the user's eye and accordingly modifies (i.e., makes it closer and smaller) the near vision and accommodation regions.

[0274] In a specific implementation, as developed in Examples 1 and 2, VA is measured at multiple distances and information 25 is inferred. In particular, a general model 23 of the type described above can be used, relating to formula (5). Moreover, at least one of the selected points is located in the accommodation region in association with one or more converging lenses respectively applied to the user's eye.

[0275] For example, such as Figure 8 What is shown represents that in S = +3D and A max=6D under specific farsighted conditions, VA (axis 432, in arcminutes reciprocal) -1 (in m units) as -1 / Dist -1 The evolution curve of the function (in units, axis 431) is 43, with the value of the near point PP3 being approximately 33cm (-1 / Dist is SA). max =-3m -1 Furthermore, the near vision zone A3-3 is separated from the accommodation zone A3-2. As the user's eye moves away from PP3, the user's visual acuity (VA) periodically improves, whereby, due to proper accommodation, VA reaches its maximum level. max This maximum level is maintained when PP3 is exceeded. More precisely, taking into account the distance Dist, A is adjusted to fully compensate for the refractive error S (A = S + 1 / Dist).

[0276] For comparison, A max A value of, for example, 8D for an emmetropic eye (without refractive error, therefore S=0) results in a curve shape similar to the aforementioned curve shape, but PP is located at 12.5cm (the value of -1 / Dist is -A). max =-8m -1 Furthermore, there was no refractive error (S=0) and A max For example, a value of 3D presbyopia also results in a shape similar to the one described above, but PP is located at approximately 33cm (-1 / Dist has a value of -A). max =-3m -1 It is worth noting that the evolution of VA is therefore difficult to distinguish from curve 43, making presbyopia and hyperopia actually have quite similar effects on the user's eye VA, but they rely on different mechanisms: hyperopia is due to insufficient refractive power of the user's eye, while presbyopia is due to insufficient accommodative ability.

[0277] For example, consider three points at corresponding distances from the user's eye: P3-2 and P3-3 in the near vision region A3-3, and P3-1 in the accommodation region A3-2. Point P3-1 is at a significantly greater distance from the user's eye compared to P3 and undergoes two different measurements: one without additional lenses (naked eye or with conventional correction, depending on the mode of execution), and the other with an applied focal power DiffS3, which is, for example, between +0.25D and +4D, or more precisely, between +0.5D and +2D. Multiple lenses with correspondingly different optical powers can then be successively applied to the user's eye. The P3-1 measurement without additional lenses can be used to directly extract VA. max On the other hand, the P3-1 measurement involving additional lenses can ensure that measurements are provided in the distance vision area beyond the accommodation area in the case of hyperopia, and distinguish hyperopia from emmetropia and presbyopia.

[0278] While P3-2 and P3-3 measurements can provide useful VA information, in variations, only one or neither is used, thus reducing testing operations by focusing on the accommodation region A3-2. In other embodiments that can be combined with the previous embodiments, the P3-1 measurement is performed with only one or more lenses, or conversely, only without lenses. In the latter case, hyperopia can be distinguished from presbyopia based on the user's age.

[0279] Example 4 - Estimation of refractive error based on VA at one or more directional angles

[0280] This embodiment is particularly suitable for considering astigmatism. Therefore, the refractive error of the user's eye can be represented by spherical defocus S, axial component α (along which the refractive power is S), and cylindrical defocus C. Furthermore, the equivalent spherical power H of the user's eye can be given by the following formula:

[0281] H = S + C / 2 (16)

[0282] The following section will use the negative cylinder notation to achieve the lowest refractive power S. min and the highest refractive power S max The values ​​are respectively:

[0283] S min =S+C, S max =S (17)

[0284] In a specific implementation, the general model 23 used by device 1 is a more complex version of the model developed in Example 1, and corresponds to equations (1) to (5), while taking into account the cylindrical lens defocus C. Therefore, it is noted that:

[0285] H' = S + C / 2 + 1 / Dist (18)

[0286] The value of defocus blur δ' in General Model 23 is:

[0287]

[0288] in:

[0289]

[0290] In formula (20), 0 <H’<A max Corresponding to the accommodation region, and in the case of myopia (S<0), H'≤0 and H'≥A max These correspond to the far-field and near-field regions, respectively. PR is given by -2 / (2S+C) according to formulas (18) and (20).

[0291] In the general model 23, two meridians are specifically considered: one perpendicular to the axial component α and the other corresponding to the lowest refractive power S. min Related to defocus blur δ1 and visual acuity VA axis1 and MAR MAR axis1 And another line along the axial component α and with the highest refractive power S max Related to defocus blur δ2 and visual acuity VA axis2 and MAR MAR axis2 .

[0292] The values ​​for defocus blur δ1 and δ2 are:

[0293] δ1=|S min +1 / Dist-A|=|S+C+1 / Dist-A| (21)

[0294] δ2=|S max +1 / Dist-A|=|S+1 / Dist-A| (22)

[0295] The defocus blur δ' is related to the defocus blurs δ1 and δ2 through the following relationship:

[0296]

[0297] (i.e., the root mean square radius of the fuzzy ellipse).

[0298] In particular, in this adjustment region (formulas (19), (21), and (22) that take into account formulas (18) and (20):

[0299] δ'=δ1=δ2=|C / 2| (24)

[0300] In the following text, the term "mean MAR" currently refers to the MAR of an astigmatic user's eye corresponding to an equivalent spherical power, i.e., this can be obtained by measuring MAR using a non-directional test target. The same notation represents "mean VA". In a manner similar to Equation (5), the mean MAR (denoted as MAR) Seq And with average VA VA Seq (Associated) can be represented as:

[0301]

[0302] Where MAR0, Q, and q are model parameters.

[0303] This model can be derived from Gómez-Pedrero’s article (especially formula (8)) by introducing the further influence of distance Dist.

[0304] MAR0 corresponds to the optimal correction VA for a given pupil size. The parameter q is similar to the parameter considered in the general model 23 of Example 1, reflecting the defocus tolerance, and is generally between 1 and 3. Regarding the parameter Q, it can be expressed in the general model 23 as equal to:

[0305] Q = (KD) q (26)

[0306] Here, D still refers to the pupil diameter of the user's eye, and K is another parameter.

[0307] Consistently, MAR axis1 and MAR axis2 They are represented as follows:

[0308]

[0309] VA Seq VA axis1 and VA axis2 Explanatory trends Figure 9 Curve 50 shows that these curves represent the conditions when S = -0.5D, C = -1D, and A max =In specific myopia cases of 2D, VA (axis 5002, in arcminutes reciprocal) -1 The distance (axis 5001, in meters) is used as a function of distance. Then, the equivalent spherical power is H = -1D (Equation (16)), which depicts the adjustment region A5-2 between PP (denoted as PP5) of 1m and PR (denoted as PR5) of 0.33m (Equations (18) and (20)). These correspond to VA respectively. axis1 VA axis2 and VA Seq The three curves 501, 502, and 503 have the same layout in the accommodation region A5-2, but differ significantly from each other in the distance vision region A5-1 (beyond PR5) and the myopia region A5-3 (below PP5). Curve 501 (VA) axis1 It has the lowest VA for distance vision and the best VA for near vision, curve 502 (VA) axis2 Conversely, it has the highest VA for distance vision and the lowest VA for near vision, and curve 503 (VA) Seq It lies between the two.

[0310] Finally, the general model 23 depends on seven parameters in determining the VA: three parameters concerning refractive error, namely S, C, and α; one parameter concerning distance Dist; and three parameters corresponding to MAR0, Q, and q. Using Dist as a variable to provide reference fuzzy data 22 leaves six unknowns for evaluation.

[0311] In some modes of device 1, the resolution depends not only on the variable distance but also on the orientation angle associated with one or more oriented test targets. Accordingly, the reference fuzzy data 22 includes both distance and angle.

[0312] Specific related implementation methods include the following aspects:

[0313] Under the first chain, VA is performed at three or more distinct distances, each associated with a different blur level. Seq VA measurement. In this regard, it is appropriate to use an optometric font without a specific orientation, such as Figure 10A The refraction font, indicated by a ring, is denoted as 55A, particularly relevant to determining the equivalent spherical power. In myopia, distance may include one distance in the accommodation region and two distances in the distance vision region. An iterative process based on an initial approximation, as developed in Example 1, can be used.

[0314] In the second chain, the axis component α is extracted. In some modes, this is done by using the parent dial (e.g., Figure 10B The basic process of the dial (denoted as 55B) is completed as is known to those skilled in the art, and the display of the dial is beyond the user's eye accommodation range (i.e., A=0).

[0315] In an alternative mode, the directional refraction font is rotated in front of the user's eyes beyond the user's eye's accommodation area (i.e., A=0) to achieve a maximum and / or minimum sharpness. This can be achieved simply by displaying the refraction font on the screen of a mobile device (e.g., a smartphone) and by rotating the mobile device until the relevant direction is reached, possibly by means of an IMU (roll measurement unit) present in the mobile device to identify and record that relevant direction. The screen of the mobile device can be pointed towards a reflector, allowing the user to determine the relevant direction by viewing the refraction font reflected in the reflector rather than directly on the screen. Therefore, a greater distance can be obtained between the user's eyes and the visualized refraction font. The directional refraction font displayed on the screen can, for example, take the form of the 55C (“C”) refraction font used in the Landau C test or the 55D (“E”) refraction font used in the Roll E test (the latter having four possible orientations, including upright, reversed, and tilted to the right or left), as shown in the respective... Figure 10C and Figure 10D As shown above. It can take the form of a grating, or include gratings, such as a parallel line disk, as... Figure 10E What is represented may be located within the Gabor spot, such as... Figure 10F What it represents.

[0316] Depending on the variant, the optometry font rotates within the user's eye's accommodation area.

[0317] In the third chain, once the axis component α is extracted via the second chain, directional refraction text is displayed to the user's eye at a predetermined distance beyond the user's eye's accommodation zone (i.e., A = 0), based on the astigmatic axis (axis component α) and / or the vertical axis (α + 90°). Then, VA is measured, generating VA at the selected distance Dist. axis1 and / or VA axis2 .

[0318] In the favorable mode, VA is measured. axis1 and VA axis2 Both. This can facilitate calculations, particularly for extracting C visible in formulas (27) and (28). In variations, only the astigmatic axis or only the vertical axis is used for measurements at two or more distances using directional optometry fonts.

[0319] As will be obvious to the reader, although the third chain depends on the second chain, the first chain can be in any order relative to the second and third chains.

[0320] Alternatively or in combination, VA measurements are used in one or more directions other than the astigmatic axis and the vertical axis. Using the general model 23 developed above, the MAR of the directional optometric font along the axis angle θ is given by the following equation:

[0321]

[0322] This relation can be used to evaluate unknowns.

[0323] In a specific mode, three measurements are performed to obtain the average VA, including one measurement within the adjustment region and two measurements outside the adjustment region. The axial component α is identified separately, and two measurements are performed along the discovered astigmatic axis and vertical axis, respectively. This provides five independent equations from which estimates of five unknowns S, C, MAR0, Q, and q can be derived. As is known to those skilled in the art, these estimates can be achieved through any suitable numerical iterative optimization method.

[0324] As described above and as can be seen from the formulas, multiple variant implementations are possible. For example, all parameters except the axial component α (which can be determined independently) can be derived from average VA measurements at different distances. In other examples, once the axial component α is known, all other parameters can be derived from VA measurements at various distances along a given orientation angle. In other modes, VA measurements are performed at three or more orientation angles and the same distances (including along the astigmatic axis and the vertical axis).

[0325] The annotations regarding parameter initialization and / or partial parameter settings in Example 1 are also relevant here. Furthermore, in cases where such initialization is at least partially based on prescription values ​​from a user who already has glasses, in addition to those for VA... max In addition to vision correction, the indicated astigmatism angle can also provide a good starting point for the axial component α.

[0326] Moreover, in a manner similar to Examples 1 and 2, the parameters can be evaluated fully or partially during the reference time period T0, and these parameters can be fully or partially utilized in a later VA measurement intended to assess refractive error.

[0327] The embodiments of this Example 4 can be further combined with the use of multiple optical power corrections, as described in Example 3.

[0328] Example 5 - Refractive error estimation combined with pupil diameter

[0329] In the series of examples above, by maintaining the same or similar user and environmental conditions (especially brightness) while continuing to perform VA measurements, the pupil diameter D of the user's eye is at least implicitly considered. This is highly desirable because the pupil diameter D is often a first-order parameter in visual acuity models, the role of which is pointed out above and developed in Gómez-Pedrero's paper.

[0330] The influence of D on VA Figure 11 As shown in the figure, the optimal corrected MAR (axis 652, in arcminutes) is schematically illustrated as a function of pupil diameter (axis 651, in mm) with a specific curve 65, i.e., the MAR without defocus blur (e.g., the MAR in Examples 1 and 2). min For example, MAR0 in Example 4, see formulas (5), (6), (25), (27), (28), and the reader can find a detailed explanation of this in Gómez-Pedrero's article. As can be seen on curve 65, for small values ​​of D up to a first size threshold of about 3 mm, the optimal corrected MAR is a strongly decreasing function of D (diffraction profile constraint). Above this first size threshold, the optimal corrected MAR increases at a slow, globally linear rate due to the presence of higher-order aberrations (denoted as "HOA"). This increase remains valid up to a second size threshold of about 6 mm, above which the MAR stabilizes, which can be explained by the Styles-Crawford effect known to those skilled in the art.

[0331] Accordingly, in the case of low to no blur, the MAR varies substantially with D of the above-mentioned optimal corrected MAR. In contrast, for sufficiently high amounts of blur (typically above 1D), the MAR is roughly proportional to the pupil diameter D and the defocus blur (δ, δ', δ1 or δ2, depending on the target measurement), while the contribution of the optimal corrected MAR becomes weaker or negligible.

[0332] All embodiments described below take into account the pupil diameter D, and can be combined, in particular, with Examples 1 to 4.

[0333] In some embodiments, light conditions are carefully controlled.

[0334] In other embodiments, it is directly ensured that the pupil diameter D of the user's eye remains constant across different VA measurements. This achievement provides better accuracy because D can change substantially over time and with varying object distances, even at the same luminance.

[0335] Based on the evaluation of D using a mobile device that includes at least a front-facing camera, the following reveals a specific mode for achieving this pupil diameter control.

[0336] In near vision, the distance Dist is typically up to about 50 cm, measured directly by a camera on a mobile device. This dimensional information can be obtained directly via a dedicated mobile application. Alternatively, pupil size can be inferred from reference dimensions such as the distance between the ears or the diameter of the iris. This latter relative estimate may prove quite reliable because the iris diameter is very stable.

[0337] In distance viewing situations, typically at a distance of at least 3 meters from the user's eyes, a setup similar to that disclosed in the aforementioned patent application WO2022 / 013410 can be used. That is, as Figure 12A and Figure 12B (Not drawn to scale) As shown, a user holds a mobile device 61 (such as a smartphone equipped with a front camera, a rear camera, and a screen) close to the user's eyes 60. The user sees the screen of the mobile device through a reflector 62 located in front of the user at a distance d from the mobile device, and the pupil diameter D is measured by the rear camera. More precisely, the measurement process can be performed, for example, using a smartphone as follows:

[0338] - The user positions himself in front of the reflector 62;

[0339] -With the smartphone 61 kept close to the eye 60, the user uses the front-facing camera and the pattern or optometry font displayed on the screen to adjust the distance Dist to twice the distance d from the smartphone 61 to the reflector 62, usually between 3m and 4m.

[0340] - While maintaining the selected position, the user moves the smartphone 61 away from the face, for example, about 20cm to 30cm, and uses the rear camera to measure the pupil diameter D of the eye 60; as with near vision, the change in D can be evaluated in particular relative to the iris diameter of the eye 60.

[0341] - Adjust the ambient lighting until the desired pupil diameter D is observed;

[0342] - The user brings the smartphone 61 back to a position close to the eyes 60, thus hiding the eyes, and continues the visual acuity test;

[0343] - The above operation may be repeated with different distance values ​​for Dist.

[0344] By controlling the lighting conditions, the same pupil diameter D can then be set in subsequent VA measurement operations.

[0345] In the variant, the pupil diameter D varies between VA measurements instead of remaining constant, but based on the appropriate expression in the general model 23, the pupil diameter is monitored and causes compensation for the variation. For example, the VA in equations (5) and (6) of Examples 1 and 2 max MAR0 and Q in equations (25) to (29) of E2 and Example 4 are expressed as functions of D. The relevant modeling representations can be found in Gómez-Pedrero’s article, especially in the relevant formulas (5), (6), and (7) in that article.

[0346] Example 6 - Estimation of refractive error based on VA under multiple pupil diameters

[0347] Although in this embodiment of the invention, Example 5 takes into account the pupil diameter D when deriving the Rx estimate from controlled variations of other entities (such as distance Dist (Examples 1 and 2), optical power correction (Example 3), or orientation angle (Example 4)), the pupil diameter D of the user's eye is itself directly used to apply the controlled variation, enabling the derivation of the Rx estimate 25A and / or parameter value 25B. In practice, different values ​​of D are part of the reference fuzzy data 22 and are associated with corresponding terms of the reference VA measurement 21.

[0348] Similarly, relevant modeling representations can be found in, for example, Gómez-Pedrero's articles.

[0349] The changes to D can be combined with changes to any or all other entities mentioned above.

[0350] The process for estimating the refractive error of a user's eye based on a VA measurement will now be described in detail in an example that may be applicable to all of the above embodiments, and this process may be performed using device 1. Figure 13 This indicates that the process includes:

[0351] - Receive reference VA measurement 21 and associated reference fuzzy data 22 (step 81);

[0352] -Concretize the general model 23 into a specific model 24 based on the received reference VA measurement 21 and reference fuzzy data 22 (step 82);

[0353] - Generate information related to refractive errors 25 (step 83);

[0354] - Optionally, inferred reference fuzzy data 22' is determined based on the received reference VA measurement 21 and reference fuzzy data 22 (step 84), and those data 22' are re-injected as additional items of reference fuzzy data 22 for use in additional items of reference VA measurement 21;

[0355] - Provide information related to refractive errors 25 (step 85).

[0356] exist Figure 14 The aforementioned device 1 is implemented in a specific device 9 that is visible above. It corresponds to, for example, a tablet computer, a smartphone, a head-mounted display (HMD), a laptop computer, a desktop computer, or a workstation.

[0357] Device 9 includes the following components, which are interconnected via an address and data bus 95, which also transmits clock signals:

[0358] - Microprocessor 91 (or CPU);

[0359] -ROM-type non-volatile memory 96;

[0360] - Random Access Memory (RAM) 97;

[0361] - One or more I / O (input / output) devices 94, such as a keyboard, mouse, joystick, webcam; other modes for introducing commands are also possible, such as voice recognition;

[0362] - Power supply 98; and

[0363] - Radio Frequency (RF) Unit 99.

[0364] Depending on the variant, power supply 98 is located outside device 9.

[0365] Device 9 also includes a display device 93 of the display screen type. According to variations, the display device is external to device 9 and is connected to the device via cable or wirelessly for transmitting display signals. Device 9 includes an interface for transmission or connection, adapted to transmit display signals to an external display device, such as an LCD (liquid crystal display) or OLED (organic light-emitting diode) screen or video projector. In this regard, RF unit 99 can be used for associated wireless transmission.

[0366] It should be noted that the word "register" used in the description of memory 97 can specify both low-capacity memory areas (some binary data) and large-capacity memory areas (enabling the storage of all or part of the entire program or representing the calculated or displayed data) in each of the mentioned memories. Furthermore, registers can be arranged and configured in any manner, and each of them does not necessarily correspond to an adjacent memory location, and can be distributed in other ways (this particularly covers the case where a register comprises several smaller registers).

[0367] RAM 97 specifically includes:

[0368] - The operating program of microprocessor 91 is located in register 970;

[0369] - In register 971, information about reference VA measurement 21 and reference fuzzy data 22 is represented;

[0370] - Register 972 contains information about the general model 23;

[0371] - Register 973 contains information about a specific model 24;

[0372] - In register 974, information related to refractive errors includes information representing Rx estimate 25A and / or parameter value 25B.

[0373] When powered on, the microprocessor 91 loads and executes the instructions of the program contained in RAM 97.

Claims

1. A device (1; 1A) for estimating the refractive error (Rx) of a user's eye based on the visual acuity of the user's eye (60). ；1B； 1C), characterized in that the device comprises: - At least one input terminal (11), said at least one input terminal being adapted to receive at least two measurements (21) of the visual acuity of the user's eye, referred to as reference visual acuity measurements, and data (22) associated with at least two different blur levels respectively associated with said reference visual acuity measurements, referred to as reference blur data, said reference visual acuity measurements being performed by applying a corresponding blur level to the user's eye during a reference measurement period (T0). - At least one processor, the at least one processor being configured to concretize a general model (23) involving theoretical visual acuity (VA), theoretical ambiguity data (Dist; θ; D), and theoretical refractive error (S, C, α) into a specific model (24) of the user's eye based on the reference visual acuity measurement and the reference ambiguity data, and to generate information (25) related to the refractive error of the user's eye based on the specific model. - At least one output terminal (15), said at least one output terminal being adapted to provide said information in relation to the refractive error of the user's eye.

2. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to claim 1, characterized in that, The resulting information related to the refractive error includes an estimate (25A) of the user's eye refractive error during the reference measurement period (T0).

3. The apparatus (1; 1B; 1C) for estimating the refractive error of the user's eye according to claim 1 or 2, characterized in that, The resulting information related to the refractive error includes at least one parameter value (25B) associated with the particular model, which enables the generation of an estimate (28) of the user's refractive error after the reference measurement period (T0) based on at least one measurement (26) of the user's visual acuity, referred to as a later visual acuity measurement, which is called a later refractive error estimate.

4. The apparatus (1; 1C) for estimating the refractive error of the user's eye according to claim 3, characterized in that, The at least one input (11) is adapted to receive the at least one later visual acuity measurement (26), the at least one processor is configured to generate the later refractive error estimate (28) according to the specific model (24) and according to the at least one later visual acuity measurement, and the at least one output (15) is adapted to provide the later refractive error estimate (28).

5. The device for estimating the refractive error of the user's eye according to claim 3 or 4 (1; 1B; 1C), characterized in that, The reference fuzzy data (22) includes at least one measurement of the user's eye refractive error during the reference measurement period.

6. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to any one of the preceding claims, characterized in that, The blur level is applied at least in part by performing the reference visual acuity measurement (21) at at least two reference distances (Dist1, Dist2, Dist3) from the user’s eye to the test target, and the theoretical blur data includes the theoretical distance (Dist) from the user’s eye.

7. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to claim 6, characterized in that, The general model (23) successively involves three regions starting from the user's eye and moving away from the user's eye, the three regions being related to the user's theoretical visual acuity and including a nearest region (A1-3; A2-3; A3-3; A5-3), an accommodation region (A1-2; A2-2; A3-2; ​​A5-2), and a farthest region (A1-1; A2-1; A5-1), the accommodation region being associated with a flat range of the theoretical visual acuity, and the nearest region and the farthest region being associated with variable values ​​of the theoretical visual acuity, the blur level being imposed at least in part by performing the reference visual acuity measurement (21) at the reference distances including at least one distance (Dist2) corresponding to the accommodation region and at least one distance (Dist1, Dist3) corresponding to at least one of the nearest region and the farthest region, respectively.

8. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to any one of the preceding claims, characterized in that, At least one of the blur levels is applied, at least in part, by performing at least one of the reference visual acuity measurements (21) at at least one reference orientation angle (0°, 90°; θ) of the orientation test target (55C), the theoretical blur data including the theoretical orientation angle relative to a predetermined axis, and the at least one processor is configured to further embody the general model (23) based on the at least one reference orientation angle.

9. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to claim 8, characterized in that, The user's eye refractive error has a spherical component, a cylindrical component, and a cylindrical axis, and the theoretical refractive error has a theoretical spherical component (S), a theoretical cylindrical component (C), and a theoretical cylindrical axis (α): - The at least one input terminal (11) is adapted to receive at least one of the reference visual acuity measurements with respect to the non-directional test target (55A), at least one of the reference visual acuity measurements at at least one reference orientation angle of the directional test target (55C), and respectively associated reference ambiguity data (22), the reference ambiguity data further including information about at least one offset (0°, 90°; θ) of the cylindrical lens axis relative to a predetermined axis. - The at least one processor is configured to concretize the general model (23) into the specific model (24) based on the reference visual acuity measurement (21) regarding both the non-directional test target and the directional test target and based on the reference fuzzy data (22), and to generate information related to the spherical and cylindrical components of the user's eye refractive error. - The at least one output terminal (15) is adapted to provide the information related to the spherical component and the cylindrical component.

10. The apparatus (1; 1B; 1C) for estimating the refractive error of the user's eye according to any one of the preceding claims, characterized in that, At least one of the blur levels is applied, at least in part, by performing at least one of the reference visual acuity measurements with at least one lens arranged in front of the user's eye, so as to cause at least one predetermined optical power correction (+0.25D, +0.5D; [-4D, -0.25D]), and the theoretical blur data includes the theoretical optical power correction (DiffS1; DiffS3) applied to the user's eye.

11. The apparatus (1; 1B; 1C) for estimating the refractive error of the user's eye according to claim 10, characterized in that, In the general model (23), the theoretical refractive error change (ΔS) is equal to the theoretical optical power correction (DiffS1).

12. The apparatus (1; 1A; 1B; 1C) for estimating the refractive error of the user's eye according to any one of the preceding claims, characterized in that: - The at least one input terminal (11) is adapted to receive pupil size data of the user's eye (60) during the reference visual acuity measurement, referred to as reference pupil data. - The general model (23) relates to a theoretical pupil size (D), and the at least one processor is configured to further materialize the general model into the specific model (24) based on the reference pupil data.

13. An automatic module for myopia control, comprising means (1; 1A) for estimating the refractive error (Rx) of a user's eye based on the visual acuity of the user's eye (60). ；1B； 1C), characterized in that the device comprises: - At least one input terminal (11), said at least one input terminal being adapted to receive at least two measurements (21) of the visual acuity of the user's eye, referred to as reference visual acuity measurements, and data (22) associated with at least two different blur levels respectively associated with said reference visual acuity measurements, referred to as reference blur data, said reference visual acuity measurements being performed by applying a corresponding blur level to the user's eye during a reference measurement period (T0). - At least one processor, the at least one processor being configured to concretize a general model (23) involving theoretical visual acuity (VA), theoretical ambiguity data (Dist; θ; D), and theoretical refractive error (S, C, α) into a specific model (24) of the user's eye based on the reference visual acuity measurement and the reference ambiguity data, and to generate information (25) related to the refractive error of the user's eye based on the specific model. - At least one output terminal (15), said at least one output terminal being adapted to provide said information in relation to the refractive error of the user's eye.

14. A method for estimating the refractive error (Rx) of a user's eye (60) by at least one processor based on the visual acuity of the user's eye, characterized in that, The method includes: - Receive (81) at least two measurements (21) of the visual acuity of the user's eye, referred to as reference visual acuity measurements, and data (22) associated with at least two different blur levels respectively associated with the reference visual acuity measurements, referred to as reference blur data, wherein the reference visual acuity measurements are performed by applying the corresponding blur levels to the user's eye during the reference measurement period (T0). -Based on the reference visual acuity measurement and the reference fuzzy data, a general model (23) involving theoretical visual acuity (VA), theoretical fuzzy data (Dist; θ; D), and theoretical refractive error (S, C, α) is concretized (82) into a specific model (24) for the user's eye, and information (25) related to the refractive error of the user's eye is generated based on the specific model. - Provide (85) the information relating to the refractive error of the user's eye.

15. A computer program comprising software code, which, when executed by a processor, is adapted to perform a method for estimating refractive errors of a user's eye based on the visual acuity of the user's eye (60), the method comprising: - Receive (81) at least two measurements (21) of the visual acuity of the user's eye, referred to as reference visual acuity measurements, and data (22) associated with at least two different blur levels respectively associated with the reference visual acuity measurements, referred to as reference blur data, wherein the reference visual acuity measurements are performed by applying the corresponding blur levels to the user's eye during the reference measurement period (T0). -Based on the reference visual acuity measurement and the reference fuzzy data, a general model (23) involving theoretical visual acuity (VA), theoretical fuzzy data (Dist; θ; D), and theoretical refractive error (S, C, α) is concretized (82) into a specific model (24) for the user's eye, and information (25) related to the refractive error of the user's eye is generated based on the specific model. - Provide (85) the information relating to the refractive error of the user's eye.

Images