Method and apparatus for detecting mismatch in optical devices

Abstract

1. An apparatus for detecting misfit of an optical device on a wearer, comprising: -Receive and store head data related to the wearer's head over time when wearing and using the optical device; - processing head data based on head data patterns associated with known mismatches of an optical device with a wearer of said optical device; - an apparatus comprising a processing circuit configured to detect a misfit of an optical device to a wearer by matching received and stored head data with a head data pattern associated with the misfit.

Description

[Technical Field]

[0001] SUMMARY The present disclosure relates to an apparatus and related method for detecting misfit of an ophthalmic lens provided to a wearer.

[0002] Additionally, the present disclosure relates to a computer program product including one or more stored sequences of instructions accessible by a processor. [Background technology]

[0003] It is known that some wearers may experience poor fit with the ophthalmic lenses they receive. This can occur over time or immediately after wearing the ophthalmic lenses. For example, fitting a progressive lens typically requires the wearer to change the alignment of their head and eyes to properly use the optical zone of the lens. One cause of poor fit is that a particular wearer does not properly use the optical zone of the progressive lens while wearing the lens.

[0004] Each wearer will adopt a specific body posture to perform a specific task, such as reading. Consider a seated wearer reading a document placed on a desk in front of them, and different wearers will tilt their head differently to read the document.

[0005] Eye trackers and motion capture are used in the lab to assess the wearer's head-eye coordination, with a smart frame measuring the wearer's descent slope at any time of day, for example while reading or performing other tasks.

[0006] The wearer must go to an optical testing laboratory to perform the task in a specific environment and determine the incompatibility of the ophthalmic lens, which may result in the incompatibility being detected too late or not at all, causing discomfort to the wearer and leading to abandonment of the ophthalmic device. Summary of the Invention [Problem to be solved by the invention]

[0007] Therefore, it is necessary to detect the incompatibility of an ophthalmic lens in the wearer's daily life environment, taking into account the wearer's body posture.

[0008] It is also necessary to accommodate wearers who are not suited to ophthalmic lenses. [Means for solving the problem]

[0009] To this end, the present disclosure proposes a device for detecting the incompatibility of an optical device to a wearer, said device comprising: -Receive and store head data related to the wearer's head over time when wearing and using the optical device; - processing head data based on head data patterns associated with known mismatches of an optical device with a wearer of said optical device; Detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch. The processing circuit is configured as follows.

[0010] Advantageously, the apparatus allows for self-detection of optical device incompatibility based on head data, without the need for the knowledge of an ophthalmologist.

[0011] Furthermore, the device allows for detection of wearer mismatches based on head data parameters such as head movement and head posture. An example of head movement is head bowing. If the recorded head data may correspond to an abnormal head data pattern associated with a known mismatch, the device detects an optical mismatch.

[0012] According to further embodiments, which may be considered alone or in combination, The head data includes at least one of head pose and movement data, gaze direction, eyelid opening, pupil size, EEG data, facial expressions; and / or When processing the head data, an indication of the wearer's activity is considered by matching the received data and stored head data with head data patterns associated with the mismatch; and / or the optical device has ophthalmic functionality, and the processing circuitry is further configured to receive an ophthalmic prescription of the wearer and to take said ophthalmic prescription into account when comparing the received and stored head data with the head data pattern; and / or the processing circuit is further configured to alert the wearer or an eye care professional that an incompatibility of the optical device has been detected; and / or The cause of the non-conformity is selected from the list consisting of the optical function of the optical device, the fitting parameters of the optical device, the use of the optical device by the wearer, the ophthalmic prescription of the wearer, the integrity of the optical device, and / or The optical device has a progressive addition power, and the head data relates to a distribution of head depression angles of the wearer when using the optical device.

[0013] The present disclosure further relates to a method performed by a processing circuit of the device, the method comprising: - receiving and storing head data relating to the wearer's head over time when wearing and using the optical device; - processing head data based on head data patterns associated with known mismatches of an optical device to a wearer of said optical device; - detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch; Includes.

[0014] Advantageously, detecting a misfit of the optical device to the wearer by matching the received and stored head data with head data patterns associated with the misfit makes it possible to determine the cause of the misfit. Depending on the cause of the misfit, a dedicated solution can be offered to the wearer, such as specific training to improve the fit to the optical device or by offering a different optical design.

[0015] This method allows the wearer to self-detect mismatches with the optical device. The wearer does not need to go to an optical testing laboratory and perform specific tasks to determine whether the optical device is compatible. Based on the head data and known mismatch head data patterns, the device can detect mismatches with the optical device by itself.

[0016] According to further embodiments, which may be considered alone or in combination, The head data includes at least one of head pose and movement data, gaze direction, eyelid opening, pupil size, EEG data, facial expressions; and / or When processing the head data, an indication of the wearer's activity is considered by matching the received data and stored head data with head data patterns associated with the mismatch; and / or the optical device has ophthalmic functionality, and the method further comprises receiving an ophthalmic prescription of the wearer and taking said ophthalmic prescription into account when comparing the received and stored head data with the head data pattern; and / or The method further comprises alerting the wearer or an eye care professional that an incompatibility of the optical device has been detected; and / or The method further comprises determining the cause of the mismatch of the optical device by comparing the received and stored head data with the head data pattern; and / or The cause of the non-conformity is selected from the list consisting of: the optical function of the optical device, the fitting parameters of the optical device, the use of the optical device by the wearer, the ophthalmic prescription of the wearer; and / or the optical device has a progressive addition power and the head data relates to the distribution of head depression angles of the wearer when using the optical device; and / or Detecting the mismatch includes determining the number of modes in the distribution of head drop angles.

[0017] The present disclosure further relates to a method for determining a head pattern, the method comprising: receiving and storing wearer data associated with a wearer of the optical device; receiving and storing optical device data relating to an optical device associated with each wearer; receiving and storing longitudinal head data associated with each wearer's head when wearing and using an associated optical device; receiving and storing fit data related to the fit of an optical device associated with each wearer; - processing the received and stored data across a large number of wearers to determine at least one head pattern associated with a misfit of the optical device using a machine learning model to find correlations between the head data (and other optional data) and the misfit; Acquiring head data, which may consist of several time series of frame measurements; Compute descriptive statistics for each time series; Grouping descriptive statistics into data vectors to summarize head data; Grouping data related to the fit of the optical device associated with each wearer into a single number representing the degree of fit, ranging from 1 (poor fit) to 10 (perfect fit); and Initiating a machine learning model where the input data are the data vectors for each wearer and the data to predict are numerical values ​​representing the goodness of fit; using a machine learning model to define head data patterns associated with optical device mismatch; and processing the Includes.

[0018] According to further embodiments, which may be considered alone or in combination, The head data includes at least one of head pose data, gaze direction, eyelid opening, and pupil size; and / or The wearer data includes the wearer's ophthalmic prescription and the optical device has ophthalmic functionality; and / or Optical device data includes data relating to the optical functions of the optical device and / or fitting parameters of the optical device and / or use of the optical device by the wearer and / or the wearer's ophthalmic prescription.

[0019] The present disclosure further relates to a non-transitory computer-readable storage medium containing computer-executable instructions that, when executed by a computer, cause the computer to perform a method performed by a processing circuit of an apparatus according to the present disclosure.

[0020] The present disclosure further relates to a non-transitory computer-readable storage medium containing computer-executable instructions that, when executed by a computer, cause the computer to perform a method for determining a head pattern.

[0021] Embodiments of the present disclosure will now be described, by way of example only, with reference to the following drawings: [Brief explanation of the drawings]

[0022] [Figure 1] 1 shows a flow chart of steps performed by a processing circuit of an apparatus for detecting a misfit of an optical device on a wearer. [Figure 2a] 1 shows a schematic diagram of the optical system of the eye and ophthalmic lens. [Figure 2b] 1 shows a schematic diagram of the optical system of the eye and ophthalmic lens. [Figure 3] 1 shows a flowchart of method steps for detecting misfit of an optical device on a wearer. [Figure 4]1 shows a histogram of the wearer's head tilt angle for the period Monday through Friday. [Figure 5] 1 shows a histogram of the wearer's head tilt angle for the Saturday and Sunday periods. [Figure 6] 1 shows a histogram of unimodal head depression angles. [Figure 7] 1 shows a histogram of bimodal head depression angles. [Figure 8] 1 shows a flow chart of steps for determining a head pattern. DETAILED DESCRIPTION OF THE INVENTION

[0023] Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the size of some of the elements in the figures may be exaggerated relative to other elements to help improve understanding of embodiments of the present disclosure.

[0024] The present disclosure relates to an apparatus for detecting the misfit of an optical device on a wearer.

[0025] A first aspect of the present disclosure relates to an apparatus for detecting the misfit of an optical device on a wearer, the apparatus comprising: -Receive and store head data related to the wearer's head over time when wearing and using the optical device; - processing head data based on head data patterns associated with known mismatches of an optical device with a wearer of said optical device; Detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch. The processing circuit is configured as follows.

[0026] FIG. 1 is a flow chart showing the various steps performed by the processing circuitry of the device.

[0027] The device including the processing circuit executes a receiving step S10, in which the processing circuit proceeds to receive head data of the wearer measured by at least one head data sensor, and stores the measured head data in a data storage medium.

[0028] Further, the device comprising the processing circuitry executes processing step S12, in which the processing circuitry proceeds to compare the measured head data of the wearer with a predetermined head data pattern associated with the cause of the misfit of the optical device to the wearer.

[0029] Finally, the device comprising the processing circuitry performs a detection step S14, in which the processing circuitry proceeds to detect a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch.

[0030] In the context of the present disclosure, the term misfit relates to a misfit of the optical device to the wearer. Said misfit may be caused by the fact that the design of the ophthalmic lens does not take into account the body behavior and body posture of the wearer. An example of body behavior and posture is the position of the wearer's head when performing a given activity or task.

[0031] According to the present disclosure, the term optical device may relate to optical lenses, more particularly ophthalmic lenses. The optical device may be a single vision lens, a bifocal lens, a trifocal lens, a progressive lens, a filter lens, or any other type of optical device that allows the wearer to improve their vision.

[0032] The term head data refers to any data that can be measured over time about the wearer's head. For example, the head data may relate to the position or orientation of the wearer's head relative to the torso.

[0033] In this way, the head data may relate to, for example, head tilt or head rotation along a horizontal axis. Head data relating to head tilt may be stored in the form of angle parameters. Such angle parameters may correspond to descent angles.

[0034] The head data can be acquired using a smart frame. The smart frame corresponds to a frame including a head data sensor that detects the movement of the wearer's head. When the considered head data corresponds to head displacement, the smart frame can be equipped with a head data sensor, such as an IMU (Inertial Measurement Unit). In this example, the head data sensor, such as an IMU, detects information related to the displacement of the wearer's head, such as the tilt or rotation of the wearer's head, and transmits it to the device.

[0035] The apparatus according to the present disclosure is configured to receive and store any displacement of the wearer's head when the wearer is wearing the optical device.

[0036] The device stores the wearer's head data over time. Once the wearer's head data is stored in the device, the wearer's head data is compared with previously acquired head data patterns associated with optical device mismatches. Upon performing the comparison, the device can detect a similarity between the measured wearer's head data and the optical device mismatch head data pattern. If a given similarity threshold is exceeded, the device determines that the optical device is not compatible with the wearer.

[0037] The similarity can be determined using a histogram. For example, a histogram representing the frequency of each data value is calculated based on the wearer's head data over time. This histogram is compared with histograms from previously acquired data. The most similar histogram is determined as the one whose overall shape is closest to the calculated one.

[0038] The similarity can be further determined using a machine learning algorithm. For example, a machine learning model is constructed based on previously acquired data to determine the misfit state of the optical device to the wearer based on the head data. The model is trained using the acquired data. For example, the model can be a decision tree, a random forest, a state vector machine, a neural network, a deep learning algorithm, or the like. The wearer's data is sent to the machine learning model to determine the misfit state or similarity state of the wearer.

[0039] In the case of a misfit of a progressive lens, the wearer can compensate for the fact that they cannot see through a certain area of ​​the progressive lens by moving their head. The certain area of ​​the optical device may correspond, for example, to the near vision zone. The wearer's head data corresponding to the unnatural head posture that compensates for the inability to see through the certain area is then compared with the misfit head data pattern of the optical device. Based on the comparison and further similarity with the misfit head data pattern of the optical device, the apparatus can determine that the optical device is not a fit for the wearer.

[0040] The poor fit of progressive lenses can also be determined based on how often the wearer finds a head position and gazes through the progressive lens that provides good vision for both near and far. Poorly fitted wearers tend to adopt the same typical normal posture.

[0041] Head data can include parameters that vary between head pose and movement data, gaze direction, eyelid opening, pupil size, EEG (electroencephalography) data, and facial expressions.

[0042] The term head pose corresponds to the position of a particular wearer's head relative to the torso in everyday life when performing a particular task, or reference position.

[0043] The term head movement data corresponds to data measured for head displacement.

[0044] The term gaze direction is defined by two angles measured with respect to a direct orthonormal basis about the center of rotation of the right or left eye.

[0045] Figures 2a and 2b are schematic diagrams of the optical system of the eye and lens, and therefore illustrate the definitions used herein. More precisely, Figure 2a represents a perspective view of such a system, showing the parameters α and β used to define the gaze direction. Figure 2b shows a vertical plane parallel to the anterior-posterior axis of the wearer's head and passing through the center of rotation of the eye, when parameter β is equal to 0.

[0046] The center of rotation of the eye is labeled Q'. The axis Q'F', shown by a dashed line in Figure 2b, is a horizontal axis that passes through the center of rotation of the eye and extends in front of the wearer, i.e., the axis Q'F' corresponds to the primary gaze view. This axis cuts the front surface of the optical device, which may be a lens, at a point called the fitting cross, which is present on the lens to allow the optometrist to align the lens in the frame. The intersection of the rear surface of the lens and the axis Q'F' is O. O can be the fitting cross if it is located on the rear surface. The upper spherical surface with center Q' and radius q' touches the rear surface of the lens at a point on the horizontal axis. As an example, a value of radius q' of 25.5 mm corresponds to a normal value and gives satisfactory results when wearing the lens.

[0047] A given gaze direction, indicated by a solid line in Fig. 2a, corresponds to the position of the eye in the rotational direction about Q' and to a point J on the vertex sphere, and the angle β is the angle formed between the axis Q'F' and the projection of the straight line Q'J onto the horizontal plane containing the axis Q'F', which angle appears in the scheme of Fig. 2b. The angle α is the angle formed between the axis Q'J and the projection of the straight line Q'J onto the horizontal plane containing the axis Q'F', which angle appears in the scheme of Fig. 2b. A given gaze view therefore corresponds to a point J on the vertex sphere, or to a pair (α, β). The more positive the value of the downward gaze angle, the more downward the gaze, and the more negative the value, the more upward the gaze.

[0048] The term eyelid opening corresponds to the space between the lower and upper eyelids when the upper eyelid is retracted. The wearer may have different eyelid openings. If the provided optical device does not fit the wearer due to a small eyelid opening, it is necessary to provide the wearer with a suitable device.

[0049] In a further aspect of the present disclosure, the device may take into account other body behaviors or postures of the wearer beyond the behavior and posture of the wearer's head in everyday life.

[0050] The optical device may have ophthalmic functionality, and the processing circuitry is further configured to receive an ophthalmic prescription of the wearer and take said ophthalmic prescription into account when comparing the received and stored head data with the head data pattern.

[0051] The term processing circuit may be any electronic control unit such as a microcontroller.

[0052] The term "ophthalmic prescription" corresponds to a prescription provided by an ophthalmologist to an optical device after the wearer has worn the optical device to correct a functional defect of the wearer. An ophthalmic function is a corrective function performed by the optical device.

[0053] Depending on the ophthalmic capabilities of the optical device and the monitored head data, the optical device may appear to be improperly used by the wearer or may appear to be ill-fitting to the wearer when performing certain activities, such as reading.

[0054] The processing circuitry is further configured to alert the wearer or an eye care professional that a mismatch of the optical device has been detected.

[0055] The optical device may be provided with a warning device. The warning may take various forms, such as a mechanical vibration, a sound, etc. The warning signal makes it possible to inform the wearer of any indication of an incompatibility of the optical device.

[0056] Thus, the warning system can quickly inform the wearer and / or eye care professional when the wearer tries on a new pair of ophthalmic lenses that the optical device is not suitable for the wearer.

[0057] Another embodiment of the present disclosure is to inform the wearer and / or an eye care professional that the optical device provided to the wearer does not address the wearer's current visual impairment and that the wearer needs a new optical device with new ophthalmic features to continue correcting the visual impairment.

[0058] The processing circuitry may be further configured to determine a cause of mismatch of the optical device by comparing the received head data to different sets of stored head data having head distribution patterns associated with the cause of the mismatch.

[0059] The optical device may be provided with a data storage medium, the data storage medium including different sets of stored head data patterns associated with causes of mismatch, and the processing circuitry compares the monitored head data of the wearer, which may also be stored on the data storage medium.

[0060] During the comparison, the processing circuitry controls whether there is a likelihood of the comparison of the head data pattern monitored by the wearer with the head data pattern associated with the cause of the mismatch exceeding a given threshold. It should be understood that the likelihood may correspond to a percentage of similarity between the measured head data and the stored head data pattern associated with the cause of the mismatch. If the threshold is exceeded, the processing circuitry determines that the cause of the mismatch of the optical device corresponds to a cause of mismatch of a similar stored head data pattern associated with the cause of the mismatch.

[0061] The cause of the non-conformance may also be determined using a machine learning model that was previously built using previously acquired data.

[0062] The cause of the mismatch may be selected from the list consisting of: the optical function of the optical device, the attachment parameters of the optical device, the use of the optical device by the wearer, the wearer's ophthalmic prescription, and the integrity of the optical device.

[0063] In the sense of the present disclosure, an optical function corresponds to a function that provides, for each direction of gaze, the effect of the optical lens on the light rays passing through the optical lens.

[0064] Optical functions may include refractive functions, light absorption, spectral functions, diffractive functions, scattering properties, polarization capabilities, contrast enhancement or reduction, or any other relevant parameters.

[0065] The refractive function corresponds to the refractive power (mean power, astigmatism, etc.) of an optical lens as a function of gaze direction.

[0066] The expression "optical design" is a widely used expression known by those skilled in the art of ophthalmology to specify a set of parameters making it possible to define the refractive function of an ophthalmic lens, and each ophthalmic lens designer has his own design, especially for progressive ophthalmic lenses. As an example, the "design" of a progressive ophthalmic lens results from the optimization of the progressive surface so as not only to restore the ability of a presbyopic person to see clearly at all distances, but also to optimally respect all physiological visual functions such as foveal vision, exofoved vision, binocular vision, etc., and to minimize undesirable astigmatism. For example, a progressive lens design may be - a refractive power profile along the main direction of gaze (meridian) used by the lens wearer during activities of daily living; the distribution of the refractive power (mean power, astigmatism, ...) on the side of the lens, i.e. away from the main direction of gaze, Includes.

[0067] These optical properties are part of the "design" defined and calculated by the ophthalmic lens designer and provided by the progressive lens.

[0068] The "optical design" of progressive lenses is subjected to rigorous testing before being commercialized.

[0069] The term mounting parameters refers to the position of the two optical lenses within the frame.

[0070] When a wearer purchases a new optical device, the mechanical aspects of the eyeglasses in the frame may not fit the wearer's head.

[0071] Another embodiment of the present disclosure relates to the mechanical aspects of eyeglass frames. The morphology of a wearer's head may grow over time. Changes in the shape of the wearer's head or changes in skin properties may induce displacement of the position of the optical elements. The mechanical structure of eyeglass frames may also change over time, for example due to wear, improper use, or storage, which may cause the position of the eyeglass frames on the wearer's face to change.

[0072] The wearer's use of the optical device corresponds to the way the wearer can use the optical device. In the case of multifocal or progressive lenses, the wearer may use a single portion of the area of ​​the optical lens. The wearer may always look through the distance vision portion of the progressive lens, even in near vision situations, for example, if the wearer does not lower their gaze direction. The wearer then compensates for this insufficient lowering of their gaze direction by further lowering their head. Based on the head data monitored during use of the optical device, the processing circuitry can determine abnormal head data when using the optical device.

[0073] Another cause of incompatibility may be the integrity of the optical device, such as twisting of the temples of the bridge between the optical lenses. Any deformation of the optical device may result in an altered optical function provided by the optical device, which may no longer be adapted to the wearer's visual impairment.

[0074] The optical device has a progressive addition power, and the head data is related to the distribution of the wearer's head lowering angle when using the optical device. As mentioned above, a wearer who has insufficient downward movement of the gaze direction of the eyes can compensate for this insufficiency by lowering their head further while using an optical element, for example, a progressive lens for near vision. The abnormal head data related to the wearer's further lowering of their head is data that makes it possible to determine the incompatibility of the optical device to the wearer.

[0075] The apparatus for detecting a misfit may comprise a head data sensor or sensors, a processing circuit, and a data storage medium. The apparatus may be removably attached to a portion of the optical device, for example, to a temple.

[0076] In an alternative embodiment, the device for detecting the misfit can be built into the eyeglass frame. The calculations can also be performed outside the optical device, in a smartphone application or computer or cloud server to which the head data is sent.

[0077] A second aspect of the present disclosure relates to a method performed by a processing circuit of an apparatus, the method comprising: - receiving and storing head data relating to the wearer's head over time when wearing and using the optical device; - processing head data based on head data patterns associated with known mismatches of an optical device to a wearer of said optical device; - detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch; Includes.

[0078] FIG. 3 is a flow chart illustrating various steps corresponding to a method performed by a processing circuit.

[0079] The method includes a receiving step S20 in which the processing circuit receives head data of the wearer measured by at least one head data sensor, and stores the measured head data in a data storage medium.

[0080] The method further includes a processing step S22 in which the processing circuitry proceeds to compare the measured wearer head data with predetermined head data patterns associated with causes of misfit of the optical device to the wearer.

[0081] Finally, the method includes a detection step S24 in which the processing circuitry proceeds to detect a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch.

[0082] The term "match" should be understood as if there is a likelihood above a given threshold between the head data monitored by the wearer and a head data pattern associated with the cause of the mismatch. It should be understood that the likelihood may correspond to a percentage of similarity between the measured head data and a stored head data pattern associated with the cause of the mismatch. Such a match can be determined using machine learning methods.

[0083] The threshold may correspond to, for example, a similarity between the measured head data pattern and a predetermined head data pattern above a given percentage, and is preferably determined using a machine learning method that automatically determines the most likely status (e.g., match / no match) based on the input head data.

[0084] The method can include an additional warning step, during which the eye care professional and / or the wearer are alerted to the incompatibility of the optical device during daily activities. The warning can also inform the eye care professional that the optical device is not suitable for the wearer. The eye care professional can then check the prescribed interpupillary distance, prescription, frame shape, lens design, or other parameters related to the optical device.

[0085] The warning can be used to check whether the wearer's posture is approaching the compatible posture over time, which is why the wearer's use of the optical device needs to be monitored over time and in their everyday life environment, rather than in a stable environment that may not correspond to the wearer's everyday life environment.

[0086] This method can take into account the number of modes in the distribution of the wearer's head tilt angles when using the optical device.

[0087] The head data sensor records the frame orientation or head down angle over time. The time of data acquisition can be predefined. The recorded frame orientation or head down angle is then transmitted for processing.

[0088] Some of the measurement data may be deleted before processing for several reasons: -If the measurement data is considered an outlier, - if the measurement data is taken outside of certain times, e.g. does not correspond to working hours, - for measurement data that does not correspond to the completion of a specific activity, and - for measurement data corresponding to moments when the wearer was not wearing the optical device correctly.

[0089] Other parameters may be considered to eliminate the measurement data.

[0090] An example can be considered of a wearer who uses an optical device while at work. Figure 4 corresponds to a histogram of head descent angles from Monday to Friday. Figure 5 corresponds to a histogram of head descent angles from Saturday and Sunday. In the following example, the wearer is considered to work from 9:00 AM to 6:00 PM from Monday to Friday and not work on weekends. In this way, the measurement data corresponding to Figure 5 is deleted and the period considered in Figure 4 corresponds to data related to head descent angles measured from 9:00 AM to 6:00 PM.

[0091] The mode corresponds to a peak where the occurrence of head down angles is more important while accomplishing a particular activity, such as working in front of a computer.

[0092] If the distribution of head tilt angles can be traced to a curve similar to a Gaussian distribution, the histogram is unimodal. If the histogram is unimodal, the optical device is well-suited to the wearer. Figure 6 shows a unimodal histogram corresponding to the wearer's head tilt angle while wearing the optical device.

[0093] The distribution of head tilt angles may include multiple peaks. This distribution is considered a multimodal histogram. Figure 7 shows a bimodal histogram corresponding to the wearer's head tilt angles while wearing the optical device. Figure 7 shows that there is not one specific range of head tilt angles, but two ranges.

[0094] A bimodal histogram indicates a wearer's inability to adapt to the optical device. Wearers with multifocal or progressive lenses may try to compensate for their inability to lower their gaze direction by further lowering their head angle.

[0095] Thus, one of the two peaks shown may correspond to anomalous head data, such as the additional head drop angle mentioned in the example above. In particular, a wearer equipped with progressive lenses and having a bimodal histogram has an optical device that does not suit them.

[0096] To use progressive lenses properly, the wearer of progressive lenses must learn to lower their eyes further than they would naturally lower their gaze when performing intermediate and near vision tasks. At the same time, the wearer must lower their head less than they would naturally lower their head when performing these tasks.

[0097] Wearers with bimodal histograms have not yet adopted the visual strategies required for progressive lenses. These progressive lens wearers tend to view near and intermediate distance objects through the distance vision zone and do not experience the appropriate add power required for the task. This results in poor fit and wearer discomfort.

[0098] Following an optional warning signal, progressive lens wearers who are not fitted with progressive lenses may be recommended to perform specific vision training for using progressive lenses.

[0099] A third aspect of the present disclosure relates to a method for determining a head pattern, the method comprising: receiving and storing wearer data associated with a wearer of the optical device; receiving and storing optical device data relating to an optical device associated with each wearer; receiving and storing longitudinal head data associated with each wearer's head when wearing and using an associated optical device; receiving and storing fit data related to the fit of an optical device associated with each wearer; - processing the received and stored data for a number of wearers to determine at least one head pattern associated with a misfit of the optical device; Includes.

[0100] FIG. 8 shows a flow chart illustrating the various steps of a method for determining a head pattern.

[0101] The method includes a first receiving step S30, in which wearer data relating to a wearer of the optical device is received and stored.

[0102] The method includes a second receiving step S32 in which optical device data relating to the optical device associated with each wearer is received and stored.

[0103] The method includes a third receiving step S34, in which head data relating to the head of each wearer when wearing and using the associated optical device is received and stored over time.

[0104] Furthermore, the method includes a fourth receiving step S36, in which fitting data relating to the fitting of the optical device associated with each wearer is received and stored.

[0105] Finally, the method includes a processing step S38 in which the received and stored data is processed to determine at least one head pattern associated with an optical device mismatch.

[0106] The head data according to the third aspect of the present disclosure may include head pose data, gaze direction, eyelid opening, and pupil size.

[0107] The wearer data may include the wearer's ophthalmic prescription, and the optical device has ophthalmic functionality.

[0108] The optical device data may include data relating to the optical functions of the optical device and / or the fitting parameters of the optical device and / or the use of the optical device by the wearer and / or the wearer's ophthalmic prescription.

[0109] Another method for determining at least one head pattern associated with a misfit of the optical device is to correlate monitored head data with wearer feedback.

[0110] The wearer can provide feedback regarding the worn optical device with a device for detecting poor fit of the optical device to the wearer. Feedback can be performed by completing a questionnaire related to the comfort of the optical device and / or the wearer's fit of the optical device in activities of daily living. Other sources of feedback are also possible, such as the use of a smartphone application. The wearer can use the application to notify the wearer of any discomfort or poor fit when using the optical device.

[0111] The new non-compatible posture pattern can be determined according to the feedback provided by the wearer and the monitoring data acquired by the at least one head data sensor to determine a posture data pattern that correlates with a fit level of the optical device to the wearer. The posture data pattern can be, for example, a head data posture pattern.

[0112] According to one embodiment, the pose data patterns are determined using a machine learning model that is used to find correlations between head data (and other optional data) and mismatches.

[0113] The head data may consist of multiple time series of the following measurements: - The pitch angle of the frame, measured by an IMU embedded in the frame, -Time derivative of the frame's yaw angle, as measured by an IMU embedded in the frame. These two time series are acquired for a large number of n wearers (e.g., i=1 to 10,000).

[0114] For each time series, - the average value of measurements over time, - standard deviation of measurements over time, - amplitude of the measurement over time (max-min), A set of descriptive statistics is calculated, including

[0115] According to one embodiment, a histogram of the measurements can further be constructed, from which the center value of the main peak and the width of the main peak can be estimated.

[0116] Descriptive statistics are grouped into data vectors summarizing the head data. There is one such vector per wearer. Such vectors are denoted by x_i.

[0117] The data related to the fit of the optical device associated with each wearer consists of a single number representing the degree of fit, ranging from 1 (poor fit) to 10 (perfect fit). There is one value per wearer. Such values ​​are denoted v_i.

[0118] A machine learning model is initiated where the input data is the data vector for each wearer and the data to predict is a numerical value representing the goodness of fit. The machine learning algorithm can be multiple linear regression, random forest, neural network, or other related algorithms.

[0119] The model is trained and tested using a dataset of {x_i} and {v_i}, relating vectors {x_i} to fitted values ​​{y_i}: Model M(x_i)→v_i

[0120] If the trained model provides good predictions, it can be used to define head data patterns associated with optical device mismatches.

[0121] The threshold T is defined based on the degree of fit. A wearer can be considered unfit if: v_i≦T

[0122] A head pattern x_non-adapt is associated with a non-adapt if, for this head data, the model predicts a value below the following threshold: M(x_non-adapt)=v_non-adapt≦T

[0123] In this example, we do not use wearer data or optical device data, which can be easily added to the data vector x_i.

[0124] Other data may be monitored and associated with the head data, such as, for example, the activity performed by the wearer when the head data is monitored. Indeed, a head pattern may be associated with a misfit when the wearer is performing a given activity and not associated with a misfit or associated with a different misfit when the wearer is performing another activity. For example, a posture pattern may be associated with a misfit when the wearer is walking but not associated with a misfit when the wearer is working at a desk. The activity performed by the wearer may be determined based on the wearer's location, e.g., GPS location, or based directly on input from the wearer.

[0125] The device can be improved for detecting mismatches over time by acquiring new mismatch posture data patterns, and new causes of mismatches can be determined through wearer feedback.

[0126] According to the different histograms and / or feedback, the wearers are segmented into various groups, such as their ability to perform a given task depending on the type of optical lens they wear and / or their suitability to use the optical device while performing a specific task, for example, reading or working in front of a computer. The optical device may be a single vision lens, a wide progressive lens, or a narrow progressive lens. In a preferred embodiment, the segmentation is performed using machine learning methods, where a model is trained to determine the segment to which the wearer belongs, and the segment may be defined by the cause of the incompatibility based on head data and other optional data.

[0127] Thanks to the wearer segmentation, the misfit detection method is able to predict misfit to progressive lenses.

[0128] A fourth aspect of the present disclosure relates to an alternative method for detecting mismatches in optical devices.

[0129] A fifth aspect of the present disclosure relates to a non-transitory computer-readable storage medium.

[0130] The non-transitory computer-readable storage medium may include computer-executable instructions that, when executed by a computer or processing circuit, cause the computer or processing circuit to perform a method for detecting a mismatch and / or a method for determining a head pattern associated with a mismatch.

[0131] The method includes a first receiving step S40, in which data relating to the wearer, such as prescription, gender and / or age, is received and stored in a database.

[0132] The method includes a second receiving step S42, in which data related to the optical device is received and stored in a database. The data related to the optical device may include optical lens design, AR coating, anti-fouling, and / or any other parameters related to the optical device.

[0133] The method includes a third receiving step S44, in which data measured by the at least one head data sensor is stored in a database, The measured data may include head movement, posture and descent angle, light intensity, distance to an object, and / or whether the wearer is wearing the optical device over time.

[0134] The method includes a fourth receiving step S46, in which feedback related to the non-conformance is received and stored in a database. The feedback may be provided by a smartphone application or an online survey.

[0135] The method may include a fifth and final receiving step S48 in which the cause of the non-conformance is received and stored in a database.

[0136] The data associated with the database may be provided by actual wearers through surveys and / or feedback via a smartphone application or via an eye care professional.

[0137] The data associated with the database may also be provided based on simulated virtual wearers using avatars with incorrect prescriptions, improperly fitted optical devices, and the head poses and / or head data for various visual tasks may be calculated.

[0138] The database may contain a combination of data relating to real and virtual wearers.

[0139] The method includes a first analysis step S50, in which data stored in a database is analyzed to determine head data patterns that are characteristic of a given case of non-match.

[0140] The method further includes a model generation step S52, where a model is generated to predict potential non-conformances and causes of non-conformances based on the database. The generation of the model can be achieved by machine learning, for example using neural networks.

[0141] The learning phase of the machine learning can obtain feedback from the wearer to improve convergence to potential mismatches and their causes. Improvement can be provided using reinforcement learning methods.

[0142] The method includes a second analysis step S54 in which data provided by the head data sensor is analyzed for a new wearer wearing an optical device having an apparatus for detecting misfit according to the present disclosure.

[0143] The method includes a warning step S56, in which a warning device of the apparatus according to the present disclosure warns the wearer or an eye care professional if the data measured by the head data sensor corresponds to an incompatible pattern.

[0144] If the cause of the non-conformance can be identified from the head data pattern, this information can be added to the database.

[0145] The method includes an optical device provision step S58, where if the tested optical device does not fit the wearer, a new optical device is proposed to the wearer, taking into account the cause of the non-fit and the data stored in the database.

[0146] Alternatively, the method may suggest realignment or vision training of the tested optical device based on the cause of the non-conformance.

[0147] Reconditioning of an optical device may involve changing the optical lens or the frame of the optical device for a different design.

[0148] The training may be specific training to improve the wearer's fit to a particular optical device, such as a progressive lens.

[0149] The method may be performed by a processing circuit or may be processed by a device external to an apparatus according to the present disclosure, for example in a cloud network.

[0150] The database may contain additional data obtained, for example, by optometric tests, such as convergence insufficiency, adaptation delay, interocular disparity, longer head posture or gaze stabilization provided by at least some of the real and / or virtual wearers.

[0151] The head data sensor may be a type of sensor other than an IMU, such as an eye tracker that measures gaze direction, which may be integrated into the frame of the optical device.

[0152] The head data sensor may measure pupil size or facial expression, including for example eyebrow position. The head data sensor for measuring pupil size or facial expression may be an integrated camera.

[0153] The head data sensor can measure EEG (electroencephalography) data.

[0154] Head data sensors can be used to determine the time the optical device is worn and the time the optical device is not worn, and optionally the type of task the wearer performs. Thanks to this parameter, incompatibility can be detected. A wearer may not wear the optical device frequently due to discomfort caused by the shape of the optical device once worn. Shape issues can be caused by the size of the frame at temple level or the shape and / or size of the nose pads. Based on these data, the eye care professional can be alerted that the optical device needs to be readjusted.

[0155] This disclosure relates to head data. However, head data may be replaced by other types of data related to the wearer's body. The data may, for example, relate to the posture or movement of a part of the wearer's body other than the head.

[0156] Many further modifications and variations will be apparent to those skilled in the art upon reference to the above exemplary embodiments, which are provided for illustrative purposes only and are not intended to limit the scope of the present disclosure, which is determined solely by the appended claims. In particular, the present disclosure can be applied to active lenses, the optical design of which can be modified to improve the fit of the wearer. The present disclosure can also be applied to active filters, whose light transmission function can be adapted to take into account the wearer's comfort.

[0157] In the claims, the word "comprise" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The fact that different features are recited in mutually different dependent claims does not indicate that a combination of these features cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope of the present disclosure.

Claims

1. 1. An apparatus for detecting misfit of an optical device on a wearer, comprising: - receiving and storing head data relating to the wearer's head over time when the optical device is worn and used, - processing head data based on head data patterns associated with known mismatches of an optical device with the wearer of the optical device; - detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch; The apparatus comprises a processing circuit configured to:

2. The apparatus of claim 1 , wherein the head data includes at least one of head pose and movement data, gaze direction, eyelid opening, pupil size, EEG data, and facial expression.

3. 3. The apparatus of claim 1, wherein the optical device has an ophthalmic function, and the processing circuitry is further configured to receive an ophthalmic prescription of the wearer and take the ophthalmic prescription into account when comparing the received and stored head data with a head data pattern.

4. 4. The apparatus of claim 1, wherein the processing circuitry is further configured to determine a cause of the mismatch of the optical device by comparing the received different sets of stored head data with a head distribution pattern associated with the cause of the mismatch.

5. 5. The apparatus according to claim 1, wherein the optical device has progressive addition power and the head data relates to a distribution of head depression angles of the wearer when using the optical device.

6. A method performed by a processing circuit of the device, comprising: - receiving and storing head data relating to the wearer's head over time when the optical device is worn and used; - processing the head data based on head data patterns associated with known mismatches of an optical device to the wearer of the optical device; - detecting a mismatch of the optical device with the wearer by matching the received and stored head data with a head data pattern associated with the mismatch; A method comprising:

7. The method of claim 6 , wherein the head data includes at least one of head pose and movement data, gaze direction, eyelid opening, pupil size, EEG data, and facial expression.

8. 8. The method of claim 6 or 7, wherein the optical device has an ophthalmic function, and the method further comprises receiving an ophthalmic prescription of the wearer, and taking the ophthalmic prescription into account when comparing the received and stored head data with a head data pattern.

9. 9. The method of claim 6, further comprising determining a cause of the incompatibility of the optical device by comparing the received and stored head data with a head data pattern.

10. 10. The method according to any one of claims 6 to 9, wherein the optical device has progressive addition power and the head data relates to a distribution of head depression angles of the wearer when using the optical device.

11. 11. The method of claim 10, wherein detecting a misfit comprises determining whether a distribution of head down angles is bimodal, a bimodal histogram indicating that the wearer has failed to fit the optical device.

12. 1. A method for determining a head pattern, comprising: - receiving and storing wearer data relating to a wearer of the optical device; - receiving and storing optical device data relating to said optical devices associated with each wearer; - receiving and storing longitudinal head data associated with each wearer's head when wearing and using said associated optical device; - receiving and storing fitting data relating to the fitting of said optical device associated with each wearer; - processing the received and stored data across multiple wearers to determine at least one head pattern associated with a misfit of the optical device using a machine learning model to find correlations between head data and misfits; A method comprising:

13. The method of claim 12 , wherein the head data includes at least one of head pose data, gaze direction, eyelid opening, pupil size, EEG data, and facial expression.

14. A non-transitory computer-readable storage medium containing computer-executable instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 6 to 11.

15. A non-transitory computer-readable storage medium containing computer-executable instructions that, when executed by a computer, cause the computer to perform the method of claim 12 or 13.

Images