Methods and devices for evaluating the visual discomfort

Abstract

The disclosure relates to a device (100) for evaluating the visual discomfort of a person (1) comprising at least one neuro-physiological sensor (5) and at least a processing unit configured to determine, with a trained supervised machine learning model, the visual discomfort of the person (1).

Description

METHODS AND DEVICES FOR EVALUATING THE VISUAL DISCOMFORTFIELD OF THE DISCLOSURE

[0001] The present disclosure relates to methods / devices for evaluating and alleviating the visual discomfort of a person, such as an eye fatigue or an eye dry.BACKGROUND OF THE DISCLOSURE

[0002] As electronic products become increasingly prevalent in modern life, people are spending extended periods gazing at digital monitors, whether for work or entertainment. Despite the many benefits of such technologies, prolonged exposure to digital devices can lead to visual fatigue, i.e., the prolonged strain on human’s visual organs, which surpasses their compensatory capabilities, resulting in a cluster of ocular and systemic symptoms (Simmerman, H. 1950. “Visual Fatigue,” Optometry and Vision Science (27:11), pp. 554-561). Extensive research has linked visual fatigue to health impairment, decreased work-related efficiency and effectiveness (Hoffman, D. M., Girshick, A. R., Akeley, K., and Banks, M. S. 2008. “Vergence-Accommodation Conflicts Hinder Visual Performance and Cause Visual Fatigue,” Journal of Vision (8:3), pp. 33-33), and even increased risks of accidents (Bando, T, lijima, A., and Yano, S. 2012. “Visual Fatigue Caused by Stereoscopic Images and the Search for the Requirement to Prevent Them: A Review,” Displays (33:2), pp. 76-83). Therefore, there is a growing need from both academics and industry to achieve precise detection and timely treatment of visual fatigue.

[0003] Traditional methods for diagnosing visual fatigue can be classified as subjective measures and objective indicators. Subjective assessment is usually performed through self-report symptom questionnaires, while objective assessment relies on optometric measurements. Although subjective assessment scales like the Simulator Sickness Questionnaire (SSQ) (Kennedy, R. S., Lane, N. E., Berbaum, K. S., and Lilienthal, M. G. 1993. “Simulator Sickness Questionnaire: An Enhanced Method for Quantifying Simulator Sickness,” The International Journal of Aviation Psychology (3:3), pp. 203-220) and the Visual Fatigue Scale (VFS) (Heuer, H., Hollendiek, G., Kroger, H., and Romer, T. 1989. “Rest Position of the Eyes and Its Effect on Viewing Distance and Visual Fatigue in Computer Display Work,” Zeitschrift Fur Experimentelle Und Angewandte Psychologie (36:4), pp. 538-566) have been widely accepted, such measures can only be applied post hoc and require the conscious effort of the tested subject.

[0004] On the other hand, optometric examinations can provide highly precise visual fatigue diagnosis, but come with the requirement for well-trained professionals, clinical- grade equipment, and highly restricted conditions.

[0005] In other words, subjective assessment can only be achieved with the consciousness of the person assessed, while objective measurement is costly and resource-intensive. Neither of them is suitable for real-time detection.SUMMARY OF THE DISCLOSURE

[0006] The present disclosure aims to propose a more objective solution for predicting, or instantly evaluating a person’s level of visual discomfort.

[0007] According to a further objective, the present disclosure aims to propose a solution that may allow the evaluation of visual discomfort in real-time, such as an occurrence of eye fatigue or eye dry.

[0008] Hence, the object of the disclosure relates to a method of training a supervised machine learning models (e.g. SVM, neural networks, convolutional networks) for determining a level of the visual discomfort of a person equipped with at least one neurophysiological sensor, carried out by a computer, characterised in that it comprises the following steps: a - obtaining measured responses of said at least one neuro-physiological sensor as function of several degrees of visual discomfort for persons and several visual stimuli so as to form a database; b - training the supervised machine learning model with the database to predict the level of the visual discomfort of a person.

[0009] Advantageously, the method according to the disclosure allows training the supervised machine learning model as an artificial intelligence so that the latter will incrementally improve itself so as to enrich the database to evaluate better the level of visual discomfort of various persons, i.e., notably as a function of her / his ametropia or conversely as a function of her / his emmetropia.

[0010] The object of the disclosure relates also to a supervised machine learning model obtained by the training method as described above, characterised in that it is configured to determine a level of the visual discomfort of a person with measured responses of at least one neuro-physiological sensor applied to the person who is stimulated with several visual stimuli.

[0011] Advantageously, the supervised machine learning model according to the disclosure is an artificial intelligence that will incrementally improve itself and enrich the database to evaluate better the level of visual discomfort of various persons, i.e., notably as a function of her / his ametropia or conversely as a function of her / his emmetropia.

[0012] The disclosure may also include one or more of the following optional features, taken alone or in combination.

[0013] The supervised machine learning model may comprise a database including measured responses of said at least one neuro-physiological sensor as a function of several degrees of visual discomfort for persons and several visual stimuli.

[0014] The database may also include tracked facial emotions of persons and / or at least one optical element reducing the visual discomfort of persons.

[0015] Another object of the disclosure relates to a method for evaluating a level of visual discomfort of a person equipped with at least one neuro-physiological sensor, carried out by a computer, characterised in that it comprises the following steps: a - stimulating the eyes of the person with changing visual stimuli which generate several degrees of visual discomfort for the person, b - measuring the responses of the person with regard to the changing visual stimuli by said at least one neuro-physiological sensor, c - evaluating the measured responses by the supervised machine learning model as described above for determining a level of the visual discomfort of the person.

[0016] Advantageously, the method according to the disclosure allows a more objective solution to instant evaluate the level of visual discomfort of the person. In other words, the method can be carried out quickly due to the trained supervised machine learning model, do not require a well-trained professional and can be performed in various devices (clinical dedicated device, commercial dedicated device, eyewear, etc.).

[0017] The method can thus be applied for testing occasionally the level of visual discomfort of a person (one particular testing) or for monitoring the level of visual discomfort of a person (permanent or prolonged testing by wearing an instrumented eyewear for example).

[0018] The disclosure may also include one or more of the following optional features, taken alone or in combination.

[0019] Said changing visual stimuli of step a may be obtained by the tasks of the person’s everyday life. In other words, notably in a monitoring application with for example an integrated portable eyewear, the method can warn about a risk of visual discomfort maybe even before the person feels or has consciousness of it.

[0020] In case of an occasional test, the step a may be configured to stimulate convergence and accommodation of the eyes of the person. As such, it can comprise at least a close and a far distance conditions of a display and / or sequence of interposing different lenses between the person’s eyes and a display.

[0021] The method may comprise, before step a, a step of establishing an optometric profile of the person, the method being configured to consider the optometric profile of the person. Advantageously, the method according to the disclosure may notablyconsider an ametropia or conversely an emmetropia of the person to adapt its steps such as in a case of a dedicated step a of an occasional test.

[0022] The object of the disclosure relates additionally to a method for decreasing visual discomfort of a person comprising the following steps:- performing a method for evaluating a level of visual discomfort of a person as described above,- selecting on the basis of the result of the evaluation method of visual discomfort for the person in a database for each eye, an optical element reducing visual discomfort of the person.

[0023] Advantageously, the method according to the disclosure allows, after a more objective solution to evaluate the level of visual discomfort of the person, to propose one (or more) dedicated optical element(s). Here again, the method can be carried out quickly due to the trained supervised machine learning model, do not require a well-trained professional and can be performed in various devices (clinical dedicated device, commercial dedicated device, eyewear, etc.).

[0024] The method can thus be applied for occasionally determining optimal optical element for a person (one particular testing) or for determining optimal optical element for a person after a prolonged testing (by wearing an instrumented eyewear for example).

[0025] Another object of the disclosure relates to a device for evaluating the visual discomfort of a person comprising:- at least one neuro-physiological sensor, and- at least a processing unit configured to determine, with the supervised machine learning model as described above, the visual discomfort of the person.

[0026] Advantageously, the device according to the disclosure allows a more objective solution to instant evaluate the level of visual discomfort of the person. In other words, the device submits a quick evaluation with help of the trained supervised machine learning model, do not require a well-trained professional and can be included in various applications (clinical dedicated device, commercial dedicated device, eyewear, etc.).

[0027] The device can thus be applied for testing occasionally the level of visual discomfort of a person (one particular testing) or for monitoring the level of visual discomfort of a person (permanent or prolonged testing by wearing an instrumented portable device for example).

[0028] Alternatively, the device may be a Virtual Reality (VR) eyewear, Augmented Reality (AR) eyewear or Mixed Reality eyewear.

[0029] The disclosure may also include one or more of the following optional features, taken alone or in combination.

[0030] Said at least one neuro-physiological sensor may be at least one sensor of the group of sensors including: an eye tracker sensor, an electrodermal sensor, an electroencephalographic sensor, a respiratory electromyographical sensor.

[0031] The device may further comprise an interface to be manipulated by the person and configured to allow the person to express a self-evaluation, the supervised machine learning model being configured to consider such self-evaluations for customising the evaluation of the visual discomfort of the person.

[0032] The device may further comprise a web application, accessible on the web, configured to submit the evaluation of the visual discomfort of the person. In other words, the device may not include all components of the disclosure and be remotely connected to a non-on-board component (such as the supervised machine learning model) to perform the method according to the disclosure.

[0033] The object of the disclosure relates additionally to a device to decrease visual discomfort of a person comprising:- a device for evaluating the visual discomfort as described above,- at least a processing unit configured to select, with the supervised machine learning model as described above, at least one adapted optical element reducing the visual discomfort of the person.

[0034] Advantageously, the device according to the disclosure allows a more accurate solution to propose one (or more) dedicated optical element(s), i.e., notably by considering a more objective evaluation of the level of visual discomfort of the person. Here again, the device submits a quick determination with help of the trained supervised machine learning model, do not require a well-trained professional and can be included in various applications (clinical dedicated device, commercial dedicated device, eyewear, etc.).

[0035] The device can thus be applied for occasionally determining optimal optical element for a person (one particular testing) or for determining optimal optical element for a person after a prolonged testing (by wearing an instrumented eyewear for example).

[0036] Another object of the disclosure relates to an eyewear for reducing the visual discomfort of a person, characterised in that it comprises said at least one adapted optical element selected by the device as described above. Alternatively, the eyewear may be a Virtual Reality (VR) eyewear, Augmented Reality (AR) eyewear or Mixed Reality eyewear.

[0037] Finally, the object of the disclosure relates to a computer program product or a computer readable storage medium comprising instructions which, when executed by a computer, causes the computer to carry out the steps of the method for evaluating a level of visual discomfort of a person or the steps of the method for decreasing visual discomfort of a person as described above.BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Other features and advantages of the present disclosure will appear more clearly upon reading the following detailed description, made with reference to the annexed drawings, provided as a non-limited description, in with:Figure 1 is in a side view a schematic example of a device for evaluating the visual discomfort of a person according to an embodiment;Figure 2 is a front view of a detail of Figure 1 ;Figure 3 is a schematic synoptic view of several components of the device for evaluating the visual discomfort of a person according to the embodiment of Figure 1 ;Figure 4 is an example of a flowchart of a method for evaluating the visual discomfort of a person according to an embodiment.DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] The embodiment(s) in the following description is(are) only to be considered as example(s). In the various Figures, the same or similar elements bear the same references, optionally added with an index. The description of their structure and their function is therefore not systematically restated.

[0040] In all the following, the orientations are the orientations of the Figures. In particular, the terms "upper", "lower", "left", "right", placed above, below, forwards and backwards in relation to the orientations of the Figures.

[0041] It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the disclosure described herein are capable of operation in other orientations than described or illustrated herein. In addition, a feature described in relationship with one embodiment may also concern another embodiment even if this is not mentioned expressively. Simple features of different embodiments may also be combined to provide further realizations.

[0042] It is to be noticed that the term “comprise(s)”, “include(s)”, “have”, “has”, “can”, “contain(s)”, and variant thereof, notably used in the claims, should not be interpreted as being restricted to the means listed thereafter. It does not exclude other elements or steps. It is thus to be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Thus, the scope of the expression “a device comprising means A and B” should not be limited to devices consisting only of components A and B. It means that with respect to the present disclosure, the only relevant components of the device are A and B.

[0043] The device and the method according to the present description can be implemented on one or several computer(s). In this context, unless specifically stated otherwise, it is appreciated that throughout this specification discussions utilising terms such as “computing”, “calculating” and “generating”, “treating” or the like, refer to the action and / or processes of a computer or computing system, or similar electronic computing device, that manipulate and / or transform data represented as physical, such as electronic, quantities within the computing system’s registers and / or memories into other data similarly represented as physical quantities within the computer system’s memories, registers or other such information storage, transmission or display devices.

[0044] A computer program product comprising one or more stored sequences of instruction that is accessible to a processor and which, when executed by the processor, causes the processor to carry out the steps of the method is also proposed.

[0045] Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD- ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), electrically programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEPROMs), magnetic or optical cards, or any other type of media suitable for storing electronic instructions, and capable of being coupled to a computer bus system. A computer-readable medium carrying one or more sequences of instructions of the computer program product is thus proposed. This enables to carry out the method in any location.

[0046] The processes and displays presented herein are not inherently related to any particular computer or other apparatus. Various general purpose systems may be used with programs in accordance with the teachings herein, or it may prove convenient to construct a more specialized apparatus to perform the desired method. The desired structure for a variety of these systems will appear from the description below. In addition, embodiments of the present disclosure are not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the disclosures as described herein.

[0047] Such a computer which is used as a processing and / or control unit, may comprise input channels for receiving signals for example from sensors or interfaces. It may also comprise output channels to output signals for example for controlling elements like a motor, a display, etc. It may comprise A / D (analogue / digital) and D / A (digital / analogue) modules.

[0048] By "visual discomfort” or “discomfort”, one understands any visual capacity reduction inducing punctual symptoms such as dry eye, visual fatigue, asthenopia, headache, eye pain, eye strain in or around the eye, occasional blur, burning or irritatedeye, sensitivity to bright light, occasional double vision, sickness, difficulty to focus or refocus at accommodate at different distances, watery eyes, visual attention, sore, tired, burning or itching eyes, concentrating difficulty. For example, a journey map of visual fatigue and dry eye is a cumulative representation of visual discomforts tagged in time.

[0049] Figure 1 shows in a side profile a person 1 whose head is maintained in a fixed position by a chinrest 3. The person 1 may be an adult or a child (male orfemale). She / He may be a person 1 wearing specific eyeglasses or not. In the following, it is assumed that the person 1 is an adult not wearing eyeglasses at all, but the following description also applies in the same way to wearer of eyeglasses, goggles, masks, etc.

[0050] Figure 1 shows also a device 100 (see also Figure 3) for evaluating the visual discomfort of the person 1. The device 100 comprises at least one neuro-physiological sensor 5 which is for example fixed onto the skin of the person 1. In order to improve precision, the use of several, i.e., at least two, neuro-physiological sensors is preferred. Alternatively, the device 100 may be a Virtual Reality (VR) eyewear, Augmented Reality (AR) eyewear or Mixed Reality eyewear.

[0051] Such neuro-physiological sensors 5 may be within the following group of sensors: an eye tracker sensor, an electrodermal sensor, an electroencephalographic sensor, a respiratory electromyographical sensor. All these sensors are able to track a reaction from person 1 in response to optical / eye stimulations to which person 1 may be exposed.

[0052] The device 100 further comprises an eye stimulating machine 7 which is partially shown on Figure 1 and surrounded with a dashed line. The eye stimulating machine 7 is configured for stimulating the eyes of the person 1 with changing visual stimuli in particular for stimulating convergence and accommodation of the eyes of the person 1 .

[0053] In the exemplary embodiment of Figure 1 , the eye stimulating machine 7 comprises a movable display 9. This movable display 9 may comprise for example a screen with printed letters or other graphical patterns like drawings for example, i.e., static display. In another embodiment, it may comprise an electronic display like an LCD, LED or OLED screen or similar which displays letters or other graphical patterns, i.e., possibly changeable / dynamic display.

[0054] The movable display 9 may for example be fixed to a gear rack 11 which is engaged with a tooth wheel 13 driven by an actuator 15 such an electrical motor with a toothed output axis. Thus, the movable display 9 can be displaced in a back and forth movement as shown by arrow 17 in Figure 1.

[0055] In particular, the movable display 9 is configured to be placed at predefined distances from the eyes of the person 1. More specifically, the movable display 9 is able to be placed at least at two distances corresponding respectively to a close distance condition (shown in Figure 1 with dashed lines) and a far distance condition (shown inFigure 1 with continuous lines). The distance between the person’s eyes and the screen was for example set to either 0.4 m (the close distance condition) or 2.0 m (the far distance condition).

[0056] In addition, the eye stimulating machine 7 comprises for each eye a series of different lenses. In the exemplary embodiment of Figures 1 and 2, the eye stimulating machine 7 comprises four different lenses LR1-LR4 for the right eye and four different lenses LL1-LL4 for the left eye. These lenses LR1 -LR4 and respectively LL1-LL4 have for example different optical powers. In a possible configuration, LL1 and LR1 , LL2 and LR2, LL3 and LR3, LL4 and LR4 are respectively identical lenses. The series are not limited to four lenses for each eye, but they may comprise a higher number (more than four such as six, eight, ten, twelve, sixteen, twenty lenses) or a smaller number (less than four such as three, two or one lens(es)). The number of levels of corrective lenses can also differ (notably as a function of the number of lenses) without departing from the disclosure. As a matter of example, the eye stimulating machine 7 may comprise sixteen lenses with four levels of corrective lenses (four different optical powers).

[0057] According to the exemplary embodiment of Figure 1 , at least one of the series of different lenses LR1-LR4, LL1-LL4 may comprise lenses of increasing optical powers, i.e., the optical powers of the lens series may respect the relation: LR1 <LR2<LR3<LR4 or LL1 >LL2>LL3>LL4.

[0058] Each optical power of the lenses LR1-LR4, LL1-LL4 may range from 1 to 45 dioptres. The lenses LR1-LR4, LL1-LL4 are for example arranged in ascending order based on their grades of optical power. As a matter of examples, lenses LR1 -LR4, LL1- LL4 can be lenses (with different optical powers with difference of 0,5 dioptre for example) of a conventional phoropter controlled by the device 100 or 104, or optical power of lenses LR1-LR4, LL1-LL4 can be modified with help of a conventional phoropter by interposing it between lenses LR1-LR4, LL1-LL4 and the person 1.

[0059] According to the example shown in Figure 2, all lenses LR1-LR4 and LL1-LL4 are placed in a lens support mechanism 19 configured to dispose one lens LR1-LR4, LL1- LL4 at a time between each eye of the person 1 and the movable display 9. On the one hand, lenses LR1-LR4 are placed in a movable rack 19R which may be driven up and down (with respect to a vertical direction such the gravity direction) by an actuator 21 R such as an electrical motor with a toothed output axis. On the other hand, lenses LL1- LL4 are placed in a movable rack 19L which may be driven up and down (with respect to a vertical direction such the gravity direction) by an actuator 21 L such as an electrical motor with a toothed output axis. The up and down movements of racks 19R and 19L are shown by arrows 23R and 23L in Figure 2. The actuators 21 R and 21 L are controlled independently so that racks 19R and 19L may move up and down independently fromeach other as it will be explained below. As a matter of example, lens LR1 may be in front of the right eye of the person 1 whereas lens LL3 may be in front of the left eye of the person 1 .

[0060] In the example of Figure 2, an opaque sheet or plate 25 having two holes 27 (only one is shown in Figure 1), one for each eye of the person 1 , is placed between the chinrest 3 and the lens support mechanism 19 so that person 1 can only watch with each eye, through its associated hole 27, only one lens LR1-LR4, LL1-LL4 at a time.

[0061] Figure 3 presents a synoptic scheme of device 100 for evaluation of visual discomfort.

[0062] In Figure 1 and 2, the actuators 15, 21 R and 21 L, as well as the neurophysiological sensor 5 are connected, for example, to a first processing and control unit 101 shown in Figure 3. The connexion may be wired or wireless. The connexion may be configured to allow transmission of power supply, control signals, power, measurement signals, in one or two ways.

[0063] In Figure 3 are also represented- a facial emotion tracker 40 for tracking facial emotions of the person 1 in response to the presented changing stimuli, comprising for example a camera,- an interface 42 which may be manipulated by the person 1 and configured to allow to the person 1 to express a self-evaluation, in particular the feeling of a visual discomfort in response to the presented changing stimuli. Interface 42 may comprise a touch screen, buttons, switches or a microphone with a voice recognition processor allowing person 1 to give feedback of her / his visual comfort.- an optometric profiler 44,- an eye tracker 46.

[0064] Other neuro-physiological sensors 5 may also be connected to processing and control unit 101.

[0065] Thus, the first processing and control unit 101 is configured for:- submitting various optical stimuli to the person 1 as described below, and- measuring the responses of person 1 to the optical stimuli by said at least one neuro-physiological sensor 5 and / or the facial emotion tracker 40, and also, if present, by any inputting data coming from interface 42 and / or the optometric profiler 44 and / or for the eye tracker 46.

[0066] Processing and control unit 101 may be realised by one machine like a computer or by several machines / computers.

[0067] The device 100 for evaluating the visual discomfort comprises furthermore a second processing unit 102. The second processing unit 102 is configured to receive inputs from the first processing and control unit 101. In particular, the second processingunit 102 is configured to receive the measurement results from neuro-physiological sensor(s) 5, the signals representing the facial emotions form the facial emotion tracker 40 in order to consider such tracked emotions for evaluating the visual discomfort of the person, and / or signals of the interface 42 to take into account any self-evaluation for evaluating the visual discomfort of the person 1. Of course, the first and the second units 101 , 102 may belong to the same computer or each to respectively two different connected computers.

[0068] The second processing unit 102 is specifically programmed with a trained supervised machine learning model {e.g. SVM, neural networks, convolutional networks) for predicting / evaluating / classifying the visual discomfort of the person 1 and is fed with the measured responses of the at least one neuro-physiological sensor 5 and inputted via the first processing and control unit 101.

[0069] The trained supervised machine learning model possesses for example one kernel among the following group: a linear kernel, a polynomial kernel, a Gaussian radial basis function kernel or a Sigmoid kernel.

[0070] Through the above mentioned processing, datasets may be structured with each row containing m physiological features (corresponding to the measurements) and one output class. One dataset will be used for training supervised machine learning models to achieve a binomial classification between the normal or visual discomfort state with the following regression formula:In which:Y refers to the visual discomfort class, coded as -1 or 1 ; xt represents the m-length physiological feature vector; f is the trained SVM model ( / .e., decision function).

[0071] Of course, others methods may be added or interchanged with respect to the hereinbefore binomial classification such as a leaning multiple regression (as a function of responses (cl ear / b lurry) in view of a stimulus change: the model will be trained by matching responses with neuro-physiological data) and / or a learning random forest (as a function of responses in view of different stimuli: such as convergence rock (answer: single / double) and / or accommodative rock (answer: clear / blurry) and / or vergence (answer: single / blurred / double), the model will be trained depending on the responses to each stimulation, a decision tree will be iterated as a function of visual discomfort profiles and neuro-physiological signals).

[0072] The development of diagnosis machine learning includes training a model in the training dataset and then estimating prediction performance in another dataset. Thispredictive approach allows instant diagnosis with high precision and uses two separate datasets, one for training and another for testing.

[0073] The output of the supervised machine learning model comprises a state corresponding to the presence or the absence of a visual discomfort. Thus, the results can be used in order to choose an adapted optical element to achieve reduction or even suppression of visual discomfort. In order to do this, the device 100 for evaluating the visual discomfort can be completed with a database 103 of optical elements in order to form a device 104 for decreasing visual discomfort of a person 1 .

[0074] Processing unit 102 can be then configured to select for the person 1 in said database 103 for each eye an optical element reducing discomfort and visual discomfort. Alternatively, an additional processing unit can be specifically used for selecting of the optical elements reducing discomfort.

[0075] The device 100 for evaluating the visual discomfort can perform a method according to the disclosure comprising the main steps 200, 202 and 204. In first step 200, person 1 is equipped with at least one (preferentially several) neuro-physiological sensor(s) 5. Then, in the second step 202, the eyes of the person 1 are stimulated with the eye stimulating machine 7. Finally, in the third step 204, the measured responses are computed in the second processing unit 102 for evaluating the visual discomfort of the person 1 . Of course, an optometric profile of the person 1 can be established before step 200.

[0076] In particular, the eye stimulating machine 7 is configured to present to person 1 changing visual stimuli in particular for stimulating convergence and accommodation of the eyes of the person. For doing this, the lenses LR1-LR4 and LL1-LL4 which are disposed respectively in their movable racks 19R / 19L, are independently driven via actuators 21 R and 21 L to predetermined positions, as well as the distance of display 9 with respect to the persons eyes by actuator 15.

[0077] Thus, stimulating the eyes of the person 1 with changing visual stimuli in particular for stimulating convergence and accommodation of the eyes of the person, comprises to locate moveable display 9 at a close distance condition and / or at a far distance condition. The close and the far distance conditions are realised for example by disposing, via actuator 15, the moveable display 9 at above mentioned predefined distances from the eyes of the person 1 .

[0078] Another stimulation of the eyes of the person 1 with changing visual stimuli in particular for stimulating convergence and accommodation of the eyes of the person may comprise interposing a sequence of different lenses LL1 -LL4 / LR1-LR4 having different optical powers, between the person 1 eyes and the moveable display 9. One possible sequence may comprise disposing lenses LL1 -LL4 / LR1 -LR4 of increasing or decreasingdifferent optical powers between the person 1 eyes and the moveable display 9. Another possible sequence may comprise disposing lenses LL1-LL4 / LR1-LR4 of randomly selected different optical powers between the person 1 eyes and the moveable display 9.

[0079] According to a sequence, two identical lenses, for example LL1 and LR1 , one lens for each eye, are disposed between the person 1 eyes and the moveable display 9. According to another sequence, two different lenses, i.e., of different optical powers, for example LL1 and LR3, one lens for each eye, are disposed between the person 1 eyes and the moveable display 9.

[0080] Following a protocol of regular optometry tests, person 1 is for example instructed to place his chin on the chinrest 3 ensuring her / his pupils to be aligned to the centre of the two lenses with the lowest dioptre, for example LR1 and LL1.

[0081] The processing and control unit 101 is then for example programmed to randomly select either the close or far distance condition by controlling actuator 15 to move the moveable display 9 to the chosen place. Once the visual eye discomfort device 100 is initialised, processing and control unit 101 may controls notably actuators 21 L and 21 R to submit person 1 randomly selected visual stimuli.

[0082] Further to the measurements, person 1 can report, via interface 42 for example, if presented information appear as clear, blurry or double. The processing and control unit 101 is programmed to scroll up to align the person’s eyes to lenses with one higher grade of dioptres, to select other visual stimuli and to repeat the procedure till the person 1 reports double vision.

[0083] The processing and control unit 101 may be programmed to present stimulus conditions for each eye which are symmetrical (for example identical lenses for left eye and right eye) or asymmetrical (for example different lenses for left eye and right eye).

[0084] In step 202 at the same time when the eyes are stimulated, the responses of the person 1 with regard to the stimuli by said at least one neuro-physiological sensor 5 are measured.

[0085] As stated above, preferentially several neuro-physiological sensors 5 are used such as at least two of the following sensors: an eye tracker sensor 46 (vergence, pupil size), an electrodermal sensor, an electrooculographic sensor, an electrocardiogram sensor, an electroencephalographic sensor and a respiratory electromyographical sensor. In addition, tracking the facial emotion 40 of the person 1 in response to the presented changing stimuli can be detected and responses by the person 1 through interface 42 can be collected.

[0086] Finally, in step 204, the measured responses are fed to the second processing unit 102 which includes a software programmed as a trained supervised machine learning model for evaluating the visual discomfort of the person as described above.

[0087] In addition, the tracked the facial emotion of the person 1 in response to the presented changing stimuli can be feed to said trained supervised machine learning model which is configured to consider such tracked emotions for evaluating the visual discomfort of the person.

[0088] Furthermore, self-evaluations detected via interface 42, in particular the feeling of a discomfort in response to the presented changing stimuli can be fed to said trained supervised machine learning model which is configured to consider these selfevaluations for evaluating the visual discomfort of the person 1 .

[0089] Finally, the optometric profile of the person 1 and stimulating the eyes of the person 1 can be fed to said trained supervised machine learning model for being considered for evaluating the visual discomfort of the person 1 .

[0090] Thus, an output of the trained supervised machine learning model may be for example a statement of presence or absence of visual discomfort of person 1 .

[0091] Clinical diagnosis of visual discomfort often relies on optometry procedures and patients’ descriptions of their visual symptoms ( / .e., near-vision difficulty, headaches). Blurring, headaches, and occasionally double vision are common internal symptoms that have been used as indicators in optometric examinations.

[0092] The trained supervised machine learning SVM model is configured in such a way that such visual symptoms which may be observed by person 1 may be linked to the measurements by the neuro-physiological sensors 5 and therefore be detected. Thus, the method and device 100 according to the disclosure allows to get more objective indicators of presence or absence of visual discomfort. In case of a monitoring application of the disclosure, the visual discomfort detected by the SVM model can generate a warning signal for example sent to the monitoring person 1 , to a person set by the monitoring person 1 , to a professional, etc. As in a particular testing, an interface can be manipulated by the person 1 and configured to allow the person 1 to express a selfevaluation for customising the evaluation of the visual discomfort of the person 1 . For example, in case of a smart eyewear, the person 1 could express her / his visual discomfort verbally (automatic phonic recognition) or by pressing a button on the frame of the smart eyewear, to indicate the presence and / or the intensity of pain in her / his eyes. Alternatively, the eyewear may be a Virtual Reality (VR) eyewear, Augmented Reality (AR) eyewear or Mixed Reality eyewear.

[0093] In an optional step 206 (shown with dashed arrow), when the presence of a visual discomfort is detected through above described method and device 100, device 104 mayselect in database 103, on the basis of the results of visual discomfort for the person 1 , an optical element reducing discomfort and visual discomfort for each eye.

[0094] The present disclosure therefore not only presents a device 100 and method for evaluation of visual discomfort, but also a device 104 and method to reduce or suppress discomfort and visual discomfort allowing to the person to gain life quality.

[0095] Of course, the present disclosure is not limited to the embodiments and variants presented but may be subject to various other embodiments and / or variants, which will be apparent to those skilled in the art. Thus, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the disclosure, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination. REFERENCE NUMERALS1 - person to evaluate3 - chinrest5 - neuro-physiological sensor7 - eye stimulating machine9 - movable display11 - gear rack13 - tooth wheel15 - actuator17 - displacement of the movable display19 - lens support mechanism 19L - left movable rack 19R - right movable rack 21 L - left actuator 21 R - right actuator23L - displacement of the left movable rack23R - displacement of the right movable rack25 - opaque plate27 - hole40 - facial emotion tracker42 - interface44 - optometric profiler46 - eye tracker100 - device for evaluating the visual discomfort of a person101 - first processing and control unit102 - second processing unit103 - database104 - device for decreasing visual discomfort200 - equipping step 202 - stimulating step204 - evaluating step206 - selecting stepLL1 - first left lensLL2 - second left lens LL3 - third left lensLL4 - fourth left lensLR1 - first right lensLR2 - second right lensLR3 - third right lens LR4 - fourth right lens

Claims

CLAIMS1. Method of training a supervised machine learning model for determining a level of the visual discomfort of a person (1) equipped with at least one neurophysiological sensor (5), carried out by a computer, characterised in that it comprises the following steps: a - obtaining measured responses of said at least one neuro-physiological sensor (5) as a function of several degrees of visual discomfort for persons (1) and several visual stimuli so as to form a database (103); b - training the supervised machine learning model with the database (103) to predict the level of the visual discomfort of a person (1).

2. Supervised machine learning model obtained by the training method according to the preceding claim, characterised in that it is configured to determine a level of the visual discomfort of a person (1) with measured responses of at least one neuro-physiological sensor (5) applied to the person (1) who is stimulated with several visual stimuli.

3. Supervised machine learning model according to the preceding claim, comprising a database (103) including measured responses of said at least one neurophysiological sensor (5) as a function of several degrees of visual discomfort for persons (1) and several visual stimuli.

4. Supervised machine learning model according to the preceding claim, in which the database (103) also includes at least one optical element reducing the visual discomfort of persons (1).

5. Method for evaluating a level of visual discomfort of a person (1 ) equipped with at least one neuro-physiological sensor (5), carried out by a computer, characterised in that it comprises the following steps: a - stimulating the eyes of the person (1) with changing visual stimuli which generate several degrees of visual discomfort for the person (1), b - measuring the responses of the person (1) with regard to the changing visual stimuli by said at least one neuro-physiological sensor (5), c - evaluating the measured responses by the supervised machine learning model according to any one of claims 2 to 4 for determining a level of the visual discomfort of the person (1).

6. Method according to the preceding claim, in which said changing visual stimuli of step a are obtained by the tasks of the person’s (1) everyday life.

7. Method according to claim 5 or 6, comprising, before step a, a step of establishing an optometric profile of the person (1), the method being configured to consider the optometric profile of the person (1).

8. Method for decreasing visual discomfort of a person (1) comprising the following steps:- performing a method for evaluating a level of visual discomfort of a person (1) according to any one of claims 5 to 7,- selecting on the basis of the result of the evaluation method of visual discomfort for the person (1) in a database (103) for each eye, an optical element reducing visual discomfort of the person (1).

9. Device (100) for evaluating the visual discomfort of a person (1) comprising:- at least one neuro-physiological sensor (5), and- at least a processing unit (102) configured to determine, with the supervised machine learning model according to any one of the claims 2 to 4, the visual discomfort of the person (1).

10. Device (100) according to the preceding claim, further comprising an interface(42) to be manipulated by the person (1) and configured to allow to the person (1) to express a self-evaluation, the supervised machine learning model being configured to consider such self-evaluations for customising the evaluation of the visual discomfort of the person (1).

11. Device (100) according to claim 9 or 10, further comprising a web application, accessible on the web, configured to submit the evaluation of the visual discomfort of the person (1).

12. Device (104) to decrease visual discomfort of a person (1) comprising:- a device (100) for evaluating the visual discomfort according to any one of claims 9 to 11 ,- at least a processing unit (102) configured to select, with the supervised machine learning model according to any one of the claims 2 to 4, at least oneadapted optical element reducing the visual discomfort of the person (1).

13. Eyewear for reducing the visual discomfort of a person (1), characterised in that it comprises said at least one adapted optical element selected by the device (104) according to the preceding claim .

14. Computer program product comprising instructions which, when executed by a computer, causes the computer to carry out the steps of the method according to any one of claims 5 to 8.

15. Computer readable storage medium comprising instructions which, when executed by a computer, causes the computer to carry out the steps of the method according to any one of claims 5 to 8.

Images