Device and method for determining data related to the progression of a person's refractive value

Abstract

The present invention relates to a processing device (120), a computer-implemented method and a computer program for determining data related to the progression (114) of a person's refractive value, and to a system, a computer-implemented method (210) and a computer program for providing data related to the progression (114) of a refractive value. Herein, the processing device (120) is configured to - receive data related to a person, the data including · the refractive state (320) of the person; · the age (314), gender (316) and ethnicity (318) of the person; and · at least one risk factor related to the person; - use at least one machine learning algorithm (132) to determine data related to the progression (114) of the person's refractive value, wherein the at least one machine learning algorithm (132) includes at least one prediction model (134) for determining the relationship between the data related to the person and the data related to the progression (114) of the person's refractive value. By using the processing device (120), the system (110), the computer-implemented method and the computer program, the prediction of the onset of myopia and the progression (114) of myopia can be improved.

Description

Technical Field

[0001] The present invention relates to a processing device, a computer-implemented method and a computer program for determining data related to the progression of a person's refractive value, and to a system, a computer-implemented method and a computer program for providing data related to the progression of a refractive value. Background Art

[0002] Pascolini D. and Mariotti S.P, Global estimates of visual impairment: 2010 [Global estimates of visual impairment: 2010], Br J Ophthalmol "British Journal of Ophthalmology" 2012; 96: 614e618.doi: 10.1136 / bjophthalmol-2011-300539 describes visual impairment as a global health and socio-economic problem, characterized by uneven distribution. The main cause of visual impairment is uncorrected refractive error, accounting for 43% of the causes. Despite the emergence of preventive policies, the situation has not changed much in the past 10 years, and uncorrected refractive error still appears on the list of the main causes of visual impairment in different regions of the world.

[0003] Grzybowski A. et al., A review on the epidemiology of myopia in schoolchildren worldwide [A review on the epidemiology of myopia in schoolchildren worldwide], BMC Ophthalmology "BMC Ophthalmology" (2020) 20: 27 points out that the prevalence of myopia in Asian countries is 60%, among which, East Asian countries show an even higher prevalence of 73%. Further, the study shows that the prevalence of myopia in European countries is 40%, while the prevalence of myopia in North American children is 42%. In contrast, African and South American countries show a prevalence of myopia below 10%.

[0004] Dong L. et al., Prevalence and time trends of myopia in children and adolescents in China [Prevalence and time trends of myopia in children and adolescents in China], Retina "Retina" 40(3); 2019, pp. 399 to 411, describes a recent meta-analysis on the prevalence and trends of myopia in Chinese children and adolescents, which estimates that the prevalence of myopia may reach 84% after 30 years.

[0005] Given the visual and pathological consequences of myopia and high myopia, it is important to address the issue at an early stage. The standard methods for correcting refractive errors such as myopia are glasses, contact lenses, or refractive surgery. However, based on current and projected global estimates of refractive errors, these methods can only correct refractive errors but may not be sufficient to slow further progression.

[0006] In addition to these mentioned correction methods, Walline J.J. et al., Interventions to slow progression of myopia in children [Interventions to slow progression of myopia in children], 2020, Cochrane Database Syst Rev. 1st ed. [Cochrane Database of Systematic Reviews, 1st edition]: CD004916; https: / / doi.org / 10.1002 / 14651858.CD004916.pub4 , have demonstrated that different solutions have successfully slowed or halted the rate of myopia progression. Examples include the administration of different doses of atropine; the use of spectacle lenses, specifically bifocal glasses, progressive multifocal lenses, peripheral defocus lenses, or multi-zone progressive optical defocus lenses; the use of multifocal contact lenses; and the use of orthokeratology.

[0007] Whether a particular type of myopia treatment can be applied to a person depends primarily on the person's risk of developing myopia and myopia progression. Morgan I.G. et al., IMI Risk Factors for Myopia [IMI Risk Factors for Myopia], Invest. Ophthalmol. Vis. Sci. [Investigative Ophthalmology & Visual Science] 2021; 62(5): 3 reported that the myopia community has invested a great deal of effort in understanding the key parameters that can influence this risk over the past few decades. Although a large number of variables affecting the onset and progression of myopia have been proposed, only a few are frequently mentioned. These parameters include ethnicity, behavioral habits, and parental myopia.

[0008] Available via https: / / bhvi.org / myopia-calculator-resources / the MyopiaCalculator provides online software for predicting the progression of a person's refractive value by using person-related data, which includes the person's age, ethnicity, and refractive error, as well as the recommended myopia treatment to be applied to the person. Based on this, the software calculates the predicted value of the percentage reduction in myopia progression compared to a standard correction procedure such as a single-vision spectacle lens, and calculates the predicted value of the person's refractive error process with or without immediate initiation of the recommended myopia treatment.

[0009] Available via https: / / myopiacare.com / myappia-myocalc / The obtained MyAppia provides another online software for predicting the progression of a person's refractive value by using human-related data, which includes age; the refractive state of the person (including the spherical equivalent of the person's eyes), and one or more recommended myopia treatments to be applied to the person. On this basis, the software simultaneously calculates the predicted value of the percentage of myopia progression slowdown and the predicted value of the person's refractive error process based on one or more of the myopia treatments of the recommended type of myopia treatment to be applied to the person.

[0010] US2018 / 160894 A1 discloses methods, systems, and computer program products for inferring the progression of the eye condition of an eye disease patient. When a patient visits an ophthalmologist, the patient or guardian may be interested in the current eye condition and the prediction of the future eye condition progression. Aspects of the present invention can be used to predict the progression of the eye condition of a patient (e.g., a child) at multiple different post-examination times after an examination. Predicting the progression of a patient's eye condition over time can be used to help an ophthalmologist customize a treatment plan for the patient and / or customize a follow-up examination schedule.

[0011] WO 2020 / 083382 A1 discloses systems, methods, devices, and media for performing the diagnosis of myopia onset and progression. Machine learning algorithms enable the automatic analysis of relevant features to generate predictions. Also disclosed are treatment methods incorporating machine learning algorithms to identify suitable treatments and predict treatment efficacy.

[0012] WO 2020 / 126513 A1 discloses a method for constructing a prediction model for predicting the evolution of at least one vision-related parameter of at least one person over time, the method comprising: obtaining continuous values respectively corresponding to repeated measurement results of at least one parameter of a first predetermined type over time for at least one member of a group of individuals; obtaining the evolution of the (multiple) vision-related parameters of the (multiple) members of the group of individuals over time; constructing a prediction model by at least one processor, including associating at least a portion of the continuous values with the evolution of the (multiple) vision-related parameters of the (multiple) members of the group of individuals obtained, the association including jointly processing at least a portion of the continuous values associated with the same parameter among the (multiple) parameters of the first predetermined type. The prediction model depends on each value in the jointly processed values to a different extent.

[0013] WO 2020 / 126514 A1 discloses a method for predicting the evolution over time of at least one vision-related parameter of at least one person, the method comprising: obtaining successive values of the person, the successive values corresponding respectively to repeated measurements over time of at least one parameter of a first predetermined type of the person; predicting, by at least one processor, the evolution over time of the person's vision-related parameter from the obtained successive values of the person by using a prediction model associated with a group of individuals; the prediction comprising associating at least a portion of the successive values of the person with the predicted evolution over time of the person's vision-related parameter, the association comprising jointly processing the successive values associated with the same parameter of the first predetermined type. The predicted evolution depends to a different degree on each of the jointly processed values.

[0014] US2021 / 145271 A1 discloses a system and method for determining a patient's corrective lens prescription using predictive calculations and refractive vision simulations. In summary, these techniques serve as a digital alternative to phoropter testing, thereby reducing the costs, time, and human error associated with eye examinations. Based on age, gender, autorefractor readings, and environmental factors, a patient-specific model is calculated and fed into a vision simulation tool. Based on this simulation, an eye care professional can determine the patient's corrective lens prescription.

[0015] CN 104751611 A discloses a method, device, and equipment for preventing and controlling myopia. The method comprises: acquiring data information on factors affecting the user's vision; processing the data information on factors affecting the user's vision to obtain a parameter value reflecting the user's eye use condition; and generating an alarm when the parameter value meets a preset threshold condition.

[0016] CN 106980748 A discloses a big data fitting-based method and system for monitoring the refractive development of adolescents. The method comprises: constructing a big data model of multi-factor dynamic refractive values; and fitting the dynamic refractive value of a tester according to the basic information of the tester and the measured value of the tester's uncorrected visual acuity by using the big data model of multi-factor dynamic refractive values, obtaining the cause of the tester's poor vision, and in the case where in-depth monitoring of the refractive development of the tester is required, fitting the static refractive value of the tester by using the big data model and performing long-term monitoring.

[0017] CN 107358036 A discloses a method, device and system for predicting the myopia risk of children. The method includes the following steps: obtaining the current detection data of the user's visual acuity and the physiological index value; and predicting the myopia risk of the user by using a visual acuity prediction model according to the current detection data and / or the physiological index value. According to this method, by obtaining the current detection data of the user's visual acuity and the physiological index value and using the visual acuity prediction model according to the current detection data and / or the physiological index value to predict the myopia risk of the user, the user and the user's parents can timely understand the myopia risk of the user and prevent or treat myopia in advance.

[0018] CN 110288266 A discloses a myopia risk assessment method and system. The method includes the following steps: obtaining the factor data of the myopia risk factors of the myopia risk assessment target, where the myopia risk factors include at least one factor; assigning values to each myopia risk factor according to the factor data to obtain an assignment result; calculating a myopia risk index according to the assignment results of all myopia risk factors; and obtaining the myopia risk result of the myopia risk assessment target according to the myopia risk index.

[0019] CN 110299204 A discloses a method and system for predicting the myopia prevention and control effect. The method includes the following steps: obtaining the myopia risk index value and the myopia prevention and control index value of the myopia prevention and control target, where the myopia risk index value is used to represent the myopia risk degree of the myopia prevention and control target, and the myopia prevention and control index value is used to represent the prevention and control intensity of the myopia prevention and control strategy of the myopia prevention and control target; calculating to obtain a myopia prevention and control efficacy value according to the myopia risk index value and the myopia prevention and control index value; and performing a prediction according to the myopia prevention and control efficacy value to obtain the myopia prevention and control effect of the myopia prevention and control target.

[0020] CN 112289446 A discloses a computer system for predicting adolescent myopia. The computer system includes a database device, a data input device, a myopia prediction device and a prediction result output device. The myopia prediction device is respectively connected to the database device and the data input device, generates a prediction model based on the feature data in the database in the database device by using a tree regression algorithm, outputs the predicted spherical lens degree based on the feature data of the subject by using the prediction model, and further outputs the prediction result through the prediction result output device.

[0021] KR 2021 0088654A and US2022 / 0028552 A1 disclose a method for constructing a prediction model for predicting the evolution of at least one vision-related parameter of at least one person over time. The method includes: obtaining continuous values that respectively correspond to repeated measurements of at least one parameter of a first predetermined type of at least one member of a group of individuals over time; obtaining the evolution of the (multiple) vision-related parameters of the (multiple) members of the group of individuals over time; constructing a prediction model by at least one processor, including associating at least a portion of the continuous values with the evolution of the (multiple) vision-related parameters of the (multiple) members of the group of individuals obtained over time, the association including jointly processing at least a portion of the continuous values associated with the same parameter among the (multiple) parameters of the first predetermined type. The prediction model depends on each value in the jointly processed values to a different degree.

[0022] Problems to be solved

[0023] In particular, with respect to the disclosures of KR 2021 0088654A or US2022 / 0028552 A1, the object of the present invention is to provide a processing device, a computer-implemented method and a computer program for determining data related to the progression of a person's refractive value, and a system, a computer-implemented method and a computer program for providing data related to the progression of the refractive value, which at least partially overcome the limitations of the prior art.

[0024] A specific object of the present invention is to provide a processing device, a system, a computer-implemented method and a computer program, which can improve the prediction of myopia onset and myopia progression and can be used in a more reliable manner by eye care professionals such as opticians, optometrists or ophthalmologists; or by end consumers such as people with myopia or related persons (especially the parents of the person or the caregiver of the person). Summary of the invention

[0025] This problem is solved by a processing device, a computer-implemented method and a computer program for determining data related to the progression of a person's refractive value, and a system, a computer-implemented method and a computer program for providing data related to the progression of the refractive value, which have the features of the independent claims. Preferred embodiments that can be implemented in isolation or in any combination are listed in the dependent claims and the following description.

[0026] In a first aspect, the present invention relates to a processing device for determining data related to the progression of a person's refractive value. According to the present invention, the processing device is configured to

[0027] - receive data related to a person, the data including

[0028] · the refractive state of the person;

[0029] · the age, gender and race of the person; and

[0030] · at least one risk factor associated with the person;

[0031] - using at least one machine learning algorithm to determine data related to the progression of the refractive value of the person, wherein the at least one machine learning algorithm includes at least one prediction model for determining the relationship between the data associated with the person and the data related to the progression of the refractive value of the person.

[0032] As used herein, the term "processing" or any of its grammatical variants refers to applying at least one algorithm to data received from at least one input file in such a way that the required data related to the progression of the refractive value of a person is provided by at least one output file for further processing. As commonly used, the term "data" refers to at least one piece of information included by at least one file, specifically by at least one input file or at least one output file. In particular with respect to the present invention, at least one piece of information included by at least one input file may be related to a person, while at least one piece of information included by at least one output file may be related to the progression of the refractive value of a person. The at least one algorithm may be configured to determine data related to the progression of the refractive value of a person based on data related to the person according to a predefined scheme by using data from at least one input file, wherein, as described in more detail below, artificial intelligence, in particular at least one machine learning algorithm, may also be applied.

[0033] As is commonly used, the term "processing device" refers to a device designated to determine data related to the progression of a person's refractive value based on person-related data received from at least one input file, which person-related data can preferably be provided to the processing device via at least one input interface, and the processing device preferably provides data related to the progression of a person's refractive value via at least one output interface, in particular by using a system described in more detail below, for further processing, for example. Specifically, the processing device may include at least one of an integrated circuit (in particular an application specific integrated circuit (ASIC)) or a digital processing device (in particular a digital signal processor (DSP), a field programmable gate array (FPGA), a microcontroller, a microcomputer, a computer) or an electronic communication unit (specifically a smartphone, a tablet computer, a personal digital assistant or a laptop computer). Additional components may be feasible, in particular at least one of a data acquisition unit, a preprocessing unit or a data storage unit. The processing device may preferably be configured to execute at least one computer program, in particular at least one line of computer program code, which computer program code is configured to execute at least one algorithm for determining data related to the progression of a person's refractive value, wherein the processing of the data may be carried out in at least one of a continuous or parallel manner.

[0034] According to the present invention, the processing device is configured to determine data related to the progression of a person's refractive value based on person-related data that can be received from at least one input file. As is commonly used, the term "determine" or any of its grammatical variants refers to the process of generating a representative result commonly denoted as "data". In particular with respect to the present invention, the data generated in this way includes multiple pieces of information related to the speculation of the refractive value of at least one eye of a person. In addition to the term "person", different terms may also be applied, such as "user", "subject", "individual" or "wearer".

[0035] As described above, the data related to a person includes the refractive state of the person. Preferably, the data related to the refractive state of the person may include at least one refractive value of at least one eye of the person. As used herein, the term "at least one eye" refers to one or both eyes of a person. As commonly used, the term "refractive value" corresponds to at least one refractive error of at least one eye of a person, and this refractive value can be particularly used to produce at least one optical lens, specifically at least one spectacle lens or at least one contact lens, and each of these lenses exhibits a diopter capable of correcting at least one refractive error of at least one eye of a person based on the refractive value. As further commonly used, the term "progression of refractive value" refers to the speculation of the temporal change, particularly the decrease, especially the monotonic decrease, of the refractive value of at least one eye of a person over a period of time. Based on Section 3.5.2 of Standard ISO 13666:2019 (referred to herein as the "Standard"), the term "spectacle lens" refers to an optical lens for correcting at least one refractive error of at least one eye of a person, wherein the optical lens is carried in front of the person's eyes so as to avoid direct contact with the person's eyes. Thus, the term "contact lens" refers to an optical lens for correcting at least one refractive error of at least one eye of a person, and this optical lens is in direct contact with the person's eyes when worn. Further, the term "glasses" refers to an element including two separate spectacle lenses and a spectacle frame, wherein each spectacle lens is ready to be received by a spectacle frame selected by a person.

[0036] Regarding the present invention, at least one refractive value may particularly be selected from spherical lens values, and preferably additionally from cylindrical lens values of the eye lens of at least one eye of a person. As defined in Section 3.12.2 of the Standard, the term "spherical power" (commonly abbreviated as "sph") refers to the back vertex power value of a spherical power lens. As defined in Section 3.13.7 of the Standard, the term "cylinder" (commonly abbreviated as "cyl") refers to the algebraic difference between the powers of the principal meridians, where the power of one principal meridian is subtracted from the power of another principal meridian selected as a reference.

[0037] Alternatively or additionally, data related to a person's refractive state can be or include at least one biometric value of at least one of the person's eyes. As is commonly used, the term "biometric value" refers to a measurement related to at least one extension of at least one characteristic of at least one of the person's eyes in three-dimensional space, and thus can generally be indicated by using a value and a corresponding unit indicating the spatial extension (such as meters). Regarding the present invention, in particular, the biometric value can simultaneously include the axial length and corneal radius of at least one of the person's eyes, wherein the anterior chamber depth or lens thickness of the eye can additionally be used as at least one of the biometric values. However, it may also be feasible to select at least one different biometric value. Further, those skilled in the art know that at least one refractive value of at least one of the person's eyes is closely related to at least one biometric value of at least one of the person's eyes. In fact, there are known relationships that can be applied to convert a first value related to at least one biometric value into a second value related to at least one refractive value or vice versa. Therefore, both at least one refractive value of a person and at least one biometric value of at least one of the person's eyes are represented by the term a person's "refractive state".

[0038] As is further commonly used, as defined in Section 3.1.10 of the standard, the term "diopter" or simply "D" refers to the ability of a lens (specifically, an eyeglass lens or a contact lens) to change at least one of the curvature or direction of an incident wavefront by refraction. As used further herein, the term "myopia" relates to a person's refractive state, wherein the value of the diopter of at least one of the person's eyes is less than -0.5 D. As is further commonly used, the term "high myopia" refers to a person's refractive state, wherein the value of the diopter of at least one of the person's eyes is less than -6.0 D. As is further commonly used, the term "myopia progression" refers to the speculation about the temporal change, particularly a decrease, especially a monotonic decrease, of the diopter of at least one of the person's eyes over a period of time. In this context, the period of time for speculation can cover many years, preferably from 1 year to 12 years, more preferably from 2 years to 10 years, particularly from 4 years to 8 years. However, it may also be feasible to use a different period of time. As used further herein, the term "myopia onset" refers to the time point at which the diopter of at least one of the person's eyes decreases from a value higher than -0.5 D to a value lower than -0.5 D during myopia progression.

[0039] As described above, the data related to a person further includes personal data, namely, the age, gender, and race of the person. As is commonly used, the term "age" refers to the time since the day the person was born, and preferably can be indicated in years. The present invention can be particularly applicable to children, adolescents, or young adults aged 4 to 24 years old, especially 5 to 20 years old. However, the present invention can also be applied to people of different ages. As is further commonly used, the term "gender" refers to the identity of a person with respect to the terms "female" and "male", and non-binary identities can also be possible. As is further commonly used, the term "race" refers to the attribution of a person to a specific group. Without wishing to be bound by theory, although Morgan I.G. et al. (see above) summarized that the evidence and causal relationship between a person's gender and race on the one hand and myopia progression on the other hand are weak or inconsistent, as stated in WO 2020 / 083382 A1, compared with a machine learning algorithm trained using a mixed population training set, a trained machine learning algorithm can still generate more accurate predictions for a person belonging to at least one of a specific gender or group.

[0040] According to the present invention, the data related to a person includes at least one risk factor related to the person. As is commonly used, the term "risk factor" refers to at least one value in a condition or process related to the person, which has been proven to exacerbate or slow down the myopia progression of at least one eye of the person and / or extend or shorten the time point of myopia onset. Without wishing to be bound by theory, according to Morgan I.G. et al. (see above), the risk factor may be useful for designing myopia treatment. Preferably, the myopia treatment is selected to be performed by demonstrating a causal relationship related to a defined mechanism between the hypothesized risk factor and the observed myopia treatment, especially by using the association with the condition or process on the one hand and the association with myopia progression on the other hand, and these associations can be demonstrated by using cross-sectional data or preferably longitudinal data on a defined group. For the definition of different types of data, reference can be made to the description below.

[0041] In a particularly preferred embodiment, the at least one risk factor can be selected from the data related to the refractive state of at least one of the person's parents. For the term "refractive state", reference can be made to the definition provided above, and this term is applicable to at least one of the person's parents with necessary modifications. According to Morgan I.G. et al. (see above), the evidence and causal relationship between the refractive state of at least one of the person's parents on the one hand and myopia progression on the other hand are strong, regardless of whether the discussion is based on genetic variation or the transmission of a myopia lifestyle that makes their children prone to myopia.

[0042] Alternatively or additionally, at least one risk factor can be selected from data related to at least one parameter related to a person's behavioral habits. As is commonly used, the term "behavioral habits" refers to the repetitive activities of a person during which at least one eye of the person is exposed to light radiation. As is further commonly used, the term "light radiation" refers to electromagnetic waves with wavelengths from 380 nm to 780 nm (defining the so-called "visible wavelength range"), wherein radiation from adjacent wavelength ranges (especially from 100 nm to below 380 nm, denoted as the "ultraviolet-A wavelength range"; and / or from above 780 nm to 1.5 μm, denoted as the "near-infrared wavelength range") can also be considered for the purposes of the present invention. Without wishing to be bound by theory, Morgan I.G. et al. (see above) have pointed out that there is strong and causal evidence and a causal relationship between a person's behavioral habits on the one hand and myopia progression on the other hand, which depends in particular on at least one parameter of the light radiation incident on at least one eye of the person, especially parameters related to the intensity, spectral distribution, and duration of the light radiation.

[0043] According to the present invention, at least one parameter related to a person's behavioral habits is selected from data related to at least one of the following:

[0044] · A first amount of time a person spends on near vision work;

[0045] · A second amount of time a person spends outdoors.

[0046] As commonly used, "near vision work" refers to a first type of repetitive activity of a person that endeavors to direct the person's eyes to an object placed at a certain distance from the person's eyes in a manner that the person's eyes accommodate to the near point. As commonly used, the term "accommodation" or any of its grammatical variants refers to the adjustment of the refraction of a person's eyes to the retinal plane of the person's eyes when imaging an object located between the near point and the far point in front of the person's eyes. The term "far point" refers to the endpoint of the refraction direction of a person's eyes without accommodation, while the term "near point" refers to the point indicating the minimum distance in front of the person's eyes at which the object can still be clearly imaged on the retinal plane of the person's eyes, and the near point is an individual quantity that is particularly dependent on the person's age. In this context, a fixed position of a person's eyes (especially a fixed position on the cornea, such as the position of an observable corneal reflex) can be used as a reference point for measuring the distance of the near point to the person's eyes. In fact, the object of a person's near vision work can be particularly a literary object or a mobile communication device. As used herein, the term "literary object" refers to an object that includes printed information, especially an object selected from books, brochures, or newspapers. Further, the term "mobile communication device" refers to at least one electronic device that is configured to present information by using an electronically driven screen, and the at least one electronic device can be carried by a person and thus can move with the person, and the at least one electronic device is especially selected from smartphones, laptops, personal digital assistants, or laptop computers. However, other types of objects can also be feasible.

[0047] As further commonly used, the term "time spent outdoors" refers to a second type of repetitive activity of a person that is carried out by the person outdoors in the open air, especially before school, during or after school, on weekends, or during holidays. As mentioned above, the intensity of light radiation incident on at least one of a person's eyes can increase significantly during the second amount of time the person spends outdoors. Without wishing to be bound by theory, Morgan I.G. et al. (see above) pointed out that the association between time spent outdoors and the slowing of myopia progression is strong and has always been observed, while there is a controversy regarding whether an increase in time spent outdoors not only slows myopia progression but also reduces the occurrence of myopia.

[0048] The term "amount of time" as further commonly used refers to the duration of the corresponding repetitive activity of a person that is carried out especially by using the average occupancy of the person expressed in hours per day or hours per week. To determine the amount of time, the time period regularly spent outside during holidays can be used; however, the value can be modified by (preferably, additionally) increasing the time period spent outside during holidays, thereby taking into account the comparison of the annual duration of holidays with the annual duration of school time or pre-school time.

[0049] In a preferred embodiment, the data related to a person may further include at least one type of myopia treatment to be applied to the person. As commonly used, the term "type of myopia treatment" refers to at least one preventive intervention that is configured to slow down the progression of myopia in at least one eye of a person or reduce at least one of the occurrences of myopia. In particular, the type of myopia treatment may include applying at least one optical lens to at least one eye of a person. Herein, the at least one optical lens may preferably be selected from spectacle lenses or contact lenses. In particular for the type of myopia, spectacle lenses may particularly be selected from bifocal lenses, progressive multifocal lenses, peripheral defocus lenses or multi-zone forward optical defocus lenses, while contact lenses may preferably be selected from multifocal contact lenses or orthokeratology lenses. As defined in Section 3.7.3 of the standard, the term "bifocal lens" refers to a specific type of spectacle lens having two parts, where each part has a different diopter value. Similarly, as defined in Sections 3.7.7 to 3.7.8 of the standard, the term "progressive multifocal lens" refers to another type of spectacle lens that has a smooth and uninterrupted change in diopter on the lens surface. Further, the term "peripheral defocus lens" refers to another type of spectacle lens that has a change in optical power from the central optical zone of the lens to the periphery, thereby causing peripheral defocus at the eccentric regions of the retina. Further, the term "multi-zone forward optical defocus lens" refers to another type of spectacle lens that includes a central optical zone for correcting distance refractive error and an annular multifocal zone that may include segments having different diopters compared to the central optical zone. Further, the term "orthokeratology lens" refers to a breathable contact lens that is configured to temporarily reshape the corneal shape, thereby changing the diopter of at least one eye of a person. However, it may also be feasible to use other types of spectacle lenses or contact lenses.

[0050] Alternatively or additionally, the type of myopia treatment may be selected from the application of at least one of a certain dose of a drug (specifically atropine) or refractive surgery. However, applying at least one of a drug or refractive surgery may only correct the current refractive error but may not be sufficient to slow down further myopia progression.

[0051] Further in accordance with the present invention, the processing device is further configured to determine data related to the progression of a person's refractive value by using at least one machine learning algorithm. As is commonly used, the term "machine learning" refers to the process of applying artificial intelligence to automatically generate a model for at least one of classification or regression. In the present context, preferably, at least one machine learning algorithm configured to generate a desired model based on a large number of training data sets can be used. As is further commonly used, the term "training" indicates improving the performance of the method steps for determining the desired data during the training phase by providing a plurality of training data sets and performing these training data sets through specific method steps. In the present context, each training data set for training purposes is similar to the expected data set, such as data related to the progression of a person's refractive value. However, these training data sets include known data. Then, specific method steps are performed with a specific training data set, and the result of the content obtained in this way is adjusted to the known data from the specific training data set. In the present context, a plurality of training data sets are iteratively applied during the training phase in order to improve the approximation of the result obtained during the execution of the specific method steps, specifically by repeating the training of the specific method steps until the deviation between the data obtained by performing the specific method steps and the known data included in each training data set can be below a threshold. After the training phase, it can be reasonably expected that the data obtained by performing the specific method steps approximates the known data in the same way as achieved during the training phase. In this way, a more accurate determination of the data can be obtained during the training phase. Therefore, after the training phase, the desired performance of the specific method steps can be obtained.

[0052] The machine learning algorithms used herein for determining the desired data related to the progression of a person's refractive value include at least one prediction model for determining the relationship between data related to a person and data related to the progression of a person's refractive value. Preferably, one or more models can be selected from linear prediction models, specifically selected from at least one of support vector regression (SVR) or Gaussian process regression (GPR). However, it may also be feasible to use one or more additional types of models.

[0053] In a particularly preferred embodiment, at least one machine learning algorithm may include applying a first prediction model using longitudinal data and a second prediction model using cross-sectional data. As commonly used herein, the terms "first" or "second" are considered as descriptions of elements, without specifying an order or a time sequence, and also without excluding the possibility that there may be other identical elements. As further commonly used, the term "longitudinal data" refers to a plurality of first item data related to a specific person, where the term "cross-sectional data" includes at least one second item data related to a plurality of different people. By way of example, longitudinal data refers to a plurality of refractive values, specifically spherical lens values, related to the same person over a period of time, thus providing the progression of refractive values as a particular person ages. In contrast, cross-sectional data is related to the same refractive values (specifically spherical lens values) of a plurality of different people of the same age and preferably having at least one of the same gender and the same race.

[0054] In this particularly preferred embodiment, the first prediction model using longitudinal data may preferably be a first linear prediction model using support vector regression (SVR), while the second prediction model using cross-sectional data may be a second linear prediction model using Gaussian process regression (GPR). As commonly used, the term "support vector regression" or "SVR" refers to a machine learning tool for classification and regression, regarded as a non-parametric technique relying on a kernel function. As further commonly used, the term "Gaussian process regression" or "GPR" refers to a machine learning tool using a probability model based on a non-parametric kernel for prediction.

[0055] In a specific embodiment, the total data input of at least one machine learning algorithm may include a first amount of longitudinal data input and a second amount of cross-sectional data input, where both the first data amount and the second data amount are used to determine the required data related to the progression of a person's refractive value. Preferably, the total data input may be distributed in such a way that the first amount may be 30% to 70%, preferably 50% to 70%, while the second amount may be 30% to 70%, preferably 30% to 50%, where the first amount and the second amount add up to 100%. However, it may also be feasible to use different kinds of distributions.

[0056] In another specific embodiment, the machine learning algorithm may include using at least two different prediction models that can be combined. Herein, the first prediction model may generate intermediate prediction data, specifically the relationship between data related to a person and the ratio of the axial length to the corneal radius data, where the intermediate prediction data (specifically the ratio of the axial length to the corneal radius data) may preferably be used as the input of the second prediction model, especially for refractive power prediction. However, it may also be feasible to use another type of intermediate prediction data.

[0057] In another specific embodiment, the processing device may further be configured to determine at least one other item of data as data related to the progression of a person's refractive value. Herein, the at least one other item of data may preferably be selected from at least one of the following items:

[0058] - The ranking of the person compared to multiple other people;

[0059] - The myopia risk of the person;

[0060] - The high myopia risk of the person.

[0061] As commonly used, the term "ranking" or any of its grammatical variants refers to comparing the results of a particular person with the results of multiple other people for whom the same type of data has been determined. In particular, the results obtained through ranking can be indicated by a number or preferably by a percentage, which indicates the position of the particular person relative to other people (specifically, people of the same age, and preferably, people having at least one of the same gender and the same race).

[0062] As further used herein, the term "myopia risk" refers to the first probability of a person developing myopia by obtaining the value at which myopia occurs during the progression of the refractive value. In particular, the myopia risk can be indicated by a qualifier selected from a list, where each item in the list refers to a specific myopia state. Specifically, the qualifier can be selected from "high" and "low", where the term "high" can indicate that, according to prediction, the refractive power value of at least one eye of the person is below -0.5 dpt during the progression of the refractive value, while the term "low" can indicate that, according to prediction, the refractive power of at least one eye of the person remains at -0.5 dpt or above during the progression of the refractive value.

[0063] Similarly, the term "high myopia risk" refers to another probability of a person developing high myopia by obtaining the refractive value defined as high myopia during the progression of the refractive value. In particular, the high myopia risk can be indicated by another qualifier selected from another list, where each item in the other list refers to a specific high myopia state. Specifically, the qualifier can be selected from "high" and "low", where the term "high" can indicate that, according to prediction, the refractive power value of at least one eye of the person is below -6.0 dpt during the progression of the refractive value, while the term "low" can indicate that, according to prediction, the refractive power of at least one eye of the person remains at -6.0 dpt or above during the progression of the refractive value.

[0064] On the other hand, the present invention relates to a system for providing data related to the progression of refractive values. As is commonly used, the term "system" refers to a combination of at least two components, each component being configured to perform a specific task. However, among them, the at least two components can cooperate and / or interact with each other to achieve the required task.

[0065] According to the present invention, the system comprises:

[0066] - at least one input interface configured to receive human-related data as described elsewhere herein;

[0067] - a processing device as described elsewhere herein; and

[0068] - at least one output interface configured to provide data related to the progression of a person's refractive value.

[0069] Regarding the processing device, reference can be made to the description thereof throughout this document.

[0070] Furthermore, the processing device may preferably include at least one communication interface configured to provide communication with the at least one input interface and the at least one output interface. As is commonly used, the term "communication interface" refers to a transmission channel designated for data transmission. Preferably, the communication interface may be arranged as a unidirectional interface configured to forward at least one item of data in a single direction, from the at least one input interface to the processing device, or from the processing device to the at least one output interface. Alternatively, the communication interface may be arranged as a bidirectional interface configured to forward at least one item of data in one of two directions, from a communication unit that may simultaneously include an input interface and an output interface to the processing device, or vice versa. For the purpose of data transmission, the communication interface may include at least one of a wired element or a wireless element, wherein the wireless element may be configured to operate by using at least one wireless communication protocol such as Wi-Fi or Bluetooth. In a particularly preferred embodiment, the communication may be or may include encrypted data transmission or encrypted data exchange. However, another communication interface may also be feasible.

[0071] As commonly used, the term "input interface" refers to a device configured to receive at least one piece of data, specifically human-related data as described in more detail above or below. For this purpose, the data can preferably be provided in the form of at least one of an input file or input data through the use of a graphical user interface (GUI), and forwarded to a processing device for determining the required data related to the progression of a person's refractive value. As commonly used, the term "graphical user interface" or "GUI" refers to a type of input interface configured to receive the required personal data from a graphical interaction with the user. In this document, the user can be selected from at least one of the following: an eye care professional, specifically an optician, an optometrist, or an ophthalmologist, or; or a person suffering from myopia or a related person, especially the person's parent or the person's caregiver. The graphical interaction can include presenting graphical icons to the user on a screen, recording the user's response, and determining the required input data by evaluating the user's response. For this purpose, at least one touch screen can be used, which is configured to provide a channel for inputting at least one piece of data, specifically human-related data. However, other devices can also be feasible, such as at least one of a camera or a scanner, which is configured to generate an input file for processing by the processing device to obtain the required input data.

[0072] As is further commonly used, the term "output interface" refers to another device that is configured to provide at least one output file including at least another item of data, specifically the required data related to the progression of a person's refractive value. In this document, the processing device may preferably be configured to provide data related to the progression of a person's refractive value via the same or a different graphical user interface that preferably displays the data included in the output file to the user, specifically by using a graphical user interface, preferably the same graphical user interface as that used for the input interface. As described above, the user may be selected from at least one of the following: an eye care professional, such as an optician, an optometrist, or an ophthalmologist; or a person suffering from myopia or a related person, especially the person's parent or the person's caregiver. Alternatively or additionally, the processing device may preferably be configured to provide data related to the progression of a person's refractive value to at least one output interface in the form of a structured output file. As is commonly used, the term "structured output file" refers to a file in which multiple items of data follow a predefined arrangement, specifically to facilitate further processing of the output file by the recipient, which recipient is specifically at least a data processing system in the office or clinic of an eye care professional or in a hospital. Further, at least one additional output interface may be feasible, specifically at least another output interface that is configured to provide the at least another item of data received (specifically the data related to the progression of a person's refractive value determined by the processing device) to another recipient, which other recipient may especially be at least one data storage unit configured to store a copy of the output data, a printer configured to print the output data, or a microphone configured to read the output data, and these data may each have different formats. However, another output interface may also be feasible.

[0073] In a particularly preferred embodiment, the system may include at least one mobile communication device or may be implemented by at least one mobile communication device. As is commonly used, the term "mobile communication device" refers to at least one of a smartphone, a tablet computer, a personal digital assistant, or a laptop computer, which can be carried by a person and thus can move with the person. However, other types of mobile communication devices may also be contemplated. Generally, at least one mobile communication device may include at least one input interface, at least one processing device, and at least one output interface. The mobile operating system running on at least one mobile communication device may be configured to facilitate the use of software, such as a graphical user interface; multimedia functions; and communication facilities, such as the Internet or at least one wireless communication protocol, such as Wi-Fi or Bluetooth. In this document, the mobile communication device may particularly be used to collect the required input data from the user, specifically by applying a graphical user interface to be used for self-input type input data known to the user.

[0074] As commonly used, the term "provide" or any grammatical variant thereof refers to forwarding a prediction of data related to the progression of a person's refractive value to at least one output interface, such as at least one output interface described in more detail above and below. As used herein, the term "prediction" refers to a prediction of data related to the progression of a person's refractive value over a period of time in the future, which period can cover many years, preferably from 1 to 12 years, more preferably from 2 to 10 years, and especially from 4 to 8 years. However, it may also be feasible to use different time periods.

[0075] Additionally, the at least one output interface may further be configured to further provide at least one percentile ranking of the refractive value and / or a corrected progression of the person's refractive value considering at least one type of myopia treatment. As commonly used, the term "percentile ranking" refers to providing a value based on population-based data covering a range of ages. Typically, the 97th percentile, 50th percentile, and 3rd percentile can be provided for multiple people of the same age; herein, the 97th percentile, 50th percentile, and 3rd percentile respectively indicate that the relevant curve covers 97%, 50%, or 3% of the population on which the percentile is based. However, alternatively or additionally, it may also be feasible to use at least one other percentile, such as at least one of the 1st percentile, 2nd percentile, 5th percentile; 95th percentile, 98th percentile, or 99th percentile.

[0076] As further used herein, the term "corrected progression" refers to the process of change of a person's refractive value in which the implementation of at least one type of myopia treatment as described in more detail above has been considered. For example, reference may be made to the figures presented below.

[0077] On the other hand, the present invention relates to a computer-implemented method for determining data related to the progression of a person's refractive value. Herein, the method comprises the following steps:

[0078] - Receiving data related to the person, the data comprising

[0079] · the refractive state of the person;

[0080] · the age, gender, and ethnicity of the person; and

[0081] · at least one risk factor related to the person;

[0082] - Using at least one machine learning algorithm to determine data related to the progression of the person's refractive value, wherein the at least one machine learning algorithm comprises at least one prediction model for determining the relationship between the data related to the person and the data related to the progression of the person's refractive value.

[0083] On the other hand, the present invention relates to a computer-implemented method for providing data related to the progression of a person's refractive value, wherein the method comprises the following steps:

[0084] - Receiving data related to a person by using at least one input interface according to a method for determining data related to the progression of a person's refractive value;

[0085] - Determining data related to the progression of a person's refractive value by using at least one processing device as described above or below according to a method for determining data related to the progression of a person's refractive value;

[0086] - Providing data related to the progression of a person's refractive value by using at least one output interface.

[0087] Various embodiments for implementing the method according to the present invention are conceivable. According to a first embodiment, all method steps can be performed by using a single processing device (such as a computer, especially a stand-alone computer) or an electronic communication unit (specifically a smart phone, a tablet computer, a personal digital assistant or a laptop computer). In this embodiment, the single processing device can be configured to specifically execute at least one computer program, especially at least one line of computer program code configured to execute at least one algorithm as used in at least one of the methods according to the method of the present invention. Herein, the computer program executed on the single processing device can include all instructions for causing the computer to execute at least one of the methods according to the method of the present invention. Alternatively or additionally, at least one method step can be performed by using at least one remote processing device, especially selected from at least one of a server or a cloud computer, which is not located at the user site when at least one method step is executed. In this other embodiment, the computer program can include at least one remote part to be executed by at least one remote processing device to execute at least one method step. Further, the computer program can include at least one interface configured to forward and / or receive data to and / or from at least one remote part of the computer program.

[0088] The above method according to the present invention is a computer-implemented method. As commonly used, the term "computer-implemented method" refers to a method involving at least one programmable device, especially from a mobile communication device. However, another programmable device can also be feasible. Herein, at least one programmable device can particularly include or access a processing device, wherein at least one feature of these methods is executed by using at least one computer program. According to the present invention, the computer program can be provided on at least one programmable device, or at least one mobile communication device can access the computer program via a network such as an internal network or the Internet.

[0089] On the other hand, the present invention relates to a computer program comprising instructions which, when executed by a computer, cause the computer to carry out the method according to any one of the foregoing method embodiments. Specifically, the computer program may be stored on a non-transitory computer-readable data carrier. Thus, specifically, any one of the method steps indicated above may be carried out by a computer or a computer network (preferably by using a computer program).

[0090] On the other hand, the present invention relates to a computer program product having program code means for carrying out the method according to the present invention when the program is executed on a computer or a computer network. Specifically, the program code means may be stored on a computer-readable data carrier.

[0091] On the other hand, the present invention relates to a data carrier having a data structure stored thereon which, after being loaded into a computer or a computer network, such as into the working memory or main memory of a computer or a computer network, can carry out any one of the methods according to one or more of the embodiments disclosed herein.

[0092] On the other hand, the present invention relates to a computer program product having program code means stored on a machine-readable carrier for carrying out the method according to one or more of the embodiments disclosed herein. As used herein, the term "computer program product" refers to a program as a tradable product. The product can generally exist in any format (such as in a paper format) or on a computer-readable data carrier. Specifically, the computer program product may be distributed on a data network such as the Internet.

[0093] On the other hand, the present invention relates to a modulated data signal comprising instructions readable by a computer system or a computer network for carrying out any one of the methods according to one or more of the embodiments disclosed herein.

[0094] On the other hand, the present invention relates to a method for producing at least one spectacle lens. Thus, producing at least one spectacle lens comprises processing at least one lens blank by using at least one manufacturing device, the at least one manufacturing device employing data related to refractive values determined by a method for determining data related to the progression of a person's refractive values, in particular by using a processing device, wherein the data is forwarded to the at least one manufacturing device by using a method for providing data related to the progression of a person's refractive values as described elsewhere herein.

[0095] For further details regarding the methods and computer programs described herein, reference may be made to the description thereof throughout this document.

[0096] Compared to the prior art, the devices, systems, computer-implemented methods and computer programs according to the present invention exhibit advantages. In particular, they are capable of improving the prediction of myopia onset and myopia progression. In the present context, the input data that can preferably be selected for the algorithm is typical input data that can generally be accessed by eye care professionals, specifically by opticians, optometrists or ophthalmologists; or by a person suffering from myopia or a related person, in particular the person's parents or the person's caregiver. Thus, the present invention provides a flexible application of the algorithm and can support eye care professionals or end consumers in accurately predicting the progression of refractive error in one or both eyes of a person. In the present context, the prediction can allow the specification of preventive strategies for myopia progression and myopia onset.

[0097] As used herein, the terms "having", "including" or "comprising" or any grammatical variants thereof are used in a non-exclusive manner. Thus, these terms can either refer to a situation where no other features are present in the entity described in the present context apart from the features introduced by these terms, or to a situation where one or more other features are present. As an example, the expressions "A has B", "A includes B" and "A comprises B" can all refer to a situation where no other elements are present in A apart from B (i.e., the situation where A consists only of B), or to a situation where one or more other elements are present in entity A apart from B, such as element C, elements C and D, or even other elements.

[0098] As further used herein, the terms "preferably", "more preferably", "particularly", "more particularly" or similar terms are used in conjunction with optional features without limiting the possibilities of alternatives. Thus, the features introduced by these terms are optional features and are not intended to limit the scope of the claims in any way. As will be recognized by the person skilled in the art, the present invention can be carried out by using alternative features. Similarly, the features introduced by "in an embodiment of the present invention" or similar expressions are intended to be optional features without any limitation regarding alternative embodiments of the present invention, without any limitation regarding the scope of the present invention, and without any limitation regarding the possibility of combining the features introduced in this way with other features of the present invention.

[0099] In summary, the following embodiments are particularly preferred within the scope of the present invention:

[0100] Embodiment 1: A processing device for determining data related to the progression of a person's refractive value, wherein the processing device is configured to

[0101] - receive data related to a person, the data comprising

[0102] · the refractive state of the person;

[0103] · The age, gender and race of the person; and

[0104] · At least one risk factor associated with the person;

[0105] - Using at least one machine learning algorithm to determine data related to the progression of the refractive value of the person, wherein the at least one machine learning algorithm includes at least one prediction model for determining the relationship between the data related to the person and the data related to the progression of the refractive value of the person.

[0106] Example 2: The processing device according to the previous embodiment, wherein the at least one risk factor refers to at least one value in a condition or process associated with the person, and the condition or process has been proven to exacerbate or delay at least one of the myopia progression or the onset time of myopia in at least one eye of the person.

[0107] Example 3: The processing device according to any one of the foregoing embodiments, wherein the at least one risk factor is selected from data related to at least one of the following:

[0108] · The refractive state of at least one of the person's parents; or

[0109] · At least one parameter related to the person's behavior habits.

[0110] Example 4: The processing device according to the previous embodiment, wherein the at least one parameter related to the person's behavior habits is selected from data related to at least one of the following:

[0111] · The first amount of time the person spends on near vision work; or

[0112] · The second amount of time the person spends outdoors.

[0113] Example 5: The processing device according to the previous embodiment, wherein the first amount of time is the first duration of the person's repeated near vision work.

[0114] Example 6: The processing device according to any one of the previous two embodiments, wherein the near vision work refers to the first type of repetitive activity of the person, and the first type of repetitive activity is dedicated to guiding the person's eyes to an object placed at a certain distance from the person's eyes in a way that the person's eyes adapt to the near point.

[0115] Example 7: The processing device according to any one of the previous three embodiments, wherein the second amount of time is the second duration of the person's time spent outdoors.

[0116] Example 8: The processing device according to any one of the first four examples, wherein the time spent outdoors is a second type of repetitive activity of the person, and the second type of repetitive activity is carried out by the person outdoors in the open air, especially before going to school, during or after school, on weekends or during holidays.

[0117] Example 9: The processing device according to the previous example, wherein the second type of repetitive activity of the person is carried out by the person before going to school or after school, on weekends or during holidays.

[0118] Example 10: The processing device according to any one of the first six examples, wherein at least one of the first time amount or the second time amount is represented by using the average occupancy of the person expressed in hours per day or hours per week.

[0119] Example 11: The processing device according to any one of the foregoing examples, wherein the refractive state is selected from at least one of the following:

[0120] · At least one refractive value of at least one eye; or

[0121] · At least one biometric value of the at least one eye.

[0122] Example 12: The processing device according to the previous example, wherein the at least one refractive value of at least one eye is selected from the spherical lens value of the eye lens of the at least one eye.

[0123] Example 13: The processing device according to the previous example, wherein the at least one refractive value of at least one eye is additionally selected from the cylindrical lens value of the eye lens of the at least one eye.

[0124] Example 14: The processing device according to any one of the previous three examples, wherein the at least one biometric value of the at least one eye is a measured value related to at least one extension of at least one feature of the at least one eye in three-dimensional space.

[0125] Example 15: The processing device according to the previous example, wherein the at least one biometric value of the at least one eye includes the axial length and corneal radius of the at least one eye.

[0126] Example 16: The processing device according to the previous example, wherein the at least one biometric value of at least one eye of the person further includes at least one of the anterior chamber depth or lens thickness of the at least one eye.

[0127] Example 17: The processing device according to any one of the preceding examples, wherein the data related to the person further includes at least one type of myopia treatment.

[0128] Example 18: The processing device according to the preceding example, wherein the at least one type of myopia treatment is selected from the application of at least one of the following:

[0129] · An optical lens selected from contact lenses or spectacle lenses;

[0130] · A certain dose of medicine; or

[0131] · Refractive surgery.

[0132] Example 19: The processing device according to the preceding example, wherein the contact lens is selected from multifocal contact lenses.

[0133] Example 20: The processing device according to any one of the two preceding examples, wherein the spectacle lens is selected from bifocal lenses, progressive multifocal lenses, peripheral defocus lenses, or multi-zone forward optical defocus lenses.

[0134] Example 21: The processing device according to any one of the preceding examples, wherein the at least one machine learning algorithm includes using a first prediction model and a second prediction model, wherein the first prediction model generates intermediate prediction data, and wherein the intermediate prediction data is used as an input to the second prediction model.

[0135] Example 22: The processing device according to the preceding example, wherein the first prediction model generates a relationship between the data related to the person and the data related to the progression of the refractive value of the person.

[0136] Example 23: The processing device according to the preceding example, wherein the first prediction model generates the ratio of the axial length of the person to the corneal radius.

[0137] Example 24: The processing device according to any one of the two preceding examples, wherein the relationship generated by the first prediction model is used as an input to the second prediction model.

[0138] Example 25: The processing device according to the preceding example, wherein the ratio of the axial length divided by the corneal radius data is used as an input to the second prediction model.

[0139] Example 26: The processing device according to any one of the five preceding examples,

[0140] o wherein the first prediction model uses longitudinal data, wherein the longitudinal data includes a plurality of first item data related to a specific person; and

[0141] Among them, the second prediction model uses cross-sectional data, where the cross-sectional data includes at least one second item of data related to a plurality of different people.

[0142] Example 27: The processing device according to the previous embodiment, where the longitudinal data refers to a plurality of refractive values related to the same person over a period of time.

[0143] Example 28: The processing device according to the previous embodiment, where the longitudinal data provides the progression of refractive values as the same person ages.

[0144] Example 29: The processing device according to any one of the previous three embodiments, where the cross-sectional data is related to the same refractive values of a plurality of different people of the same age.

[0145] Example 30: The processing device according to any one of the previous four embodiments, where the first prediction model is a first linear prediction model using support vector regression (SVR).

[0146] Example 31: The processing device according to any one of the previous five embodiments, where the second prediction model is a second linear prediction model using Gaussian process regression (GPR).

[0147] Example 32: The processing device according to any one of the previous six embodiments, where the total data input of the at least one machine learning algorithm includes a first amount of longitudinal data input and a second amount of cross-sectional data input.

[0148] Example 33: The processing device according to the previous embodiment, where the first amount is 30% to 70%.

[0149] Example 34: The processing device according to any one of the previous two embodiments, where the second amount is 30% to 70%.

[0150] Example 35: The processing device according to any one of the previous three embodiments, where the first amount and the second amount add up to 100%.

[0151] Example 36: The processing device according to any one of the previous embodiments, where the processing device is further configured to determine at least one of the following:

[0152] · The ranking of the person compared to a plurality of other people;

[0153] · The myopia risk of the person;

[0154] · The high myopia risk of the person.

[0155] Example 37: The processing device according to the previous example, wherein the ranking refers to comparing the person's results with the results of multiple other people for whom data of the same type has been determined.

[0156] Example 38: The processing device according to any one of the previous two examples, wherein the myopia risk refers to determining the probability that the person will develop myopia by obtaining the value at which myopia occurs during the progression of these refractive values.

[0157] Example 39: The processing device according to the previous example, wherein myopia is related to the refractive state of the person, and wherein the refractive power of at least one eye of the person is a value lower than -0.5 dpt.

[0158] Example 40: The processing device according to any one of the previous two examples, wherein the value at which myopia occurs is the time point at which the refractive power of at least one eye of the person decreases from a value higher than -0.5 dpt to a value lower than -0.5 dpt during the progression of myopia.

[0159] Example 41: The processing device according to any one of the previous five examples, wherein the high myopia risk refers to determining another probability that the person will develop high myopia by obtaining the refractive values defined as high myopia during the progression of these refractive values.

[0160] Example 42: The processing device according to the previous example, wherein high myopia refers to the refractive state of the person, and wherein the refractive power of at least one eye of the person is a value lower than -6.0 dpt.

[0161] Example 43: A system for providing data related to the progression of a person's refractive values, the system comprising

[0162] - at least one input interface configured to receive data related to the person according to one of the previous examples;

[0163] - a processing device according to one of the previous examples; and

[0164] - at least one output interface configured to provide data related to the progression of the person's refractive values.

[0165] Example 44: The system according to the previous example, wherein the at least one output interface is further configured to further provide at least one of the following:

[0166] o at least one percentile location of these refractive values;

[0167] o the corrected progression of the person's refractive values in consideration of implementing the at least one type of myopia treatment.

[0168] Example 45: The system according to the previous embodiment, including a graphical user interface, which is designated as at least one of the input interface and the output interface.

[0169] Example 46: The system according to any one of the previous system embodiments, wherein the at least one percentile positioning is provided for population-based data covering a range of ages.

[0170] Example 47: The system according to any one of the previous system embodiments, wherein the at least one percentile positioning includes at least one of the following:

[0171] - The 1st percentile, the 2nd percentile, the 3rd percentile or the 5th percentile;

[0172] - The 50th percentile; and

[0173] - At least one of the 95th percentile, the 97th percentile, the 98th percentile or the 99th percentile.

[0174] Example 48: The system according to any one of the previous system embodiments, wherein the corrected progression refers to the change process of the refractive value of the person in consideration of implementing the at least one type of myopia treatment.

[0175] Example 49: A computer-implemented method for determining data related to the progression of a person's refractive value, wherein the method includes the following steps:

[0176] - Receiving data related to the person, the data including

[0177] · The refractive state of the person;

[0178] · The age, gender and race of the person; and

[0179] · At least one risk factor related to the person;

[0180] - Using at least one machine learning algorithm to determine data related to the progression of the person's refractive value, wherein the at least one machine learning algorithm includes at least one prediction model for determining the relationship between the data related to the person and the data related to the progression of the person's refractive value.

[0181] Example 50: The method according to the previous embodiment, wherein the at least one machine learning algorithm includes using a first prediction model and a second prediction model, wherein the first prediction model generates intermediate prediction data, and wherein the intermediate prediction data is used as the input of the second prediction model.

[0182] Example 51: The method according to the previous embodiment,

[0183] o wherein the first prediction model uses longitudinal data, and wherein the longitudinal data includes a plurality of first items of data related to a specific person; and

[0184] o wherein the second prediction model uses cross-sectional data, and wherein the cross-sectional data includes at least one second item of data related to a plurality of different people.

[0185] Example 52: A computer-implemented method for providing data related to the progression of a person's refractive value, the method comprising the steps of:

[0186] - Receiving data related to the person via at least one input interface according to one of the preceding method claims;

[0187] - Determining data related to the progression of the person's refractive value via at least one processing device according to any one of the preceding method claims; and

[0188] - Providing the data related to the progression of the person's refractive value via at least one output interface.

[0189] Example 53: A computer program comprising instructions which, when executed by a computer, cause the computer to perform the method according to any one of the preceding method embodiments.

[0190] Example 54: A method for producing at least one spectacle lens, wherein producing the at least one spectacle lens includes processing at least one lens blank by using data related to the refractive value, and the data is determined by the method for determining data related to the progression of a person's refractive value according to any one of the preceding embodiments of the method for determining data related to the progression of a person's refractive value. BRIEF DESCRIPTION OF THE DRAWINGS

[0191] Preferably in conjunction with the dependent claims, other alternative features and embodiments of the present invention are disclosed in more detail in the following description of the preferred embodiments. Among them, as those skilled in the art will recognize, the corresponding alternative features can be implemented in isolation and in any feasible combination. It is emphasized here that the scope of the present invention is in no way limited by the preferred embodiments. In the drawings:

[0192] Figure 1 An exemplary embodiment of a system for providing data related to the progression of a refractive value according to the present invention is illustrated;

[0193] Figure 2Illustrates an exemplary embodiment of a computer-implemented method for providing data related to the progression of a person's refractive value according to the present invention; and

[0194] Figure 3 Illustrates an exemplary embodiment of a graphical user interface designated as an input interface and an output interface. Detailed Description

[0195] Figure 1 Illustrates an exemplary embodiment of a system 110 for providing output data 112 according to the present invention, the output data including a prediction of the progression 114 of a person's refractive value. Herein, the person can be a child, adolescent or young adult aged 4 to 24 years, especially 5 to 20 years; however, it may also be feasible to apply the present invention to persons of different ages. The prediction can cover a period of time, especially many years, preferably 1 to 12 years, more preferably 2 to 10 years, particularly 4 to 8 years; however, it may also be feasible to use different time periods.

[0196] In particular, the prediction can be used as a speculation about the myopia progression and / or myopia onset of one or both eyes of a person. Herein, as described above, myopia progression indicates a decrease, especially a monotonic decrease, in the refractive power of one or both eyes of a person over a period of time. Further, myopia describes a refractive state in which the refractive power of one or both eyes of a person is less than -0.5 dpt, while high myopia describes a refractive state in which the refractive power of one or both eyes of a person is less than -6.0 dpt. Further, myopia onset refers to the time point at which the refractive power of one or both eyes of a person decreases from a value higher than -0.5 dpt to a value lower than -0.5 dpt during the myopia progression.

[0197] As Figure 1 schematically depicted, the system 110 includes at least one input interface 116, a processing device 120, and an output interface 122. The input interface is configured to receive input data 118 related to a person, the processing device is configured to determine output data 112 related to the progression 114 of the person's refractive value, and the output interface is configured to provide the output data 112 related to the progression 114 of the person's refractive value to one or more recipients 124. The one or more recipients 124 can be eye care professionals, such as opticians, optometrists or ophthalmologists; or end consumers, such as persons with myopia or related persons, especially the parents of the person or the caregiver of the person.

[0198] The input interface 116 is configured to receive input data 118, preferably in the form of an input file. In particular, in order to obtain the input data 118, the input interface 116 can preferably be implemented as a graphical user interface 126, which can be configured to obtain the required input data 118, for example, by allowing one or more recipients 124 to input the required input data 118, for example, by using a keyboard, a touch screen, and / or a microphone; however, other possibilities are conceivable. Preferred examples of the graphical user interface 126 are illustrated below in Figure 3 as shown.

[0199] Furthermore, the processing device 120 can preferably be configured to provide output data 112 in the form of a structured output file to the output interface 122. For this purpose, the output data 112 can be provided to one or more recipients 124 by using a screen, a printer, and / or a speaker. Preferably, the output interface 122 can be implemented by using the same graphical user interface 126, which can be further configured to provide the required output data 112 to one or more recipients 124. However, it is also possible to use a different graphical user interface.

[0200] The first communication interface 128 can be configured to provide communication between the input interface 116 and the processing device 120, while the second communication interface 130 can be configured to provide communication between the processing device 120 and the output interface 122. As Figure 1 schematically illustrated, each communication interface 128, 130 can be implemented as a unidirectional interface, which can be configured to forward the corresponding multiple data items in a wired element and / or wireless manner, preferably via encrypted data transmission, to the indicated single direction. However, another communication interface is also feasible.

[0201] According to the present invention, the processing device 120 is configured to receive human-related input data 118, wherein the input data 118 includes

[0202] · the refractive state of a person;

[0203] · the age, gender, and race of a person; and

[0204] · one or more risk factors related to a person.

[0205] In addition, the input data 118 can include one or more data items, preferably one or more types of myopia treatments to be applied to a person.

[0206] Further, the processing device 120 is configured to determine output data 112 related to the progression 114 of a person's refractive value by means of a plurality of machine learning algorithms 132, which include one or more prediction models 134 for determining the relationship between input data 118 related to the person and output data 112 related to the progression 114 of the person's refractive value. As described in more detail above and below, in addition to a person's refractive state, age, gender, and ethnicity, using one or more risk factors related to the person and / or one or more types of myopia treatment to be applied to the person as input data 118 significantly improves the determination of the progression 114 of the person's refractive value as determined by using the processing device 120.

[0207] As Figure 1 schematically depicted further, the machine learning algorithms 132 may preferably use: a first prediction model 136 that uses longitudinal data as input data 118 and a second prediction model 138 that uses cross-sectional data as input data 118. For additional details regarding different types of prediction models 136, 138 and different kinds of data, reference may be made to the description below.

[0208] Figure 2 An exemplary embodiment of a computer-implemented method 210 for providing output data 112 related to the progression 114 of a person's refractive value in accordance with the present invention is illustrated.

[0209] In a receiving step 212, input data 118 related to the person is received by an input interface 116, such as by a recipient using a graphical user interface 126; however, other possibilities are conceivable. Herein, the input data 118 may preferably be compiled in the form of an input matrix x and forwarded to the processing device 120. By way of example, the input matrix x may include the following entries:

[0210] - A value indicating the current refractive state of the person, preferably the spherical lens values of the person's two eyes;

[0211] - A value indicating the age of the person;

[0212] - A number indicating the gender of the person;

[0213] - A number indicating the ethnicity of the person;

[0214] - Values indicating the current refractive states of both parents of the person, preferably the spherical lens values of both eyes of the person's parents;

[0215] - A value indicating a first amount of time the person spends on near vision work;

[0216] - A value indicating a second amount of time the person spends outdoors; and

[0217] - A value indicating an optical lens type selected from a specific contact lens or a specific spectacle lens as the type of myopia treatment to be applied to a person.

[0218] Alternatively or additionally, the input matrix x may include other entries or additional entries, provided that the minimum number of entries according to the present invention is included herein.

[0219] In the determination step 214, output data 112 related to the progression 114 of the refractive value of a person is determined based on the input data 118. Herein, the input data 118 is used, which is preferably compiled in the form of an input matrix x and particularly forwarded to the processing device 120 via the first communication interface 128. For this purpose, a machine learning algorithm 132 is used, which includes one or more prediction models 134 for determining the relationship between the input data 118 related to a person and the output data 112 related to the progression 114 of the refractive value of the person. As described above, the machine learning algorithm 132 may preferably use a first prediction model 136 that employs longitudinal data as the input data 118 and a second prediction model 138 that employs cross-sectional data as the input data 118.

[0220] In a particularly preferred embodiment, the determination step 214 may include a first prediction step 216, wherein the machine learning algorithm 132 may use the first prediction model 136 to predict the ratio R of the axial length divided by the corneal radius data, specifically by using support vector regression (SVR). For this purpose, equation (1) may be used

[0221]

[0222] wherein,

[0223] - R is the ratio of the axial length divided by the corneal radius data predicted for a given input matrix as described above;

[0224] - N is the total number of trained support vectors, where each support vector includes the matrix x n and the Lagrange multiplier α of this vector n and α n * ;

[0225] - G(xn,x) is the kernel function for the first prediction model 136, where the kernel function may be selected from a linear function or a non-linear function; and

[0226] - b is the bias value, which is determined and stored during the support vector training process.

[0227] Furthermore, in this particularly preferred embodiment, the determination step 214 may include a second prediction step 218, wherein the machine learning algorithm 132 may use the second prediction model 138 to predict the progression 114 of a person's refractive value, specifically by using Gaussian Process Regression (GPR) according to the following equation (2):

[0228] g = K(y, y') * A, (Equation 2)

[0229] wherein,

[0230] - y is the input vector of the GPR, wherein the input vector corresponds to the input matrix x. However, the value indicating the current refractive state of the person, preferably the spherical lens values of the person's two eyes, is replaced by the ratio R of the axial length divided by the corneal radius data, and this ratio is obtained by using the aforementioned first prediction model 136 according to Equation 1 as intermediate prediction data;

[0231] - y' is the trained active set vector of the Gaussian process regression;

[0232] - A is the weight vector of each trained active set vector;

[0233] - * indicates the inner product calculation; and

[0234] - K(y, y') is the kernel for the Gaussian process regression, and for this purpose, various functions can be used, preferably the quadratic rational kernel according to Equation (3):

[0235]

[0236] wherein, σ f 、σl and α are the parameters of the quadratic rational kernel, and all of them are calculated and stored during the training process.

[0237] Furthermore, the determination step 214 may include a risk consideration step 220, which may be designated to consider one or more risk factors included in the input matrix x, specifically

[0238] - the values indicating the current refractive states of both parents of the person, preferably the spherical lens values of both eyes of the person's parents;

[0239] - the value indicating the first amount of time the person spends on near vision work; and

[0240] - the value indicating the second amount of time the person spends outdoors.

[0241] Furthermore, the determination step 214 may include a myopia treatment consideration step 222, which may be designated to consider one or more types of myopia treatment included in the input matrix x, specifically

[0242] - A value indicating an optical lens type selected from a specific contact lens or a specific spectacle lens as the type of myopia treatment to be applied to a person.

[0243] In the providing step 224, data related to the refractive value 112 is provided to one or more recipients 124 via the output interface 122 in the manner described in more detail above, particularly via the graphical user interface 126, specifically for further processing.

[0244] Figure 3 An exemplary embodiment of the graphical user interface 310 designated herein as the input interface 116 and the output interface 122 is illustrated.

[0245] Accordingly, the graphical user interface 310 has a first partition 312 designed as the input interface 116, which is configured to receive input data 118 related to a person. Here, the input data 118 (specifically the age 314; gender 316; race 318; refractive state 320, particularly the refractive power; refractive state 322 of at least one parent, particularly several myopic parents; a first amount of time spent on near vision work 324; a second amount of time the person spends outdoors 326; and the proposed myopia treatment 328) can be adjusted for input into the input interface 116.

[0246] After pressing the determination button 330 further included in the graphical user interface 310, a second partition 332 of the graphical user interface 310 presents output data 112 related to the progression 114 of the person's refractive value, particularly the transmitted ranking 334, general myopia status 336, myopia risk 338, and high myopia risk 340, and presents a graph 342 that shows a prediction of the progression 114 of the person's refractive value as a function of the person's age 314, where, in addition to the proposed myopia treatment 328, the input data 118 mentioned above is also considered. Here, the ranking 334 can indicate the position of the person compared to others of the same age. The general myopia status 336 can be selected from the qualifiers "good", "medium", or "poor", depending on whether the prediction of the progression 114 of the person's refractive value predicts no myopia ("good"), myopia ("medium"), or high myopia ("poor"). The values of the myopia risk 338 and the high myopia risk 340 are determined in the manner described in more detail above.

[0247] As Figure 3As further depicted in, the chart 342 shown in the second partition 332 of the graphical user interface 310 further presents reference curves 344, 346, 348, which represent the 97th percentile, the 50th percentile, and the 3rd percentile of multiple people of the same age 314, as well as the corrected progression 350 of the refractive values of the people affected by the proposed myopia treatment 328 and input into the input interface 116.

[0248] In addition, the inventors conducted a study which showed that, using machine learning 132 and a large input data set 118 obtained for Chinese children, an algorithm for predicting the spherical refractive power as a function of age 314 could be developed. Among the algorithms showing acceptable performance, support vector regression (SVR) and Gaussian process regression (GPR) were used as the first prediction model 136 and the second prediction model 138, respectively. Performance evaluation showed that the correlation value between the prediction and the measured real data was acceptable, the bias value was far lower than 0.25 dpt, and the degree of consistency was very high, allowing for an easy distinction of children at risk of myopia development and progression.

[0249] List of reference numerals

[0250] 110 System

[0251] 112 Output data

[0252] 114 Progression of the refractive value of a person

[0253] 116 Input interface

[0254] 118 Input data

[0255] 120 Processing device

[0256] 122 Output interface

[0257] 124 Receiver

[0258] 126 Graphical user interface

[0259] 128 First communication interface

[0260] 130 Second communication interface

[0261] 132 Machine learning algorithm

[0262] 134 Prediction model

[0263] 136 First prediction model

[0264] 138 Second prediction model

[0265] 210 Computer-implemented for providing output data related to the progression of the refractive value of a person

[0266] method

[0267] 212 receiving step

[0268] 214 determination step

[0269] 216 first prediction step

[0270] 218 second prediction step

[0271] 220 risk consideration step

[0272] 222 myopia treatment consideration step

[0273] 224 providing step

[0274] 310 graphical user interface

[0275] 312 first partition

[0276] 314 age

[0277] 316 gender

[0278] 318 race

[0279] 320 refractive state of a person

[0280] 322 refractive state of at least one parent of a person

[0281] 324 first amount of time spent on near vision work

[0282] 326 second amount of time a person spends outdoors

[0283] 328 myopia treatment (type)

[0284] 330 determination button

[0285] 332 second partition

[0286] 334 ranking

[0287] 336 general myopia status

[0288] 338 myopia risk

[0289] 340 high myopia risk

[0290] 342 chart

[0291] 344 97th percentile

[0292] 346 50th percentile

[0293] 348 3rd percentile

[0294] Corrected Progression of Refractive Values for 350 Persons

Claims

1. A processing device (120) for determining data related to the progression (114) of a person's refractive value, wherein, The progression (114) of these refractive values is a speculation of the temporal change of the refractive value of at least one eye of the person over a period of time, wherein the processing device (120) is configured to: o Receive at least one input file including data related to the person, the data including · The refractive state (320) of the person; · The age (314), gender (316), and ethnicity (318) of the person; and · At least one risk factor related to the person, o Provide at least one output file including data related to the progression (114) of the refractive value of the person determined by using at least one machine learning algorithm (132), wherein the at least one machine learning algorithm (132) is configured to determine, in a determination step (214), data related to the progression (114) of the refractive value of the person based on the data from the at least one input file according to the data related to the person, wherein the at least one machine learning algorithm (132) includes at least one prediction model (134) for determining the relationship between the data related to the person and the data related to the progression (114) of the refractive value of the person, Characterized in that The at least one machine learning algorithm (132) includes using a first prediction model (136) and a second prediction model (138), wherein, in a first prediction step (216) of the determination step (214), the first prediction model (136) generates intermediate prediction data including the ratio of the axial length of the person to the corneal radius by using support vector regression (SVR), wherein the intermediate prediction data is used as the input of the second prediction model (138), wherein, in a second prediction step (218) of the determination step (214), the second prediction model (138) predicts the progression (114) of the refractive value of the person, wherein the second prediction model (138) is a second linear prediction model using Gaussian process regression (GPR).

2. The processing device (120) according to claim 1, wherein, The at least one risk factor is selected from data related to at least one of the following: · The refractive state (322) of at least one parent of the person; · At least one parameter related to the behavior of the person.

3. The processing device (120) according to claim 2, wherein, The at least one parameter related to the behavior of the person is selected from data related to at least one of the following: · A first amount of time the person spends on near vision work (324); · A second amount of time the person spends outdoors (326).

4. The processing device (120) according to any one of claims 1 to 3, wherein, The data related to the person further includes at least one type of myopia treatment (328), wherein the at least one type of myopia treatment (328) is selected from the application of at least one of the following: · An optical lens selected from contact lenses or spectacle lenses; · A certain dose of medicine; · Refractive surgery, And wherein the refractive state (320) is selected from at least one of the following: · At least one refractive value of at least one eye; · At least one biometric value of the at least one eye.

5. The processing device (120) according to any one of claims 1 to 3, o Among them, The first prediction model (136) uses longitudinal data, where the longitudinal data includes a plurality of first item data related to a specific person; o Wherein, the second prediction model (138) uses cross-sectional data, where the cross-sectional data includes at least one second item data related to a plurality of different people.

6. The processing device (120) according to claim 5, wherein, The total data input of the at least one machine learning algorithm (132) includes a first amount of longitudinal data input and a second amount of cross-sectional data input, where the first amount is 30% to 70%, and the second amount is 30% to 70%, and the first amount and the second amount add up to 100%.

7. The processing device (120) according to any one of claims 1 to 3, wherein, The processing device (120) is further configured to determine at least one of the following: · The ranking (334) of the person compared to a plurality of other people; · The myopia risk (338) of the person; · The high myopia risk (340) of the person.

8. A system (110) for providing data related to the progression (114) of a person's refractive value, wherein, The progression (114) of these refractive values is a speculation of the temporal change of the refractive values of at least one eye of the person over a period of time, where the system includes: - At least one input interface (116), the at least one input interface being configured to receive at least one input file including data related to a person according to one of claims 1 to 7; - A processing device (120), where the processing device (120) is configured to o receive the at least one input file including the data related to the person, the data including · The refractive state (320) of the person; · The age (314), gender (316) and race (318) of the person; and · At least one risk factor related to the person; o Provide at least one output file, the at least one output file including data related to the progression (114) of the refractive values of the person determined by using at least one machine learning algorithm (132), where the at least one machine learning algorithm (132) is configured to determine, in the determination step (214), the data related to the progression (114) of the refractive values of the person according to the data from the at least one input file based on the data related to the person, where the at least one machine learning algorithm (132) includes at least one prediction model (134) for determining the relationship between the data related to the person and the data related to the progression (114) of the refractive values of the person, - At least one output interface (122), the at least one output interface being configured to provide data related to the progression (114) of the refractive values of the person, Characterized in that The at least one machine learning algorithm (132) includes using a first prediction model (136) and a second prediction model (138), wherein, in a first prediction step (216) of the determination step (214), the first prediction model (136) generates intermediate prediction data including a ratio of the axial length of the person divided by the corneal radius by using support vector regression (SVR), wherein the intermediate prediction data is used as an input to the second prediction model (138), wherein, in a second prediction step (218) of the determination step (214), the second prediction model (138) predicts the progression (114) of the refractive value of the person, and wherein the second prediction model (138) is a second linear prediction model using Gaussian process regression (GPR).

9. The system (110) according to claim 8, wherein, The at least one output interface is further configured to provide at least one of the following: · At least one percentile location of these refractive values, wherein the at least one percentile location is provided for population-based data covering a range of ages; · A corrected progression (350) of the refractive value of the person in consideration of implementing the at least one type of myopia treatment (328).

10. A computer-implemented method for determining data related to the progression (114) of a person's refractive value, wherein, The progression (114) of these refractive values is a speculation of the temporal change of the refractive value of at least one eye of the person over a period of time, and the method includes the following steps: o Receiving at least one input file including data related to the person, the data including · The refractive status (320) of the person; · The age (314), gender (316), and race (318) of the person; and · At least one risk factor related to the person; - Providing at least one output file including data related to the progression (114) of the refractive value of the person determined by using at least one machine learning algorithm (132), wherein the at least one machine learning algorithm (132) is configured to determine, in a determination step (214), data related to the progression (114) of the refractive value of the person based on the data from the at least one input file according to the data related to the person, and wherein the at least one machine learning algorithm (132) includes at least one prediction model (134) for determining the relationship between the data related to the person and the data related to the progression (114) of the refractive value of the person, characterized in that The at least one machine learning algorithm (132) includes using a first prediction model (136) and a second prediction model (138), wherein, in a first prediction step (216) of the determination step (214), the first prediction model (136) generates intermediate prediction data including a ratio of the axial length of the person divided by the corneal radius by using support vector regression (SVR), wherein the intermediate prediction data is used as an input to the second prediction model (138), and wherein, in a second prediction step (218) of the determination step (214), the second prediction model (138) predicts the progression (114) of the refractive value of the person, and wherein the second prediction model (138) is a second linear prediction model using Gaussian process regression (GPR).

11. The method according to claim 10, wherein, Using the at least one machine learning algorithm (132) includes using a first prediction model (136) and a second prediction model (138), wherein the first prediction model (136) generates intermediate prediction data, and wherein the intermediate prediction data is used as an input to the second prediction model (138).

12. A computer-implemented method (210) for providing data related to the progression (114) of a person's refractive value, wherein, The progression (114) of these refractive values is a speculation of the temporal change of the refractive value of at least one eye of the person over a period of time, and wherein the method (210) includes the following steps: - receiving, by using at least one input interface (116), at least one input file including data related to the person according to one of claims 10 to 11 of the method claims; - determining, by using at least one processing device (120), data related to the progression (114) of the refractive value of the person according to any one of claims 10 to 11 of the method claims; - providing, by using at least one output interface (122), the data related to the progression (114) of the refractive value of the person.

13. A computer program product, the computer program product including a computer program, the computer program including instructions which, when the program is executed by a computer, cause the computer to perform at least one step of the method according to any one of claims 10 to 12 of the method claims.

Images