KR102520600B1 - Method and device for recommending vision correction surgery - Google Patents

Abstract

The present invention relates to a method for recommending refractive surgery using artificial intelligence, which is performed by a computing device according to an aspect of the present invention. contains a value - obtains ; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data acquired from examination data of the subject into a second prediction model when the subject is suitable for refractive surgery; If the subject can undergo refractive surgery using a laser, the third group data obtained from the examination data of the subject is input into a third predictive model to be used to determine whether custom refractive surgery is necessary, thereby determining the subject's standard refractive surgery after standard refractive surgery. calculating a corneal shape factor predicted value and a corneal shape factor predicted value after custom refractive surgery; and if the subject can undergo vision correction surgery using a laser, inputting the fourth group data obtained from the examination data of the subject into a fourth predictive model and proposing a vision correction surgery corresponding to the subject. 4 The predictive model is learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and visual acuity after the refractive surgery of the plurality of recipients.

Description

Vision correction recommendation method and device {METHOD AND DEVICE FOR RECOMMENDING VISION CORRECTION SURGERY}

It relates to a method and device for recommending vision correction surgery, and more particularly, to a method and device for recommending vision correction surgery to a subject using artificial intelligence.

Vision correction procedures such as LASIK and LASEK are attracting much attention from people with poor eyesight, regardless of age or gender. Interest in vision correction surgery is increasing day by day to such an extent that there is a statistic that the number of people who have undergone vision correction surgery reaches 100,000.

However, it is difficult for a subject who wants to undergo vision correction surgery to determine which vision correction surgery is right for him or her. Basically, the subject has to choose the type of surgery such as LASIK, LASEK, SMILE, and lens implantation. In addition, within LASIK, selection according to surgical equipment such as Eye LASIK, Da Vinci LASIK, Crystal LASIK, Z LASIK, VIJU LASIK, Opti LASIK, and surgical methods such as Contura Vision, Extra LASIK, and Wavefront LASIK must be made. In addition, since the amount of corneal ablation recommended by hospitals and doctors varies, it is more difficult to select, and it is a reality that there is no choice but to rely on the words of a doctor or counselor about the quality of vision or side effects after surgery.

US 2018-0161098 A1

One task is to assist doctors in making judgments or to provide objective corrective surgery-related information to doctors, counselors, and subjects.

Another task is to recommend vision correction surgery to doctors, counselors, and subjects.

Another task is to provide reasons for recommending vision correction surgery to doctors, counselors, and subjects.

The problem to be solved is not limited to the above-described problem, and problems not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings.

According to one aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of an examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data acquired from examination data of the subject into a second prediction model when the subject is suitable for refractive surgery; If the subject can undergo refractive surgery using a laser, the third group data obtained from the examination data of the subject is input into a third predictive model to be used to determine whether custom refractive surgery is necessary, thereby determining the subject's standard refractive surgery after standard refractive surgery. calculating a corneal shape factor predicted value and a corneal shape factor predicted value after custom refractive surgery; and if the subject can undergo vision correction surgery using a laser, inputting the fourth group data obtained from the examination data of the subject into a fourth predictive model and proposing a vision correction surgery corresponding to the subject. 4 The predictive model is to provide a refractive surgery recommendation method learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after refractive surgery of the plurality of recipients. can

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of an examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data obtained from examination data of the subject into a second predictive model when the subject is suitable for vision correction surgery; predicting whether or not the subject needs custom refractive surgery by inputting third group data obtained from examination data of the subject into a third predictive model when the subject can undergo refractive surgery using a laser; and if the subject can undergo vision correction using a laser, inputting the fourth group data obtained from the examination data of the subject to a fourth predictive model and proposing a corresponding vision correction surgery to the subject, wherein the custom In the step of predicting the need for refractive surgery, the need for custom refractive surgery is predicted based on the predictive value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery of the subject, and the fourth predictive model, A method for recommending refractive surgery learned based on at least one of examination data of a plurality of recipients who have undergone corrective surgery, refractive surgery corresponding to the plurality of recipients, and vision after refractive surgery of the plurality of recipients may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data acquired from examination data of the subject into a second prediction model when the subject is suitable for refractive surgery; and if the subject can undergo vision correction using a laser, inputting the third group data obtained from the examination data of the subject into a third predictive model and proposing a corresponding vision correction surgery to the subject, wherein the visual acuity In the proposing of the corrective surgery, the optometry is suggested based on the predictive value of the corneal shape factor after the standard refractive surgery and the predicted value of the corneal shape factor after the custom refractive surgery of the subject, and the third prediction model is performed on a plurality of recipients who have undergone the refractive surgery. A method for recommending refractive surgery that is learned based on at least one of checkup data, refractive surgery corresponding to the plurality of recipients, and eyesight after refractive surgery of the plurality of recipients may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; and, when corrective surgery is suitable for the subject, inputting the second group data obtained from the examination data of the subject into a second predictive model, and suggesting a corrective surgery corresponding to the subject. The step of performing is to propose vision correction surgery based on the predicted value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery of the subject, and the second predictive model is the examination data of a plurality of subjects who have undergone refractive surgery. , Revision surgery recommendation method learned based on at least one of vision correction corresponding to the plurality of subjects and eyesight after vision correction of the plurality of subjects may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; and inputting the group data obtained from the examination data of the subject into a predictive model to suggest a vision correction operation corresponding to the subject, wherein the step of suggesting the vision correction operation is performed on the cornea after standard vision correction surgery of the subject. Corrective surgery is proposed based on the predicted value of the shape factor and the predicted value of the corneal shape factor after custom refractive surgery. A method for recommending refractive surgery learned based on at least one of eyesight after the refractive surgery of the subject may be provided.

The solution to the problem is not limited to the above-described solution, and solutions not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings. .

According to an embodiment, a method for assisting a doctor's judgment or providing objective corrective surgery-related information to a doctor, counselor, and an examinee, and an apparatus for performing the method may be provided.

According to another embodiment, vision correction surgery may be recommended to doctors, counselors, and subjects using artificial intelligence.

According to another embodiment, it is possible to provide reasons for recommending vision correction surgery to doctors, counselors, and subjects using artificial intelligence.

Effects are not limited to the above effects, and effects not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings.

1 is a diagram of a vision correction assisting system according to an embodiment.
2 is a diagram for explaining a learning device/prediction device according to an exemplary embodiment.
3 is a diagram for explaining a server device and a client device according to an exemplary embodiment.
4 is a diagram of configurations of a server device and a client device according to an embodiment.
5 is a diagram of a model related to vision correction surgery according to an embodiment.
6 is a diagram for explaining a learning step and a prediction step of a model related to vision correction surgery according to an embodiment.
7 is a diagram related to pre-processing of input/output data according to an embodiment.
8 is a diagram of a model related to vision correction including serially connected submodels according to an embodiment.
9 is a diagram of a model related to vision correction including sub-models connected in parallel according to an embodiment.
10 is a diagram of an expected visual acuity image according to an exemplary embodiment.
11 is a diagram of an expected visual acuity image generation model using a filter according to an exemplary embodiment.
12 is a diagram of a corneal topography image according to an exemplary embodiment.
13 is a diagram related to an analysis model for a cause of predictive result calculation including an optometry-related model according to an embodiment.
14 is a diagram related to a cause of calculating a prediction result according to an embodiment.
15 is a diagram related to serial connection of models related to vision correction surgery according to an embodiment.
16 to 19 are diagrams for output calculation of a model related to vision correction surgery based on an output of a corneal shape factor prediction model according to an embodiment.
20 is a view related to calculating an output of a model related to refractive surgery based on an output of a custom refractive surgery necessity prediction model according to an embodiment.
21 and 22 are diagrams for calculating an output of a model related to vision correction surgery based on an output of a proposed model for vision correction surgery according to an embodiment.
23 is a view related to calculating an output of a model related to vision correction surgery based on an output of a surgical parameter proposal model according to an embodiment.
24 to 25 are diagrams for calculating an output of a model related to vision correction based on an output of a vision prediction model according to an embodiment.
26 to 27 are diagrams related to output calculation of an optometry-related model based on an output of a corneal topography image prediction model according to an embodiment.
28 is a diagram for explaining a combination of three or more vision correction-related models according to an embodiment.
29 to 30 are diagrams for output calculation of an optometry-related model based on outputs of a corneal shape factor prediction model and a visual acuity prediction model according to an embodiment.
31 is a diagram for explaining merging of vision correction related models according to an embodiment.
32 to 34 are diagrams for implementation examples of merging models related to vision correction according to an embodiment.
35 is a diagram related to a first embodiment of a vision correction surgery recommendation method according to an embodiment.
36 is a diagram related to a second embodiment of a vision correction surgery recommendation method according to an embodiment.
37 is a diagram related to a third embodiment of a method for recommending vision correction surgery according to an embodiment.
38 is a diagram related to a fourth embodiment of a vision correction surgery recommendation method according to an embodiment.
39 is a diagram related to a fifth embodiment of a vision correction surgery recommendation method according to an embodiment.
40 is a diagram related to a sixth embodiment of a vision correction surgery recommendation method according to an embodiment.
41 is a diagram related to a first embodiment of a method for providing vision correction visualization information according to an embodiment.
42 is a diagram related to a second embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.
43 is a diagram related to a third embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.
44 is a diagram related to a fourth embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.

The embodiments described in this specification are intended to clearly explain the spirit of the present invention to those skilled in the art to which the present invention belongs, and the present invention is not limited by the embodiments described in this specification, and the present invention The scope of should be construed as including modifications or variations that do not depart from the spirit of the present invention.

The terms used in this specification have been selected as general terms that are currently widely used as much as possible in consideration of the functions in the present invention, but they may vary depending on the intention, custom, or the emergence of new technologies of those skilled in the art in the technical field to which the present invention belongs. can However, in the case where a specific term is defined and used in an arbitrary meaning, the meaning of the term will be separately described. Therefore, the terms used in this specification should be interpreted based on the actual meaning of the term and the overall content of this specification, not the simple name of the term.

The drawings accompanying this specification are intended to easily explain the present invention, and the shapes shown in the drawings may be exaggerated as necessary to aid understanding of the present invention, so the present invention is not limited by the drawings.

If it is determined that a detailed description of a known configuration or function related to the present invention in this specification may obscure the gist of the present invention, a detailed description thereof will be omitted if necessary. In addition, numbers (eg, first, second, etc.) used in the description process of this specification may mean different configurations, but may correspond to the same configuration.

Hereinafter, a method and apparatus for recommending vision correction surgery to a subject based on examination data will be described. In particular, a method and apparatus for generating a model recommending vision correction surgery using artificial intelligence and recommending vision correction surgery to a subject using the generated model will be described.

In addition, a method and apparatus for providing visualized corrective surgery-related information to a subject based on examination data will be described. In particular, by using artificial intelligence, a model that provides visualization information for refractive surgery, such as providing an expected visual acuity image after refractive surgery, predicting a corneal topography image, or analyzing the cause of the prediction result, is created, and the generated model is used to provide information to the subject. A method and apparatus for providing vision correction visualization information will be described.

In this specification, vision correction surgery is a surgery that corrects the eyesight of the subject, and includes not only surgery to correct vision through corneal ablation using lasers such as LASIK, LASEK, and SMILE (small incision lenticule extraction, SMILE), but also lens It should be interpreted broadly, including vision correction procedures that do not use lasers, such as implantation.

In addition, in the present specification, visual acuity includes visual acuity that can be measured based on a subject's judgment and visual acuity that can be measured through an eye test or the like. For example, visual acuity may be measured through an eye test chart. Or, vision is lower-order aberrations, which are basic refractive errors such as myopia, hyperopia, and astigmatism, and higher-order aberrations, such as spherical aberration, coma aberration, and trefoil aberration. aberrations) may be included. In addition, visual acuity may include uncorrected visual acuity and corrected visual acuity.

According to one aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of an examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data acquired from examination data of the subject into a second prediction model when the subject is suitable for refractive surgery; If the subject can undergo refractive surgery using a laser, the third group data obtained from the examination data of the subject is input into a third predictive model to be used to determine whether custom refractive surgery is necessary, thereby determining the subject's standard refractive surgery after standard refractive surgery. calculating a corneal shape factor predicted value and a corneal shape factor predicted value after custom refractive surgery; and if the subject can undergo vision correction surgery using a laser, inputting the fourth group data obtained from the examination data of the subject into a fourth predictive model and proposing a vision correction surgery corresponding to the subject. 4 The predictive model is to provide a refractive surgery recommendation method learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after refractive surgery of the plurality of recipients. can

Here, in the suggesting of the vision correction surgery, the vision correction surgery may be suggested based on the predicted visual acuity value after the vision correction surgery of the subject.

Here, in the proposing the vision correction surgery, the vision correction surgery may be proposed by calculating a visual acuity prediction value after the vision correction surgery of the subject corresponding to each of the plurality of vision correction procedures.

Here, in the proposing the vision correction surgery, the vision correction surgery may be suggested by calculating a plurality of predictive values of visual acuity corresponding to a plurality of different times.

Here, the refractive surgery recommendation method predicts whether or not the subject needs custom refractive surgery based on the predicted corneal shape factor predicted values after standard refractive surgery and the calculated corneal shape factor predicted values after custom refractive surgery; may further include.

Here, in the proposing of the refractive surgery, the refractive surgery may be suggested based on the calculated corneal shape factor predicted value after standard refractive surgery and the calculated corneal shape factor predicted value after custom refractive surgery.

Here, when the fourth prediction model is learned in consideration of the plurality of subjects' preference for corrective surgery, the fourth group data may include the subject's preference for corrective surgery.

Here, at least one of the first prediction model, the second prediction model, the third prediction model, and the fourth prediction model includes a plurality of sub-models, and a result value is determined based on the results of the plurality of sub-models. can be calculated

Here, at least one of the first group data, the second group data, the third group data, and the fourth group data includes at least a portion of the examinee's examination data as it is or at least a portion of the examinee's examination data. It may be characterized by including a new type of data calculated from.

Here, the corneal shape factor prediction value may include at least one of an index of height decentration (IHD) prediction value, an index of surface variance (ISV) prediction value, and an index of vertical asymmetry (IVA) prediction value.

Here, the examination data may further include genetic information.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of an examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data obtained from examination data of the subject into a second predictive model when the subject is suitable for vision correction surgery; predicting whether or not the subject needs custom refractive surgery by inputting third group data obtained from examination data of the subject into a third predictive model when the subject can undergo refractive surgery using a laser; and if the subject can undergo vision correction using a laser, inputting the fourth group data obtained from the examination data of the subject to a fourth predictive model and proposing a corresponding vision correction surgery to the subject, wherein the custom In the step of predicting the need for refractive surgery, the need for custom refractive surgery is predicted based on the predictive value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery of the subject, and the fourth predictive model, A method for recommending refractive surgery learned based on at least one of examination data of a plurality of recipients who have undergone corrective surgery, refractive surgery corresponding to the plurality of recipients, and vision after refractive surgery of the plurality of recipients may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; predicting whether or not the subject can undergo refractive surgery by inputting second group data acquired from examination data of the subject into a second prediction model when the subject is suitable for refractive surgery; and if the subject can undergo vision correction using a laser, inputting the third group data obtained from the examination data of the subject into a third predictive model and proposing a corresponding vision correction surgery to the subject, wherein the visual acuity In the proposing of the corrective surgery, the optometry is suggested based on the predictive value of the corneal shape factor after the standard refractive surgery and the predicted value of the corneal shape factor after the custom refractive surgery of the subject, and the third prediction model is performed on a plurality of recipients who have undergone the refractive surgery. A method for recommending refractive surgery that is learned based on at least one of checkup data, refractive surgery corresponding to the plurality of recipients, and eyesight after refractive surgery of the plurality of recipients may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the examination data of the subject into a first prediction model; and, when corrective surgery is suitable for the subject, inputting the second group data obtained from the examination data of the subject into a second predictive model, and suggesting a corrective surgery corresponding to the subject. The step of performing is to propose vision correction surgery based on the predicted value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery of the subject, and the second predictive model is the examination data of a plurality of subjects who have undergone refractive surgery. , Revision surgery recommendation method learned based on at least one of vision correction corresponding to the plurality of subjects and eyesight after vision correction of the plurality of subjects may be provided.

According to another aspect, in a method for recommending refractive surgery using artificial intelligence performed by a computing device, the method comprising: acquiring examination data of a subject, wherein the examination data includes medical examination data and eyeball characteristic data measurement values; and inputting the group data obtained from the examination data of the subject into a predictive model to suggest a vision correction operation corresponding to the subject, wherein the step of suggesting the vision correction operation is performed on the cornea after standard vision correction surgery of the subject. Corrective surgery is proposed based on the predicted value of the shape factor and the predicted value of the corneal shape factor after custom refractive surgery. A method for recommending refractive surgery learned based on at least one of eyesight after the refractive surgery of the subject may be provided.

According to another aspect, in a method for providing an image of expected visual acuity after refractive surgery using artificial intelligence performed by a computing device, examination data of an examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values, is obtained step; The first group data obtained from the examination data of the subject is input into a first predictive model to predict the subject's eye characteristic data after refractive surgery - the eye characteristic data prediction value includes at least one of a visual acuity prediction value and a corneal shape factor prediction value - Calculating ; and generating an expected visual acuity image based on the prediction value of the eyeball characteristic data, wherein the first prediction model is performed on the eyeball characteristic data measurement values of a plurality of recipients who have undergone refractive surgery before surgery, and the plurality of recipients. A method for providing a predicted visual acuity image learned based on at least one of a corrective surgery parameter and a measured value of eyeball characteristic data after surgery of the plurality of recipients may be provided.

Here, the generating of the expected vision image may include calculating or selecting a filter based on the predicted value of the eyeball characteristic data, and applying the filter to an original image to generate the expected vision image. .

Here, the eyeball characteristic data prediction value includes a first eyeball characteristic data prediction value and a second eyeball characteristic data prediction value, and the predicted visual acuity image includes a first predicted visual acuity image generated based on the first eyeball characteristic data predicted value and the predicted visual acuity image. A second expected visual acuity image generated based on the predicted value of the second eyeball characteristic data and different from the first predicted visual acuity image may be included.

Here, each of the first predicted eyeball characteristic data value and the second predicted eyeball characteristic data value corresponds to an eyeball characteristic data predicted value after standard refractive surgery and an eyeball characteristic data predicted value after custom refractive surgery, and the first predicted visual acuity image and the second predicted vision image. Each of the expected visual acuity images may correspond to an expected visual acuity image after standard refractive surgery and an expected visual acuity image after custom refractive surgery.

Here, the expected visual acuity image may include information on at least one of sharpness, light smearing, contrast sensitivity, night vision, glare, double vision, and afterimages of the subject's expected visual acuity after refractive surgery.

Here, the method of providing the predicted visual acuity image further includes predicting a corneal topographic image of the subject after corrective surgery by inputting the second group data obtained from the examination data of the subject to a second prediction model, 2 The prediction model may be learned based on at least one of preoperative corneal topographic images of a plurality of recipients who have undergone refractive surgery, refractive surgery parameters performed on the plurality of recipients, and postoperative corneal topographic images of the plurality of recipients. there is.

Here, the expected visual acuity image providing method may further include calculating the dependence of the predicted value of the eyeball characteristic data on the first group data.

Here, the predicted eye characteristic data includes a first predicted eye characteristic data and a predicted second eye characteristic data, the dependence includes a dependence coefficient corresponding to at least a part of the first group data, and the dependence coefficient is , a first dependency coefficient corresponding to the predicted value of the first eyeball characteristic data and a second dependent coefficient corresponding to the predicted value of the second eyeball characteristic data but different from the first dependent coefficient.

Here, each of the first predicted eyeball characteristic data value and the second predicted eyeball characteristic data value corresponds to the predicted eyeball characteristic data value after standard refractive surgery and the predicted eyeball characteristic data value after custom refractive surgery, and the first dependence coefficient and the second dependent coefficient. Each of the coefficients may correspond to the dependence of the predicted value of eyeball characteristic data after standard refractive surgery on the first group data and the dependence of the predicted value of eyeball characteristic data after custom refractive surgery on the first group data.

Here, the expected visual acuity image providing method may further include outputting a dependency coefficient greater than a predetermined value among the dependency coefficients or outputting a predetermined number of dependency coefficients.

Here, the corneal shape factor prediction value may include at least one of an index of height decentration (IHD) prediction value, an index of surface variance (ISV) prediction value, and an index of vertical asymmetry (IVA) prediction value.

Here, the visual acuity prediction value may include at least one of a prediction value of lower-order aberrations and a prediction value of higher-order aberrations.

An optometry assistance system according to an embodiment may include a learning device and a prediction device. Here, the learning device and the prediction device may be computing devices including at least one control unit. Examples of the computing device may include, but are not limited to, desktops, laptops, tablet PCs, and smart phones.

1 is a diagram of an optometry assistance system 10 according to an embodiment. Referring to FIG. 1 , the learning apparatus 100 may learn and/or generate a model for calculating information related to vision correction surgery (hereinafter referred to as “revision surgery related model”) based on learning data. Here, the learning data is data necessary for learning and/or generating a model related to vision correction, such as numbers, letters, and images, and the expression method is not limited. For example, the learning data may include examination data and surgical parameters of a person undergoing refractive surgery.

The prediction device 300 may calculate a prediction result, which is information related to vision correction surgery, based on the corrective surgery related model and input data generated through the learning device 100 . Here, the input data is data that is the basis for calculating the prediction result, such as numbers, letters, and images, and the expression method is not limited.

Specific examples of learning data, input data, prediction results, and vision correction related models will be described later.

Although the learning device 100 and the prediction device 300 are shown as separate devices in FIG. 1 , the learning device 100 and the prediction device 300 may be the same device. For example, it is possible to learn/create a model related to vision correction within the same device and calculate a prediction result using the model. Alternatively, at least some components of the learning device 100 and at least some components of the prediction device 300 may have the same configuration.

Also, the vision correction assisting system according to an embodiment may include a plurality of learning devices and/or a plurality of prediction devices.

2 is a diagram for explaining a learning device/prediction device according to an exemplary embodiment. Referring to FIG. 2 , a learning device/prediction device according to an embodiment may include a memory unit 5000 and a control unit 1000.

A learning device/prediction device according to an embodiment may include a controller 1000 for controlling its operation. The control unit 1000 may include one or more of a CPU (Central Processing Unit), a RAM (Random Access Memory), a GPU (Graphic Processing Unit), one or more microprocessors, and other electronic components capable of processing input data according to predetermined logic. can include

The control unit 1000 may read system programs and various processing programs stored in the memory unit 5000 . For example, the control unit 1000 may develop a data processing process for performing a learning and prediction step of a model related to vision correction surgery, which will be described later, in RAM, and perform various processes according to the developed program. For example, the controller 1000 may perform learning of a model related to vision correction surgery. As another example, the controller 1000 may generate a prediction result using a model related to vision correction surgery.

A learning device/prediction device according to an embodiment may include a memory unit 5000 . The memory unit 5000 may store data necessary for learning, a learning model, and a learned vision correction related model. The memory unit 5000 may store parameters and variables of a model related to vision correction surgery.

The memory unit 5000 may include a non-volatile semiconductor memory, a hard disk, a flash memory, a RAM, a read only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), or other tangible non-volatile recording media. can be implemented as

The memory unit 5000 may store various processing programs, parameters for performing program processing, or such processing result data. For example, the memory unit 5000 includes a data processing process program for performing learning and prediction steps of a model related to vision correction, which will be described later, a diagnosis process program, parameters for executing each program, and data obtained according to the execution of these programs. (eg, processed data or predicted results), etc. may be stored.

A learning device/prediction device according to an embodiment may further include a communication unit 9000. The communication unit 9000 may communicate with an external device. For example, the communication unit 9000 of the learning device may communicate with the communication unit 9000 of the prediction device. The communication unit 9000 may perform wired or wireless communication. The communication unit 9000 may perform bi-directional or unidirectional communication.

The learning device/prediction device shown in FIG. 2 is only an example, and the configuration of the learning device/prediction device is not limited thereto.

An optometry assistance system according to an embodiment may include a server device and a client device. 3 is a diagram for explaining a server device 500 and client devices 700a and 700b according to an exemplary embodiment.

The server device 500 according to an embodiment may correspond to the aforementioned learning device/prediction device. The server device 500 according to an embodiment may learn, store, and/or execute an optometry-related model.

The client devices 700a and 700b according to an embodiment may correspond to the aforementioned learning device/prediction device. The client devices 700a and 700b according to an embodiment may learn, store, and/or execute a model related to vision correction surgery.

The client devices 700a and 700b according to an embodiment may obtain the learned correction surgery-related model from the server device 500 . For example, the client devices 700a and 700b may download a model related to vision correction from the server device 500 through a network.

The server device 500 according to an embodiment may calculate a prediction result based on input data obtained from the client devices 700a and 700b. For example, the client devices 700a and 700b may receive information about the examinee and transmit the information to the server device 500, and the server device 500 may calculate a prediction result through an optometry-related model based on the information about the examinee. there is. The server device 500 according to an embodiment may obtain input data from a plurality of client devices 700a and 700b.

The server device 500 according to an embodiment may transmit the calculated prediction result to the client devices 700a and 700b. For example, the client devices 700a and 700b may provide prediction results obtained from the server device 500 to doctors, counselors, and examinees. The server device 500 according to an embodiment may obtain feedback from the client devices 700a and 700b. The server device 500 according to an embodiment may transmit prediction results to a plurality of client devices 700a and 700b.

Client devices 700a and 700b according to an embodiment may request a prediction result from the server device 500 .

The client devices 700a and 700b according to an embodiment may transmit input data to the server device 500 . The client devices 700a and 700b according to an embodiment may change at least a part of the input data and transmit the changed data to the server device 500 . The client devices 700a and 700b according to an embodiment may provide the prediction results obtained from the server device 500 to doctors, counselors, and examinees. The client devices 700a and 700b according to an embodiment may change at least a part of the prediction results obtained from the server device 500 and provide the changed information to doctors, counselors, and examinees.

3 shows a relationship between one server device 500 and two client devices 700a and 700b, but is not limited thereto, and similarly in one or more server devices 500 and one or more client devices 700a and 700b. can be applied

4 is a diagram of configurations of a server device 500 and a client device 700 according to an embodiment. Referring to FIG. 4 , the server device 500 and the client device 700 may include memory units 5000a and 5000b, control units 1000a and 1000b, and communication units 9000a and 9000b. The server device 500 and the client device 700 may transmit and obtain information through the communication units 9000a and 9000b. As an example, the client device 700 may acquire a model related to vision correction surgery learned from the communication unit 9000a of the server device 500 through the communication unit 9000b. As another example, the client device 700 may transfer input data to the communication unit 9000a of the server device 500 through the communication unit 9000b, and the server device 500 may transfer the predicted result to the client through the communication unit 9000a. It can be transmitted to the communication unit 9000b of the device 700.

As described above, the refractive surgery related model is a model that calculates various information that can be considered before and after refractive surgery.

5 is a view of a model M related to vision correction surgery according to an embodiment. Referring to FIG. 5, the corrective surgery related model (M) includes a surgery suitability prediction model (M10), a laser surgery availability prediction model (M11), a corneal shape factor prediction model (M12), a custom refractive surgery necessity prediction model (M13), Corrective surgery proposal model (M14), surgical parameter proposal model (M15), visual acuity prediction model (M16), predicted visual acuity image generation model (M17), corneal topography image prediction model (M18), and cause analysis model for prediction results (M19) can include A detailed description of each model will be described later.

At least some of the corrective surgery related models M may be learned and/or generated in the same learning device. For example, the surgical suitability prediction model M10 and the laser surgery suitability prediction model M11 may be trained and/or generated in the same learning device. Alternatively, at least some of the vision correction related models M may be learned and/or generated in different learning devices.

At least some of the corrective surgery related models M may be executed in the same prediction device. For example, the surgical suitability prediction model M10 and the laser surgery suitability prediction model M11 may be executed in the same prediction device. Alternatively, at least some of the corrective surgery related models M may be executed in different prediction devices.

The model related to vision correction according to an embodiment may be learned and/or implemented as an artificial intelligence model/algorithm, and the learning and/or implementation method is not limited. For example, models related to vision correction include classification algorithms, regression algorithms, supervised learning, unsupervised learning, reinforcement learning, and support vector machines. ), decision tree, random forest, LASSO, AdaBoost, XGBoost, artificial neural network, etc. Learning and/or implementation through various machine learning models/algorithms and deep learning models/algorithms It can be.

A model related to vision correction surgery according to an embodiment may be learned and/or generated through a learning step. The refractive surgery-related model learned and/or generated through the learning step may calculate a prediction result through the prediction step.

6 is a diagram for explaining a learning step (S10) and a prediction step (S30) of a model related to vision correction surgery according to an embodiment. Referring to FIG. 6 , the learning step (S10) may include a learning data acquisition step (S110) and a model learning step (S150).

The learning data acquisition step ( S110 ) may be a step in which the learning device acquires learning data, which is data for learning and/or generating an optometry-related model.

The model learning step ( S150 ) may be a step of learning and/or generating a model related to vision correction surgery. In the model learning step (S150), the learning device may learn and/or generate the model based on the learning data obtained in the learning data acquisition step. For example, model parameters constituting the vision correction related model may be changed in the model learning step ( S150 ). Accuracy of the model may be improved according to the change of the parameter.

The refractive surgery-related model may be learned based on refractive surgery-related information of a subject who has undergone refractive surgery. For example, the model is based on at least one of examination data before vision correction surgery of a plurality of subjects who received refractive surgery, vision correction surgery parameters performed on the plurality of subjects, and examination data after vision correction surgery of the plurality of subjects. It can be.

Referring to FIG. 6 , the prediction step (S30) may include an input data acquisition step (S310) and a model execution step (S350).

The step of acquiring input data (S310) may be a step of acquiring input data, which is data that can be used for calculating a prediction result by using an optometry-related model, by the prediction device.

The model execution step (S350) may be a step of calculating a prediction result based on the corrective surgery model and input data learned and/or generated in the learning step (S10). For example, the prediction device may output a prediction result based on input data and model parameters obtained from the learning device.

Input data of the model related to vision correction may include all information acquired about the subject. Alternatively, the input data may include at least a part of information acquired about the subject. In addition, the input data may be the same or different depending on the vision correction related model.

The prediction result of the model related to vision correction may vary depending on the learning data. Alternatively, the accuracy of the vision correction related model may vary depending on the learning data.

Prediction results may vary depending on the type of true value included in the training data. For example, the prediction result may vary depending on whether the true value considers the subjective intention of the subject. Here, the subjective intention of the examinee may include preference, such as whether a specific vision correction surgery is preferred, and payment ability, such as how much money can be paid for the vision correction surgery. If the subjective intention of the subject is not considered, the predicted result may be a medically suggested result. On the other hand, when the subjective intention of the subject is considered, the predicted result may be the same as or different from the medically suggested result.

When the first learning data and the second learning data are obtained from different hospitals, from different doctors, or are learning data at different times, the first model learned and/or generated based on the first learning data and the first learning data 2 The second models learned and/or generated based on the learning data may be different from each other. As a result, the first prediction result output by the first model and the second prediction result output by the second model may be different from each other. Also, the accuracy of the first model and the second model may be different from each other.

However, even if the learning data is different, the same vision correction related model may be learned and/or generated. Alternatively, the same prediction result may be calculated even if the model related to vision correction is learned and/or generated with different learning data.

Learning data, input data, and predicted results (hereinafter referred to as "input/output data") may include variables such as examination data and surgical parameters. Examination data and surgical parameters can be expressed in various ways, such as numbers, letters, images, etc., and there is no limitation on the expression method. In addition, the image is not limited in its dimension, such as a 2-dimensional image or a 3-dimensional image.

Examination data may include medical examination data, which is information obtained by asking questions without going through equipment or examination, eyeball characteristic data, which is information about the eyeball, and genetic information, obtained through equipment or examination.

The medical examination data includes variables such as living environment such as gender, age, race, and region of residence, demographic characteristics such as occupation, income, education, education, and family size, medical and family history such as hypertension and diabetes, and preference for refractive surgery. can include

The eyeball characteristic data may include all kinds of information related to the eyeball. For example, the eyeball characteristic data may include variables such as visual acuity, intraocular pressure, and/or retinal examination results.

The eyeball characteristic data may include information about the physical shape of the eyeball. For example, the eyeball characteristic data includes white-to-white (WTW) distance, angle-to-angle (ATA) distance, internal anterior chamber depth (ACD), sulcus-to-sulcus (STS) distance, pupil size You can include variables such as As another example, the eyeball characteristic data may include information about the shape of the cornea, such as a corneal shape factor and a corneal topography image. Here, the corneal shape factor is a numerical value representing the physical shape of the cornea, such as index of surface variance (ISV), index of vertical asymmetry (IVA), keratoconus index (KI), central keratoconus index (CKI), minimum radius of curvature (Rmin ), index of height asymmetry (IHA), index of height decentration (IHD), and central cornea thickness. In addition, the corneal topography image is an image related to the shape of the cornea, and may include a corneal topography, an anterior corneal curvature image, a posterior corneal curvature image, a corneal thickness map, and the like.

Eyeball characteristic data may be obtained through tomography, topography, optical coherence tomography (OCT), ultrasound biomicroscopy (UBM) equipment such as Pentacam, CASIA2, AL-Scan, and OQAS, but is not limited thereto. Information obtainable through the above equipment and similar equipment and information derived therefrom may be included in the eyeball characteristic data.

Genetic information may be obtained through genetic testing or the like. Genetic information may be used to determine whether corrective surgery is suitable for a subject. Alternatively, genetic information may be used to predict side effects after vision correction surgery. For example, it is possible to predict the occurrence of corneal dystrophy through genetic information.

Surgical parameters are variables related to performing refractive surgery, such as LASIK, LASEK, SMILE, lens implantation, standard refractive surgery, custom refractive surgery, etc., flap thickness, flap diameter, flap Variables that can be modified during vision correction surgery, such as side cut angle, corneal cutting profile, suction time, optic zone, hinge structure such as hinge position, hinge angle, and hinge width, can be included.

Standard refractive surgery may mean refractive surgery for correcting lower order aberrations, and custom refractive surgery may mean refractive surgery for correcting lower order aberrations and higher order aberrations.

The types of surgical parameters considered may differ depending on the subject of standard vision correction surgery and custom vision correction surgery. For example, types of surgical parameters considered according to users of custom refractive surgery may include types of surgical parameters considered according to users of standard refractive surgery.

The number of surgical parameters that change according to the subject of standard refractive surgery may be different from that of custom refractive surgery. Standard refractive surgery and custom refractive surgery can be classified according to the number of surgical parameters that change in correspondence with the recipient. The number of surgical parameters (hereinafter referred to as "reference values") that change in correspondence to the subject of surgery who distinguishes standard vision correction surgery and custom vision correction surgery may be a predetermined value. For example, custom refractive surgery may be performed if the number of surgical parameters that change in correspondence with the subject is greater than or equal to a reference value, and standard refractive surgery may be performed if the number of surgical parameters that change according to the subject is less than or equal to the reference value.

Standard refractive surgery and custom refractive surgery may vary depending on the hospital, doctor, and/or time of performing the refractive surgery. For example, the reference value may vary depending on hospitals, doctors, and/or timing.

Custom refractive surgery may include contour vision surgery and wavefront surgery.

In the case of performing custom refractive surgery, the quality of vision may be improved compared to the case of performing standard refractive surgery. Quality of vision is a comprehensive term meaning good or bad visual acuity, visual acuity measured through an optometry chart, low and high order aberrations, as well as sharpness of vision, light smearing, contrast sensitivity, night vision, glare, double vision, afterimages, It can be judged based on other inconveniences.

The input/output data may include not only measurement values that can be acquired/measured through medical examination and examination, but also predicted values that can be calculated through an optometry-related model. For example, the ocular characteristic data may include not only the measured value of the subject's ocular characteristic data obtained/measured through an examination before the refractive surgery, but also the predicted value of the subject's ocular characteristic data after the refractive surgery calculated through a model related to the refractive surgery. there is.

The input/output data may be input to a model related to vision correction surgery through preprocessing. For example, the learning device may pre-process acquired input/output data and then use the acquired input/output data to learn a model related to vision correction surgery. Alternatively, the prediction device may calculate a prediction result by preprocessing the obtained input/output data and inputting the acquired input/output data to a model related to vision correction surgery.

Preprocessing should be broadly interpreted to include all changes applied to input/output data and is not limited to the examples disclosed herein.

Preprocessing may include selecting at least some of variables included in input/output data, such as feature selection. For example, preprocessing may include t-test, Gini index, information gain, relief, DistAUC, signal to noise, MRMR, Fisher score, Laplacian score, SPEC, and the like.

Preprocessing may include generating new variables from at least some of variables included in input/output data, such as feature extraction. For example, preprocessing may include principle component analysis, linear discriminant analysis, canonical correlation analysis, singular value decomposition, ISOMAP, locally linear embedding, and the like. As another example, preprocessing may include generating a spectrum from a numerical value or generating an image from a numerical value.

The pre-processing may include a processing method when input/output data does not include a variable required by an optometry-related model (or a learning device and/or a prediction device), such as processing a missing value. For example, preprocessing may include processing missing values as the average or mode value of the variable.

Accuracy of models related to vision correction can be improved through preprocessing. For example, the accuracy of an optometry-related model that includes the feature selection step may be better than that of a model that does not include the feature selection step. As another example, the accuracy of an optometry-related model that generates an image from numerical values and calculates a prediction result based on the image may be better than the accuracy of an optometry-related model that calculates a predictive result based on the numerical value.

7 is a diagram related to pre-processing of input/output data according to an embodiment. Referring to FIG. 7 , input/output data may be input to a model related to vision correction through a preprocessing step (S500). The vision correction related model may calculate a prediction result based on preprocessed input/output data (S700).

A model related to vision correction may include a plurality of sub-models (sub-models). A plurality of submodels may calculate prediction results based on input data.

The vision correction related model may include serially and/or parallelly connected submodels such as an ensemble model.

The vision correction related model may include a plurality of sub-models connected in series. Here, that the submodels are connected in series includes that the output of at least one submodel is calculated based on the output of at least one submodel, such that the output of the first submodel becomes the input of the second submodel. can do. Alternatively, that the submodels are serially connected may include sequentially passing through a plurality of submodels in order to obtain a prediction result from input data through an optometry-related model.

8 is a diagram of a model M related to vision correction including serially connected submodels M1 and M2 according to an embodiment. Referring to FIG. 8 , the corrective surgery related model M may include a first sub-model M1 and a second sub-model M2 connected in series. The first submodel M1 may calculate an output based on input/output data, and the second submodel M2 may calculate a prediction result based on the output of the first submodel M1.

The vision correction related model may include a plurality of sub-models connected in parallel. Here, that the submodels are connected in parallel may include that the output of a submodel does not affect the output of another submodel, such as the output of a first submodel not depending on the output of a second submodel.

Inputs of a plurality of submodels connected in parallel may be the same. Alternatively, inputs of a plurality of submodels connected in parallel may be different. For example, first examination data input to the first sub-model may be different from second examination data input to the second sub-model.

The first examination data may include at least some other variables compared to the second examination data. For example, the first checkup data may include a preference for corrective surgery, but the second checkup data may not include a preference for corrective surgery.

The first examination data includes the same type of variables as the second examination data, but the values may be different. For example, the first examination data and the second examination data include corneal thickness, but the values may be different due to different methods of acquiring the values (for example, measuring the corneal thickness with different devices). there is.

The vision correction related model may include an output submodel that calculates an output based on outputs of a plurality of submodels connected in parallel. For example, when outputs of a plurality of submodels are the same, the output submodels may provide the same output. As another example, when the outputs of the plurality of submodels are different, the output submodel may output a result of considering the outputs of the plurality of submodels at a constant ratio or provide a specific output among the plurality of outputs. As another example, the output submodel may output a result generated based on outputs of a plurality of submodels.

9 is a diagram of a model M related to vision correction including submodels M1 and M2 connected in parallel according to an embodiment. Referring to FIG. 9 , the corrective surgery related model M may include a first sub-model M1 and a second sub-model M2 connected in parallel. In addition, the corrective surgery related model M may include an output submodel M3 that calculates an output based on the outputs of the first submodel M1 and the second submodel M2. The first sub-model (M1) and the second sub-model (M2) calculate a first output and a second output, respectively, based on the input/output data, and the output sub-model (M3) is based on the first output and the second output. Prediction results can be calculated.

An example of a model related to vision correction including submodels connected in parallel may be an ensemble, but is not limited thereto.

Hereinafter, individual examples of vision correction-related models will be described.

The surgical suitability prediction model can predict whether or not corrective surgery is suitable for a subject. The surgical suitability prediction model may predict the surgical suitability based on the input data.

Applicability of vision correction surgery may mean medical suitability. Accordingly, the input data of the surgical suitability prediction model may not include the subject's preference for corrective surgery. Alternatively, the surgical suitability prediction model may predict surgical suitability without considering the subject's preference for corrective surgery.

Suitability for surgery may include whether surgery is possible and whether surgery is required. For example, the suitability prediction model for surgery may determine that a subject who does not need surgery is unsuitable for surgery because of his good eyesight. As another example, the surgical suitability prediction model may determine that a subject whose visual acuity cannot be increased through refractive surgery is unsuitable for surgery.

Suitability for surgery may be output as suitable/unsuitable for surgery. Alternatively, suitability for surgery may be output by digitizing or visualizing suitability for surgery.

The laser surgery availability prediction model can predict whether or not laser vision correction surgery is possible for a subject. Here, vision correction using a laser may refer to surgery for correcting vision through corneal ablation using a laser. The laser surgery availability prediction model may predict laser surgery availability based on input data.

The availability of laser vision correction surgery may mean medical availability. Accordingly, the input data of the laser surgery availability prediction model may not include the subject's preference for corrective surgery. Alternatively, the laser surgery availability prediction model may predict laser surgery availability without considering the subject's preference for refractive surgery.

Whether laser surgery is possible or not may be output as laser surgery possible/non-available. Alternatively, the possibility of laser surgery may be output by digitizing or visualizing the possibility of surgery.

The corneal shape factor prediction model can predict the corneal shape factor of a subject after refractive surgery. The model can predict one or more corneal shape factors. The corneal shape factor prediction model may predict the corneal shape factor based on the input data.

Input data of the corneal shape factor prediction model may include surgical parameters. For example, the input data may include the type of surgery as a surgical parameter. The model can predict corneal shape factors corresponding to input surgical parameters. For example, when the input data includes LASIK, LASEK, and SMILE as surgical parameters, the output of the model may include a post-LASIK corneal shaping factor prediction value, a post-LASIK corneal shaping factor prediction value, and a post-Smile corneal shaping factor prediction value. As another example, when the input data includes standard refractive surgery and custom refractive surgery as surgical parameters, the output of the model may include corneal shape factor predicted values after standard refractive surgery and corneal shape factor predicted values after custom refractive surgery.

The corneal shape factor prediction model can predict a corneal shape factor corresponding to a predetermined surgical parameter regardless of whether the input data includes the surgical parameter. For example, when the model is trained to predict the corneal shape factor after standard refractive surgery and the corneal shape factor after custom refractive surgery, even if the input data of the model does not include the surgical parameter, the model can predict the corneal shape factor after standard refractive surgery. The predicted value and the predicted value of the corneal shape factor after custom refractive surgery can be output.

Table 1 is an output of a corneal shape factor prediction model according to an embodiment. Referring to Table 1, the corneal shape factor prediction model may output IHD, ISV, and IVA values of the subject. In addition, the corneal shape factor prediction model includes the subject's IHD, ISV, and IVA measurement values obtained before refractive surgery, expected IHD, ISV, and IVA predicted values after standard refractive surgery, and predicted IHD, ISV, and IVA predicted values after custom refractive surgery. can do.

item today Expectations after Standard Vision Correction Surgery Expectations after custom vision correction surgery IHD 0.015 0.021 0.018 ISV 20.0 27.8 25.0 IVA 0.19 0.23 0.22

The custom refractive surgery necessity prediction model can predict whether a subject needs custom refractive surgery. The custom refractive surgery necessity prediction model may predict the custom refractive surgery necessity based on the input data.

The custom refractive surgery necessity prediction model can predict the need for custom refractive surgery without considering the subject's refractive surgery preference. Alternatively, the custom refractive surgery necessity prediction model may predict the need for custom refractive surgery in consideration of the subject's refractive surgery preference. Depending on whether the subject's preference for refractive surgery is considered, the output of the custom refractive surgery necessity prediction model may vary.

The custom refractive surgery necessity prediction model may determine whether or not custom refractive surgery is required for the subject based on ocular characteristic data such as a corneal shape factor and a corneal topography image.

The custom refractive surgery necessity prediction model may determine the need for custom refractive surgery based on the absolute value of the corneal shape factor. For example, the custom refractive surgery necessity prediction model may predict that the subject needs custom refractive surgery when the corneal shape factor is outside a certain range.

The custom refractive surgery necessity prediction model may determine the need for custom refractive surgery based on the relative values of corneal shape factors. For example, the custom refractive surgery necessity prediction model may determine the need for custom refractive surgery by comparing a difference between a corneal shape factor after standard refractive surgery and a corneal shape factor after custom refractive surgery.

The vision correction surgery proposal model may suggest vision correction surgery corresponding to the subject. For example, the model may output one vision correction procedure. As another example, the model may output a plurality of vision correction procedures. As another example, the model may output a plurality of vision correction procedures together with information about their priorities.

The refractive surgery proposal model may suggest refractive surgery based on input data.

The refractive surgery corresponding to the subject may mean the refractive surgery calculated by the refractive surgery proposed model without considering the subject's refractive surgery preference. Alternatively, the refractive surgery corresponding to the subject may mean the refractive surgery calculated by the refractive surgery proposed model in consideration of the subject's refractive surgery preference. The output of the proposed model for refractive surgery may vary depending on whether or not the subject's preference for refractive surgery is considered.

The output of the refractive surgery proposal model may be one that considers the need for custom refractive surgery. For example, the output may be a standard LASIK, custom LASIK, standard LASIK, custom LASIK, standard smile, or custom smile considering the need for custom refractive surgery.

Alternatively, the output of the refractive surgery proposal model may be one in which the need for custom refractive surgery is not considered. For example, the output may be one that does not consider the necessity of custom refractive surgery such as LASIK, LASEK, SMILE, or lens implantation.

The refractive surgery proposal model may suggest refractive surgery based on the quality of vision after refractive surgery. For example, the model may suggest vision correction surgery based on a visual acuity prediction value after vision correction surgery corresponding to a plurality of vision correction surgeries.

Table 2 is an output of an optometry proposed model according to an embodiment. Referring to Table 2, the proposed model for refractive surgery can output LASIK, LASEK, and Smile surgeries together with information about priorities such as fitness. In the case of Table 2, the value corresponding to smile is greater than the value corresponding to LASIK and LASEK, which may mean that the corrective surgery proposal model proposes smile as the first priority. In addition to the method of Table 2, information about the priority may be output in various ways indicating the priority.

Types of Vision Correction Surgery goodness of fit LASIK 15.45% Lasik 13.30% smile 71.25%

Surgical Parameter Proposal The model may suggest surgical parameters. The model may suggest one or more surgical parameters. The surgical parameter proposal model may propose surgical parameters based on input data.

The input data of the surgical parameter proposal model may include image-type data. Here, the image may be an image obtained through measurement and/or examination using equipment such as a corneal topography. Alternatively, the image may be an image generated through interpolation, extrapolation, or artificial intelligence based on measured values. For example, the image may be an image generated from corneal shape factor.

The results of surgery after refractive surgery performed based on the surgical parameters suggested by the surgical parameter proposal model based on images may be better than the surgical results after refractive surgery performed based on the surgical parameters suggested based on numerical values. For example, the result of surgery after refractive surgery performed based on the surgical parameters proposed by the model based on the corneal topographic image is the refractive surgery performed based on the surgical parameters suggested based on corneal shape factors such as IHD, ISV, and IVA. The results after surgery may be better. Here, the surgical result may mean the quality of visual acuity after surgery. Alternatively, the surgical result may refer to the shape of the cornea after surgery.

The visual acuity prediction model can predict the visual acuity of a subject after refractive surgery. The vision prediction model may output a vision prediction value based on input data. The model may predict one or more visions.

Input data of the vision prediction model may include surgical parameters. For example, the input data may include the type of surgery as a surgical parameter. The model can predict visual acuity corresponding to the input surgical parameters. For example, when the input data includes LASIK, LASEK, and SMILE as surgical parameters, the output of the model may include a predicted visual acuity value after LASIK, a predicted visual acuity value after LASEK, and a predicted visual acuity value after a smile. As another example, when the input data includes standard refractive surgery and custom refractive surgery as surgical parameters, the output of the model may include a predicted visual acuity value after standard refractive surgery and a predicted visual acuity value after custom refractive surgery.

The visual acuity prediction model can predict visual acuity corresponding to a predetermined surgical parameter regardless of whether the input data includes the surgical parameter. For example, when the model is trained to predict visual acuity after standard refractive surgery and visual acuity after custom refractive surgery, even if the input data of the model does not include a surgical parameter, the model can predict visual acuity after standard refractive surgery and custom refractive surgery. It is possible to output the visual acuity prediction value.

The visual acuity prediction model may predict visual acuity corresponding to a plurality of different views. For example, the model may predict visual acuity corresponding to the first vision and the second vision after vision correction surgery. Examples of multiple different views include, but are not limited to, one day, one week, one month, six months, and one year after vision correction surgery.

The visual acuity prediction model may predict the visual acuity recovery speed of the subject based on the visual acuity prediction values corresponding to a plurality of different times. For example, the model may predict vision recovery speed based on the first vision and the second vision after refractive surgery.

The predictive vision image generating model can predict the subject's visual field after refractive surgery. The model may generate an image (hereinafter referred to as “expected visual acuity image”) visualizing the quality of visual acuity of the subject after refractive surgery. As the model outputs the expected visual acuity image, it is possible to more easily explain vision correction to the subject. As the model visualizes and outputs the visual field after vision correction surgery, the subject can more clearly understand the expected results after vision correction surgery, and through this, can receive help in choosing the vision correction surgery.

Input data of the predictive visual acuity image generation model may include surgical parameters. For example, the input data may include the type of surgery as a surgical parameter. The model may predict an expected visual acuity image corresponding to the input surgical parameters. For example, when the input data includes LASIK, LASEK, and SMILE as surgical parameters, the output of the model may include an expected visual acuity image after LASIK, an expected visual acuity image after LASIK, and an expected visual acuity image after smile. As another example, when the input data includes standard corrective surgery and custom optoscopic surgery as surgical parameters, the output of the model may include an expected visual acuity image after standard refractive surgery and an expected visual acuity image after custom refractive surgery. As another example, when the input data includes a plurality of optic zones, the output of the model may include a plurality of expected visual acuity images corresponding to the plurality of optic zones.

The predicted visual acuity image generating model may generate an expected visual acuity image corresponding to a predetermined surgical parameter regardless of whether the input data includes the surgical parameter. For example, when the model is trained to generate an expected visual image after standard corrective surgery and an expected visual image after custom corrective surgery, even if the input data of the model does not include a surgical parameter, the model can generate an expected visual image after standard corrective surgery. and an expected visual acuity image after custom refractive surgery may be output.

The expected visual acuity image may include information on at least one of sharpness, light smearing, contrast sensitivity, night vision, glare, double vision, and afterimages of the subject's expected visual acuity after refractive surgery.

10 is a diagram of an expected visual acuity image according to an exemplary embodiment. Referring to a of FIG. 10 , the predicted visual acuity images I1 and I2 may visualize and express information about sharpness of visual acuity. Referring to b of FIG. 10 , the predicted visual acuity images I3 , I4 , I5 , and I6 may visualize and express light blur information.

A plurality of expected visual acuity images may correspond to different surgical parameters. Referring to a of FIG. 10 , the first expected vision image I1 may correspond to the sharpness of vision after custom refractive surgery, and the second expected vision image I2 may correspond to the sharpness of vision after standard refractive surgery. Referring to b of FIG. 10 , the third to sixth expected vision images I3 to I6 may be predicted vision images corresponding to different optic zones. For example, the surgical range of the third expected visual acuity image I3 may be greater than the surgical ranges of the fourth to sixth expected visual acuity images I4 to I6.

An expected visual acuity image according to an embodiment may be generated through filtering using a filter. Here, filtering is a concept used in a general image processing field and may mean generating a filtered image by convolving an image and a filter. Examples of filters include average filter, weighted average filter, low-pass filter, Gaussian filter, median filter, bilateral filter ), blurring filter, high-pass filter, unsharp masking, high-boost filter, sharpening filter, etc. It is not limited to this.

11 is a diagram of an expected visual acuity image generating model M17 using a filter according to an exemplary embodiment. Referring to FIG. 11 , the expected visual acuity image generating model M17 may include a first sub-model M171 and a second sub-model M172.

The first submodel M171 may calculate and/or select a filter based on input data. For example, the input data may include predicted values of eye characteristic data after corrective surgery of the subject, and the first submodel M171 may calculate and/or select a filter based on the predicted values of eye characteristic data. As another example, the input data may include measurement values of the subject's eye characteristics data and surgical parameters, and the first submodel M171 may calculate and/or select a filter based on the measurement values of the eye characteristics data and the surgical parameters. .

The second sub-model M172 may generate an expected visual acuity image based on the filter calculated and/or selected by the first sub-model M171. For example, the second sub-model M172 may generate an expected visual acuity image by applying the filter to an original image. Here, the original image is an image that is a basis for generating the expected vision image, and may be input from the outside of the expected vision image generation model M17 or may be included in the model M17.

The corneal topography image prediction model can predict the subject's corneal topography image after refractive surgery. The model can predict one or more corneal topography images. The corneal topography image prediction model may generate a corneal topography image based on input data.

Input data of the corneal topography image prediction model may include surgical parameters. For example, the input data may include the type of surgery as a surgical parameter. The model can predict the corneal topography image corresponding to the input surgical parameters. For example, when the input data includes LASIK, LASEK, and SMILE as surgical parameters, the output of the model may include a corneal topography image after LASIK, a corneal topography image after LASEK, and a corneal topography image after SMILE. As another example, when the input data includes standard corrective surgery and custom optoscopic surgery as surgical parameters, the output of the model may include a corneal topographic image after standard corrective surgery and a corneal topographic image after custom corrective surgery.

The corneal topography image prediction model may generate a corneal topography image corresponding to a predetermined surgical parameter regardless of whether the input data includes the surgical parameter. For example, if the model is trained to generate a corneal topographic image after standard corrective surgery and a custom corrective surgery, even if the input data of the model does not include a surgical parameter, the model can generate the standard corrective corneal image and custom corneal topographic image after corrective optoscopic surgery. Corneal topography image can be output after vision correction surgery.

Input data of the corneal topography image prediction model may include the subject's corneal topography image measured before refractive surgery. 12 is a diagram of a corneal topography image according to an exemplary embodiment. Referring to FIG. 12 , the corneal topography image prediction model M18 may predict the subject's corneal topography image CI2 after the refractive surgery based on the subject's corneal topography image CI1 measured before the refractive surgery. 12 shows that only the corneal topography image CI1 is input to the model M18, but other input data may be input together.

The cause analysis model for the prediction result can analyze the cause of the prediction result generated by the refractive surgery-related model. The model may calculate dependence on input data of a model related to vision correction surgery. Here, the dependence may include how a specific variable of the input data has an effect on the prediction result.

The analysis model of the cause of the prediction result may output the cause of the prediction result. The model may output one or more prediction result production causes. In the following, for convenience of description, a case in which the cause of the prediction result is expressed in a numerical value such as a dependency coefficient will be described, but the cause of the prediction result is not limited to this, and there are restrictions on the expression method such as numerical values, images, texts, and combinations thereof. does not exist.

The causal analysis model for calculating the predicted result may output at least a part of the dependency coefficient. For example, the model may output all of the calculated dependency coefficients.

The cause analysis model for predicting results can output a dependency coefficient belonging to a certain range among the calculated dependency coefficients. For example, the model may output a greater dependence coefficient than a predetermined value among the calculated dependence coefficients. Alternatively, the model may output when the absolute value of the calculated dependency coefficient is greater than a predetermined value.

The causal analysis model for predicting results can output a certain number of dependence coefficients. For example, the model may output a predetermined number of dependence coefficients.

The cause analysis model for predicting result calculation is a cause analysis model for predicting the suitability of surgery, a cause analysis model for predicting the suitability of laser surgery, a cause analysis model for predicting corneal shape factor, a cause analysis model for predicting the need for custom refractive surgery, a cause analysis model for suggesting refractive surgery, a cause for suggesting surgical parameters An analysis model, a visual acuity prediction cause analysis model, a predicted visual acuity image generation cause analysis model, and a corneal topography image prediction cause analysis model may be included. For example, the vision prediction causal analysis model may calculate how the vision prediction model predicts vision based on the influence of variables included in the input data of the vision prediction model. Alternatively, the vision prediction cause analysis model may calculate the dependence of the vision prediction value calculated by the vision prediction model on input data of the vision prediction model.

The analysis model for the cause of the predicted result may include at least some of the models related to refractive surgery. For example, the visual acuity prediction causal analysis model may include a visual acuity prediction model.

FIG. 13 is a diagram of an analysis model M19 for predicting a result calculation cause including a model M192 related to vision correction surgery according to an embodiment. Referring to FIG. 27 , the prediction result calculation cause analysis model M19 may include an input data disturbance model M191, an optometry-related model M192, and a prediction result analysis model M193.

The input data disturbance model M191 may output disturbed input data based on the input data. The model M191 can output one or more perturbed input data. Here, the perturbation of the input data may mean changing the input data, such as changing values of at least some variables included in the input data. For example, if a variable is a number, perturbation can mean increasing or decreasing that number. Alternatively, when the variable is an image, perturbation may mean increasing or decreasing pixel values of at least some pixels of the image.

The vision correction related model M192 may output a first prediction result corresponding to the input data and a second prediction result corresponding to the perturbed input data based on the input data and the perturbed input data. For example, when the corrective surgery related model M192 included in the prediction result calculation cause analysis model M19 is a vision prediction model, the first prediction result and the second prediction result may be different vision prediction values. In addition, when there are a plurality of disturbed input data, the model may output a plurality of prediction results corresponding to the plurality of disturbed input data.

The prediction result analysis model M193 may output the cause of the prediction result based on the prediction result. For example, the model M193 may calculate the cause of the prediction result based on the difference between the first prediction result calculated from unperturbed input data and the second prediction result calculated from disturbed input data. Specifically, the dependence of the first prediction result on the first variable may be calculated based on a difference between the first prediction result and the second prediction result calculated from input data in which the first variable is disturbed.

14 is a diagram related to a cause of predicting result calculation according to an embodiment, and specifically, a diagram related to a visual acuity prediction cause. Referring to FIG. 14 , the cause of the prediction result may be expressed in numbers and images. Characters such as Cornea_Back_Rmin (V1), Astigmatism (V2), Mono (V3), Nearsightedness (V4), Op_flag (V5), and Pupil_Dia (V6) represented in FIG. 14 may correspond to variables included in input data. 5.95(N1), -0.5(N2), 0(N3), -1.37(N4), 2(N5), 3.1(N6), etc. expressed to correspond to the above characters correspond to the numerical values of the variables included in the input data. It can be.

The dependence of prediction results on variables can be visualized and expressed. For example, the dependence may be expressed by the length, color, direction, etc. of an arrow. Referring to FIG. 14 , the length of an arrow may correspond to an absolute value of a dependency coefficient. Also, the direction and color of the arrow may correspond to the sign of the dependence coefficient. 14 shows that the visual acuity predictive value (OV) is 1.18, Cornea_Back_Rmin (V1), Astigmatism (V2), Mono (V3), and Nearsightedness (V4) have a positive effect on the visual acuity predictive value, and Op_flag (V5) and Pupul_Dia (V6) are It can be interpreted as having a negative effect on visual acuity predictive value.

An example of a cause analysis model for predicting result production may include LIME (Local Interpretable Model-agnostic Explanations), but is not limited thereto.

Models related to vision correction can be combined with each other. The models may be combined by at least one of serial connection and parallel connection.

The fact that the corrective surgery related models are serially connected may include calculating an output of at least one other corrective surgery related model based on an output of at least one corrective surgery related model.

That the refractive surgery-related models are connected in parallel may include that the output of the refractive surgery-related model does not affect the output of other refractive surgery-related models.

Hereinafter, an embodiment of a combination of models related to vision correction will be described.

15 is a diagram related to serial connection of models related to vision correction surgery according to an embodiment. Referring to FIG. 15 , the first corrective surgery related model (Ma) and the second corrective surgery related model (Mb) may be combined through serial connection. The first corrective surgery related model (Ma) may calculate a first predicted result based on the first input data and the second predicted result output by the second corrective surgery related model (Mb) receiving the second input data. there is.

16 to 19 are diagrams for output calculation of a model related to vision correction surgery based on an output of a corneal shape factor prediction model according to an embodiment.

Referring to FIG. 16, the custom refractive surgery necessity prediction model M13 is based on the corneal shape factor and the first input data output by the corneal shape factor prediction model M12 receiving the second input data, and the custom refractive surgery of the subject. needs can be predicted.

Referring to FIG. 17 , the corrective surgery proposed model M14 performs the corrective surgery corresponding to the subject based on the corneal shape factor and the first input data output by the corneal shape factor prediction model M12 receiving the second input data. can suggest

Referring to FIG. 18 , the predicted visual acuity image generation model M17 predicts the subject's vision after corrective surgery based on the corneal shape factor outputted by the corneal shape factor prediction model M12 after receiving the second input data and the first input data. Sight images can be created.

Referring to FIG. 19 , the corneal topography image prediction model M18 is based on the corneal shape factor and the first input data output by the corneal shape factor prediction model M12 receiving the second input data, and the cornea after the subject's vision correction surgery. Terrain images can be predicted.

20 is a view related to calculating an output of a model related to refractive surgery based on an output of a custom refractive surgery necessity prediction model according to an embodiment. Referring to FIG. 20 , the corrective surgery proposal model M14 is based on the need for custom refractive surgery and the first input data, which the custom refractive surgery necessity prediction model M13 receives and outputs the second input data, and the visual acuity corresponding to the subject. Corrective surgery may be suggested.

21 and 22 are diagrams for calculating an output of a model related to vision correction surgery based on an output of a proposed model for vision correction surgery according to an embodiment.

Referring to FIG. 21 , the corneal shape factor prediction model M12 predicts the subject's corneal shape after the refractive surgery based on the refractive surgery and the first input data output by the refractive surgery proposal model M14 receiving second input data. factors can be predicted.

Referring to FIG. 22 , the vision prediction model M16 calculates a visual acuity prediction value after the vision correction surgery of the subject based on the vision correction surgery and the first input data output by the vision correction proposal model M14 receiving second input data. can do.

The predicted visual acuity image generating model may generate an expected visual acuity image after the refractive surgery of the subject based on the refractive surgery and the first input data that the refractive surgery proposal model receives and outputs the second input data.

The corneal topography image prediction model may predict the corrective surgery olfactory corneal topography image of the subject based on the corrective surgery and the first input data output by the corrective surgery proposed model after receiving the second input data.

23 is a view related to calculating an output of a model related to vision correction surgery based on an output of a surgical parameter proposal model according to an embodiment.

Referring to FIG. 23 , the visual acuity prediction model M16 calculates a visual acuity prediction value after the subject's refractive surgery based on the surgical parameter and the first input data output by the surgical parameter proposal model M15 receiving the second input data. can

The corneal shape factor prediction model can predict the corneal shape factor after refractive surgery of the subject based on the surgical parameter and the first input data that the surgical parameter proposal model receives and outputs the second input data.

The expected visual acuity image generation model may generate an expected visual acuity image after the subject's refractive surgery based on the surgical parameter and the first input data that the surgical parameter proposal model receives and outputs the second input data.

The corneal topography image prediction model can predict the corneal topography image after corrective surgery on the subject based on the surgical parameters and the first input data that the surgical parameter proposal model receives the second input data and outputs.

24 to 25 are diagrams for calculating an output of an optometry-related model based on an output of a vision prediction model according to an embodiment.

Referring to FIG. 24 , the corrective surgery proposal model M14 may suggest a corrective surgery corresponding to the subject based on the predicted visual acuity value and the first input data output by the second input data received by the vision prediction model M16. there is.

Referring to FIG. 25 , the predicted visual acuity image generation model M17 receives the second input data from the visual acuity prediction model M16 and outputs the predicted visual acuity image based on the first input data and the predicted visual acuity image after the subject's refractive surgery. can create

26 to 27 are diagrams related to output calculation of an optometry-related model based on an output of a corneal topography image prediction model according to an embodiment.

Referring to FIG. 26, the custom refractive surgery necessity prediction model M13 is based on the first input data and the corneal topography image output by the corneal topography image prediction model M18 receiving the second input data, and the custom refractive surgery of the subject. needs can be predicted.

Referring to FIG. 27 , the corrective surgery proposed model M14 performs the corrective surgery corresponding to the subject based on the first input data and the corneal topographic image output by the corneal topographic image prediction model M18 receiving the second input data. can suggest

The corneal shape factor prediction model may predict the corneal shape factor after corrective surgery of the subject based on the corneal topography image and the first input data output by the corneal topography image prediction model receiving the second input data.

The predicted visual acuity image generation model may generate an expected visual acuity image after corrective surgery of the subject based on the corneal topographic image and the first input data output by the corneal topographic image prediction model receiving the second input data.

Three or more refractive surgery-related models may be combined in series and/or parallel with each other. 28 is a diagram for explaining a combination of three or more vision correction-related models according to an embodiment. Referring to FIG. 28 , the first corrective surgery-related model (Ma) is the second prediction result obtained by the second corrective surgery-related model (Mb) receiving the second input data, and the third corrective surgery-related model (Mc). A third prediction result obtained by receiving and outputting the third input data and a first prediction result may be calculated by receiving the first input data. Here, the first corrective surgery related model (Ma) can be seen as serially connected to the second corrective surgery related model (Mb) and the third corrective surgery related model (Mc). In addition, the second corrective surgery related model Mb and the third corrective surgery related model Mc can be regarded as connected in parallel.

29 to 30 are diagrams for output calculation of an optometry-related model based on outputs of a corneal shape factor prediction model and a visual acuity prediction model according to an embodiment.

Referring to FIG. 29 , in the corrective surgery proposed model M14, the corneal shape factor prediction model M12 receives and outputs second input data, and the vision prediction model M16 receives third input data. Based on the output predicted visual acuity value and the first input data, a vision correction operation corresponding to the subject may be proposed.

Referring to FIG. 30 , the predicted visual acuity image generation model M17 inputs the corneal shape factor inputted by the corneal shape factor prediction model M12 and outputs the second input data, and the visual acuity prediction model M16 inputs the third input data. Based on the received and output visual acuity prediction value and the first input data, an expected visual acuity image after refractive surgery of the subject may be generated.

Orthopedic related models can be merged with each other. A plurality of models related to vision correction may be merged into one model to perform at least some of the functions of the plurality of individual models.

31 is a diagram for explaining merging of vision correction related models according to an embodiment. Referring to FIG. 31 , a first corrective surgery related model and a second corrective surgery related model may be merged to form one model Mab. The one model Mab may calculate a prediction result based on input data. Here, the prediction result may include information corresponding to at least one of a first prediction result corresponding to the output of the first corrective surgery related model and a second prediction result corresponding to the output of the second corrective surgery related model. For example, the prediction result may include at least one of a first prediction result and a second prediction result. Alternatively, the prediction result may include information on at least one of the first prediction result and the second prediction result.

Although the case of merging two models has been described in FIG. 31, it is possible to merge three or more models without being limited thereto.

32 to 34 are diagrams for implementation examples of merging models related to vision correction according to an embodiment.

Referring to FIG. 32 , a corneal shape factor prediction model and a custom corrective surgery necessity prediction model may be merged. The merged model M25 may calculate a prediction result based on input data. The prediction result may include information corresponding to at least one of a corneal shape factor and the need for custom refractive surgery. For example, the prediction result may include at least one of a corneal shape factor and the need for custom refractive surgery.

Referring to FIG. 33 , a custom corrective surgery necessity prediction model and a corrective surgery proposal model may be merged. The merged model M27 may calculate a prediction result based on input data. The prediction result may include information corresponding to at least one of a need for custom vision correction surgery and vision correction surgery. For example, the prediction result may include at least one of a need for custom vision correction surgery and vision correction surgery.

Referring to FIG. 34 , a laser surgery availability prediction model, a custom corrective surgery necessity prediction model, and a corrective surgery proposal model may be merged. The merged model M38 may calculate a prediction result based on input data. The prediction result may include information corresponding to at least one of laser surgery availability, necessity of custom vision correction surgery, and vision correction surgery. For example, the prediction result may include at least one of laser surgery availability, necessity of custom vision correction surgery, and vision correction surgery.

Hereinafter, an embodiment of a method for recommending vision correction surgery will be described.

The refractive surgery recommendation method may be implemented using one or more refractive surgery related models. When the method is implemented with a plurality of refractive surgery related models, whether or not at least one refractive surgery related model is executed may depend on a prediction result of at least one other refractive surgery related model. For example, whether the second refractive surgery-related model is executed may depend on the prediction result of the first refractive surgery-related model.

Each step of the refractive surgery recommendation method described below may be performed by a prediction device.

35 is a diagram related to a first embodiment of a vision correction surgery recommendation method according to an embodiment.

Referring to FIG. 35 , the method for recommending refractive surgery according to an embodiment includes acquiring examination data of a subject (S1100), predicting whether or not the subject is suitable for refractive surgery (S1200), and refractive surgery using a laser of the subject is possible. It may include a step of predicting whether or not (S1300), a step of calculating a predicted value of the subject's corneal shape factor (S1400), and a step of suggesting a corresponding vision correction surgery to the subject (S1500).

Obtaining examination data of the examinee ( S1100 ) may include acquiring examination data including medical examination data and measurement values of eyeball characteristic data by a computing device.

Predicting whether or not the subject is suitable for corrective surgery ( S1200 ) may include predicting whether or not the subject is suitable for corrective surgery by inputting the first group data obtained from the checkup data of the subject to the first prediction model. The first prediction model may be a surgical fit prediction model. The surgical suitability prediction model may predict the corrective surgery suitability of the subject based on the first group data.

The step of predicting whether the subject can undergo laser vision correction surgery (S1300) includes predicting whether the subject can undergo laser vision correction surgery by inputting the second group data obtained from the subject's examination data into a second prediction model. can do. Whether or not the step S1300 is performed may depend on whether or not corrective surgery is suitable for the subject. For example, the step (S1300) may be performed when vision correction is suitable for the subject. The second prediction model may be a laser surgery availability prediction model. The laser surgery availability prediction model may predict the subject's laser surgery availability based on the second group data.

In the step of calculating the predictive value of the subject's corneal shape factor (S1400), the third group data obtained from the examination data of the subject is input into the third predictive model, and the predicted value of the subject's corneal shape factor after standard refractive surgery and the corneal shape after custom refractive surgery It may include calculating factor predictions. Whether or not the step S1400 is performed may depend on whether the subject can undergo vision correction surgery using a laser. For example, the step (S1400) may be performed when the subject can undergo vision correction using a laser. The third prediction model may be a corneal shape factor prediction model. The third prediction model may predict the corneal shape factor based on the third group data. In this case, the need for custom refractive surgery may be determined based on the corneal shape factor.

Suggesting the corresponding vision correction surgery to the subject (S1500) may include proposing the corresponding vision correction surgery to the subject by inputting the fourth group data obtained from the examination data of the subject to the fourth prediction model. Whether or not the step (S1500) is performed may depend on whether the subject can undergo vision correction surgery using a laser. For example, the step (S1500) may be performed when the subject can undergo vision correction using a laser. The fourth predictive model may be learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients. The fourth predictive model may be an optometry proposed model. The fourth predictive model may suggest a vision correction operation corresponding to the subject based on the fourth group data.

Referring to FIG. 35 , the refractive surgery may not consider the predicted value of corneal shape factor and/or the need for custom refractive surgery. For example, the vision correction surgery may include LASIK, LASEK, and SMILE. In addition, even if laser vision correction is possible, lens implantation is not impossible, so the vision correction may include lens implantation, which is the same in other embodiments and embodiments of the present specification. In this case, refractive surgery considering the predicted value of the corneal shape factor and/or the need for custom refractive surgery may be determined by a doctor and/or a counselor. For example, a doctor and/or counselor may select a custom vision such as standard LASIK, standard LASIK, standard smile, custom LASIK, custom LASIK, or custom smile based on the corneal shape factor predicted value output by the refractive surgery recommendation method of FIG. 35 and refractive surgery. Vision correction surgery can be determined considering the need for correction surgery.

36 is a diagram related to a second embodiment of a vision correction surgery recommendation method according to an embodiment.

Referring to FIG. 36 , the method for recommending refractive surgery according to an embodiment includes acquiring examination data of the subject (S2100), predicting whether or not the subject is suitable for refractive surgery (S2200), and refractive surgery using a laser of the subject is possible. Predicting whether or not (S2300), calculating a predicted value of the subject's corneal shape factor (S2400), and suggesting a corresponding vision correction surgery to the subject (S2500) may be included.

Since the vision correction recommendation method of FIG. 36 is similar to that of FIG. 35, differences from FIG. 35 will be mainly described.

Referring to FIG. 36 , in the step of proposing a corrective surgery corresponding to the subject (S2500), the predicted value of the corneal shape factor after standard refractive surgery and the predicted value of the corneal shape factor after custom refractive surgery calculated in the step of calculating the predicted corneal shape factor of the subject (S2500). Based on this, vision correction surgery can be suggested. The vision correction surgery may be one in which the need for custom vision correction surgery is considered. For example, the vision correction surgery may include standard LASIK, standard LASIK, standard smile, custom LASIK, custom LASIK, and custom smile. In addition, the vision correction surgery may include lens implantation.

37 is a diagram related to a third embodiment of a method for recommending vision correction surgery according to an embodiment.

Referring to FIG. 37 , the method for recommending refractive surgery according to an embodiment includes obtaining examination data of the subject (S3100), predicting whether or not the subject is suitable for refractive surgery (S3200), and refractive surgery using a laser of the subject is possible. It may include a step of predicting whether or not (S3300), a step of predicting whether or not the subject needs custom refractive surgery (S3400), and a step of suggesting a corresponding refractive surgery to the subject (S3500).

Since the vision correction recommendation method of FIG. 37 is similar to that of FIG. 35, differences from FIG. 35 will be mainly described.

Referring to FIG. 37 , in the step of predicting whether the subject needs custom refractive surgery (S3400), third group data obtained from the subject's examination data is input into a third prediction model to predict whether the subject needs custom refractive surgery. may include Whether or not the step S3400 is performed may depend on whether the subject can undergo vision correction surgery using a laser. For example, the step (S3400) may be performed when the subject can undergo vision correction using a laser. The third prediction model may be a custom corrective surgery necessity prediction model. The third predictive model may predict the need for custom refractive surgery based on the third group data. In the above step, it is possible to predict the need for custom refractive surgery based on the predicted corneal shape factor values after standard refractive surgery and the predicted corneal shape factor values after custom refractive surgery.

The step of predicting whether the subject needs custom refractive surgery (S3400) and the step of suggesting a corresponding refractive surgery to the subject (S3500) may be connected. For example, the output of the step of proposing the corrective surgery to the subject (S3500) may be calculated based on the output of the step of predicting whether the subject needs custom refractive surgery (S3400). Alternatively, the output of the step S3400 of predicting whether or not the subject needs custom vision correction surgery may be calculated based on the output of the step S3500 of suggesting the corresponding corrective surgery to the subject.

For example, in the step of proposing the subject to have a corresponding vision correction surgery (S3500), the subject may propose the vision correction surgery based on the need for the custom vision correction surgery calculated in the step of predicting whether or not the subject needs the custom vision correction surgery (S3400). For example, the vision correction surgery may include standard LASIK, standard LASIK, standard smile, custom LASIK, custom LASIK, and custom smile. In addition, the vision correction surgery may include lens implantation.

As another example, in the step of predicting whether the subject needs custom refractive surgery (S3400), the second refractive surgery may be output based on the first refractive surgery calculated in the proposing the refractive surgery corresponding to the subject (S3500). . Here, the first vision correction surgery may not consider the need for custom vision correction surgery, and the second vision correction surgery may consider the need for custom vision correction surgery.

As another example, in the step of predicting whether the subject needs custom refractive surgery (S3400), whether or not to perform the refractive surgery may be determined based on the type of the first refractive surgery calculated in the step of proposing the corresponding refractive surgery to the subject (S3500). . For example, when the first corrective surgery is the first type, the custom corrective surgery necessity prediction model may not be executed, and when the first corrective surgery is the second type, the custom corrective surgery necessity prediction model may be executed.

The first type and the second type may be classified according to whether the cornea is ablated using a laser. For example, the first type may be non-laser vision correction surgery such as lens implantation, and the second type may be laser vision correction surgery such as LASIK, LASEK, and SMILE.

The first type and the second type may be classified according to whether custom surgery is possible. For example, the first type may be vision correction surgery in which custom surgery is not possible, and the second type may be vision correction surgery in which custom surgery is possible. Here, a pre-determined criterion may exist as to whether or not the custom surgery is possible vision correction. However, these standards may vary depending on the development of technology, hospital, surgical equipment, and the situation and judgment of the doctor. For example, if custom smile surgery is not possible, the first type includes smile and lens implantation, and the second type includes LASIK and LASEK. However, if custom smile surgery is possible, the first type includes lens implantation and Type 2 may include LASIK, LASEK, and SMILE.

38 is a diagram related to a fourth embodiment of a vision correction surgery recommendation method according to an embodiment.

Referring to FIG. 38 , the method for recommending refractive surgery according to an embodiment includes obtaining examination data of the subject (S4100), predicting whether or not the subject is suitable for refractive surgery (S4200), and refractive surgery using a laser of the subject is possible. It may include a step of predicting whether or not (S4300) and a step of suggesting a corresponding vision correction surgery to the subject (S4400).

The step of acquiring the examination data of the subject of FIG. 38 (S4100), the step of predicting whether the subject is suitable for vision correction surgery (S4200), and the step of predicting whether the subject can undergo vision correction surgery using a laser (S4300) are the same as those of FIG. 35 Therefore, the description thereof will be omitted.

Suggesting the corresponding vision correction surgery to the subject (S4400) may include proposing the corresponding vision correction surgery to the subject by inputting the third group data obtained from the examination data of the subject to the third prediction model.

Whether or not the step S4400 is performed may depend on whether the subject can undergo vision correction surgery using a laser. For example, the step (S4400) may be performed when the subject can undergo vision correction using a laser. In the step (S4400), refractive surgery may be suggested based on the predictive value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery.

The third predictive model may be a model in which a custom corrective surgery necessity prediction model and an corrective surgery proposal model are merged. The third predictive model may be learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients. The third predictive model may output information on at least one of a need for custom refractive surgery and refractive surgery based on the third group data. For example, the output of the third predictive model may include at least one of a need for custom refractive surgery and refractive surgery.

39 is a diagram related to a fifth embodiment of a vision correction surgery recommendation method according to an embodiment.

Referring to FIG. 39 , the method for recommending refractive surgery according to an embodiment includes acquiring examination data of a subject (S5100), predicting whether or not the subject is suitable for refractive surgery (S5200), and suggesting a refractive surgery corresponding to the subject It may include a step (S5300).

Obtaining examination data of the subject of FIG. 39 ( S5100 ) and predicting whether or not the subject is suitable for vision correction surgery ( S5200 ) are the same as those of FIG. 35 , so description thereof will be omitted.

Suggesting the corresponding vision correction surgery to the subject (S5300) may include proposing the corresponding vision correction surgery to the subject by inputting the second group data obtained from the examination data of the subject to the second prediction model.

Whether or not the step S5300 is performed may depend on whether or not corrective surgery is suitable for the subject. For example, the step (S5300) may be performed when vision correction is suitable for the subject. In the step (S5300), refractive surgery may be suggested based on the predictive value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery.

The second predictive model may be a model in which a laser surgery availability prediction model, a custom corrective surgery necessity prediction model, and a corrective surgery proposal model are merged. The second predictive model may be learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients. The second prediction model may output information on at least one of laser surgery availability, necessity of custom vision correction surgery, and vision correction surgery based on the second group data. For example, the output of the second predictive model may include at least one of laser surgery availability, necessity of custom vision correction surgery, and vision correction surgery.

40 is a diagram related to a sixth embodiment of a vision correction surgery recommendation method according to an embodiment.

Referring to FIG. 40 , the method for recommending refractive surgery according to an embodiment may include obtaining examination data of a subject (S6100) and suggesting a refractive surgery corresponding to the subject (S6200).

The step of acquiring examination data of the examinee in FIG. 40 ( S6100 ) is the same as that in FIG. 35 , so a description thereof will be omitted.

The step of suggesting the corresponding vision correction surgery to the subject (S6200) may include proposing the corresponding vision correction surgery to the subject by inputting the group data obtained from the examination data of the subject to the predictive model. In the step (S6200), refractive surgery may be suggested based on the predictive value of corneal shape factor after standard refractive surgery and the predicted value of corneal shape factor after custom refractive surgery.

The prediction model may be a model in which a surgery suitability prediction model, a laser surgery availability prediction model, a custom corrective surgery necessity prediction model, and a corrective surgery proposal model are merged. The predictive model may be learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients. The predictive model may output information about at least one of surgery suitability, laser surgery availability, necessity of custom vision correction surgery, and vision correction surgery based on input data. For example, the output of the prediction model may include at least one of surgery suitability, laser surgery availability, necessity of custom refractive surgery, and refractive surgery.

The above-described combination and/or merging of the refractive surgery recommendation method and the refractive surgery-related model is just an example, and the refractive surgery recommendation method may be implemented in various ways or the refractive surgery-related models may be combined and/or merged.

Hereinafter, an embodiment of a method for providing vision correction visualization information will be described.

The method for providing vision correction visualization information may be implemented using one or more vision correction related models. When the method is implemented with a plurality of refractive surgery related models, whether or not at least one refractive surgery related model is executed may depend on a prediction result of at least one other refractive surgery related model. For example, whether the second refractive surgery-related model is executed may depend on the prediction result of the first refractive surgery-related model.

The method for providing vision correction visualization information may include a method for providing an expected visual acuity image, a method for providing a corneal topography image, and a method for providing a cause for predicting results.

A method for providing an expected vision image may be implemented through a model for generating an expected vision image. The corneal topography image providing method may be implemented through a corneal topography image prediction model. The method for providing the cause of the prediction result may be implemented through an analysis model of the cause of the prediction result.

Each step of the vision correction visualization information providing method described below may be performed by a prediction device.

41 is a diagram related to a first embodiment of a method for providing vision correction visualization information according to an embodiment.

Referring to FIG. 41 , a method for providing visualization information for refractive surgery according to an embodiment includes obtaining examination data of a subject (S7100), calculating a predicted value of eyeball characteristic data after refractive surgery of the subject (S7200), and an expected visual acuity image. It may include a step of generating (S7300). Also, although not shown, the method of providing vision correction visualization information according to an exemplary embodiment may further include outputting an expected visual acuity image.

Obtaining examination data of the examinee ( S7100 ) may include acquiring examination data including medical examination data and measurement values of eyeball characteristic data by a computing device.

In the step of calculating the predictive value of eyeball characteristic data after refractive surgery of the subject (S7200), the first group data obtained from the examination data of the subject is input into the first predictive model, and at least among the predicted value of visual acuity and the predicted corneal shape factor after the refractive surgery of the subject is obtained. It may include calculating an eyeball characteristic data prediction value including one.

The first predictive model is based on at least one of a pre-surgery eye characteristic data measurement value of a plurality of recipients who have undergone refractive surgery, a vision correction surgery parameter performed on the plurality of recipients, and a post-surgery eye characteristic data measurement value of the plurality of recipients can be learned based on The first prediction model may include at least one of a visual acuity prediction model and a corneal shape factor prediction model. Alternatively, the first prediction model may be a model in which a visual acuity prediction model and a corneal shape factor prediction model are merged. The first predictive model may calculate predicted values of eyeball characteristic data after the subject's vision correction surgery based on the first group data.

Generating the predicted visual acuity image ( S7300 ) may include generating the predicted visual acuity image based on the predicted value of the eyeball characteristic data.

42 is a diagram related to a second embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.

Referring to FIG. 42 , the method for providing vision correction visualization information according to an embodiment further includes calculating and/or selecting a filter based on predicted values of eyeball characteristic data (S7600) and applying the filter to an original image (S7700). can include

Calculating and/or selecting a filter based on the predicted value of eyeball characteristic data (S7600) may be performed by the first submodel M171 of FIG. 11 .

Applying the filter to the original image ( S7700 ) may include applying the filter to the original image to generate the expected vision image. The step S7700 may be performed by the second submodel M172 of FIG. 11 .

43 is a diagram related to a third embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.

Referring to FIG. 43 , the method for providing visualization information for refractive surgery according to an embodiment may further include predicting a corneal topographic image of a subject after refractive surgery (S7400).

The step of predicting the subject's corneal topographic image after refractive surgery (S7400) may include predicting the subject's corneal topographic image after refractive surgery by inputting the second group data obtained from the subject's examination data into a second prediction model. there is.

The second predictive model is to be learned based on at least one of preoperative corneal topographic images of a plurality of recipients who have undergone refractive surgery, refractive surgery parameters performed on the plurality of recipients, and postoperative corneal topographic images of the plurality of recipients. can The second prediction model may be a corneal topography image prediction model. The second predictive model may predict a corneal topographic image of the subject after refractive surgery based on the second group data.

44 is a diagram related to a fourth embodiment of a method for providing visualization information for vision correction surgery according to an embodiment.

Referring to FIG. 44 , the method for providing vision correction visualization information according to an embodiment may further include calculating the dependence of the predicted value of eyeball characteristic data on the first group data (S7500). Also, although not shown, the method for providing vision correction visualization information according to an embodiment may further include outputting a dependence coefficient.

The step of outputting the dependence coefficient may include outputting a dependence coefficient greater than a predetermined value among the dependence coefficients or outputting a predetermined number of dependence coefficients.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computing devices and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - Includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media and ROM, RAM, flash memory, etc. Examples of program instructions include high-level language codes that can be executed by a computing device using an interpreter or the like as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

In the above, the configuration and characteristics of the present invention have been described based on the examples, but the present invention is not limited thereto, and various changes or modifications can be made within the spirit and scope of the present invention to those skilled in the art. It is obvious, and therefore such changes or modifications are intended to fall within the scope of the appended claims.

10: Vision correction assist system
100: learning device
300: prediction device
500: server device
700: client device
1000: control unit
5000: memory unit
9000: Ministry of Communications

Claims (12)

In the vision correction recommendation method using artificial intelligence performed by a computing device,
obtaining examination data of the examinee, wherein the examination data includes medical examination data and eyeball characteristic data measurement values;
The first group data obtained from the examination data of the subject is input into the first predictive model to be used for determining whether the subject needs custom refractive surgery. A first indicating a predicted value of a corneal shape factor after standard refractive surgery of the subject. Calculating a predicted corneal shape factor value and a second corneal shape factor predicted value indicating a corneal shape factor predicted value after custom refractive surgery; and
inputting the second group data obtained from the examination data of the subject into a second predictive model and providing a corresponding vision correction procedure to the subject; Including the step of providing vision correction based on the predicted value of the second corneal shape factor - including;
The second prediction model is learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients,
The predicted value of the first corneal shape factor,
Including at least one of the index of height decentration (IHD) predicted value after the standard refractive surgery, the index of surface variance (ISV) predicted value after the standard refractive surgery, or the index of vertical asymmetry (IVA) predicted value after the standard refractive surgery,
The predicted value of the second corneal shape factor,
At least one of an index of height decentration (IHD) predicted value after the custom refractive surgery, an index of surface variance (ISV) predicted value after the custom refractive surgery, or an index of vertical asymmetry (IVA) predicted value after the custom refractive surgery,
The standard refractive surgery includes refractive surgery to correct lower-order aberrations,
The custom refractive surgery includes refractive surgery to correct the lower and higher-order aberrations,
How to recommend vision correction surgery.
According to claim 1,
In the proposing of the vision correction surgery, the vision correction surgery is suggested based on the predicted visual acuity value after the vision correction surgery of the subject.
How to recommend vision correction surgery.
According to claim 2,
The proposing of the vision correction surgery may include calculating a visual acuity prediction value after the subject's vision correction surgery corresponding to each of the plurality of vision correction procedures, and suggesting the vision correction surgery.
How to recommend vision correction surgery.
According to claim 2,
The step of proposing vision correction may include calculating a plurality of visual acuity prediction values corresponding to a plurality of different times and proposing vision correction.
How to recommend vision correction surgery.
According to claim 1,
Predicting whether or not the subject needs custom refractive surgery based on the predicted value of the first corneal shape factor and the predicted value of the second corneal shape factor; further comprising
How to recommend vision correction surgery.
According to claim 1,
The proposing of the vision correction procedure may include proposing vision correction based on the predicted value of the first corneal shape factor and the predicted value of the second corneal shape factor.
How to recommend vision correction surgery.
According to claim 1,
When the second prediction model is learned in consideration of the preference of the plurality of subjects for corrective surgery, the second group data includes the preference for corrective surgery of the subject
How to recommend vision correction surgery.
According to claim 1,
At least one of the first prediction model and the second prediction model includes a plurality of submodels and calculates a result value based on results of the plurality of submodels.
How to recommend vision correction surgery.
According to claim 1,
At least one of the first group data and the second group data includes at least a portion of the examinee's examination data as it is or a new type of data calculated from at least a portion of the examinee's examination data.
How to recommend vision correction surgery.
According to claim 1,
predicting whether the subject is suitable for corrective surgery by inputting the third group data obtained from the examination data of the subject into a third predictive model;
predicting whether or not the subject can undergo refractive surgery by inputting fourth group data obtained from examination data of the subject into a fourth prediction model when the subject is suitable for refractive surgery; and
calculating at least one of the predicted value of the first corneal shape factor and the predicted value of the second corneal shape factor when the subject can undergo vision correction surgery using a laser; and
Further comprising the step of proposing the vision correction corresponding to the subject if vision correction using a laser is possible for the subject,
How to recommend vision correction surgery.
According to claim 1,
The examination data further includes genetic information
How to recommend vision correction surgery.
In a computing device,
including at least one processor, wherein the at least one processor comprises:
Examination data of the examinee - the examination data including medical examination data and ocular characteristic data measurement values - is acquired,
The first group data obtained from the examination data of the subject is input into the first predictive model to be used for determining whether the subject needs custom refractive surgery. A first indicating a predicted value of a corneal shape factor after standard refractive surgery of the subject. Calculate a second corneal shape factor predicted value representing the corneal shape factor predicted value and the corneal shape factor predicted value after custom refractive surgery, and
The second group data obtained from the examination data of the subject is input into a second predictive model to provide the subject with a corresponding vision correction surgery, wherein providing the vision correction surgery is based on the predicted value of the first corneal shape factor and the first 2 Including providing vision correction based on the predicted value of corneal shape factor;
The second prediction model is learned based on at least one of examination data of a plurality of recipients who have undergone refractive surgery, refractive surgery corresponding to the plurality of recipients, and vision after vision correction of the plurality of recipients,
The predicted value of the first corneal shape factor,
Including at least one of the index of height decentration (IHD) predicted value after the standard refractive surgery, the index of surface variance (ISV) predicted value after the standard refractive surgery, or the index of vertical asymmetry (IVA) predicted value after the standard refractive surgery,
The predicted value of the second corneal shape factor,
At least one of an index of height decentration (IHD) predicted value after the custom refractive surgery, an index of surface variance (ISV) predicted value after the custom refractive surgery, or an index of vertical asymmetry (IVA) predicted value after the custom refractive surgery,
The standard refractive surgery includes refractive surgery to correct lower-order aberrations,
The custom refractive surgery includes refractive surgery for correcting the lower and higher-order aberrations,
computing device.

Images