Myopia regression prediction electronic device and operating method thereof

Abstract

The myopia regression prediction electronic device according to one embodiment of the present invention comprises: a memory; and a processor connected with the memory and configured to execute commands contained in the memory, the processor can collect first target data of a subject and second target data of the subject, the first target data is used as input data of a first machine learning model, and the second target data is used as input data of a second machine learning model. And extracting a first result value as output data of the first machine learning model, and judging the myopia regression possibility of the subject based on the first result value.

Description

TECHNICAL FIELD

[0001] The present application relates to a myopia regression prediction electronic device and an operating method thereof. BACKGROUND

[0002] Nowadays, vision correction surgeries such as excimer laser photorefractive keratectomy, excimer laser sub-Bowman keratomileusis, and full-femtosecond laser small incision lenticule extraction are widely used for myopia correction in people of all ages and genders due to their excellent efficacy. With the popularity of smartphones, tablets, and other IT devices, the number of people who undergo vision correction surgeries to correct poor vision is increasing every year. However, the long-term efficacy and side effects of these surgeries are not yet clear, leading many people to worry about long-term complications.

[0003] The most common long-term complication after vision correction surgery is myopia regression. Since myopia regression usually progresses slowly, it is difficult to diagnose complications if long-term observation is not performed. The exact cause of myopia regression after vision correction surgery is not yet known, and since each person's situation is different, it is considered unpredictable. Myopia regression is accompanied by huge human, social, and economic costs, including the need to replace glasses, wear contact lenses, and perform additional corrective surgeries after vision correction surgery.

[0004] If patients with a high likelihood of myopia regression after vision correction surgery can be identified, the human, social, and economic costs associated with complications can be reduced. Therefore, a technology that can determine the likelihood of myopia regression after vision correction surgery is needed. SUMMARY

[0005] TECHNICAL PROBLEM

[0006] According to the present application, the present application aims to provide an electronic device that predicts the likelihood of myopia regression after vision correction surgery by applying an artificial intelligence machine learning model constructed based on preoperative data and fundus photographs of patients, and an operating method thereof.

[0007] TECHNICAL SOLUTION

[0008] The myopia regression prediction electronic device of an embodiment of the present application comprises: a memory; and a processor connected with the memory and configured to execute commands contained in the memory, the processor can collect first target data of a subject and second target data of the subject, use the first target data as input data of a first machine learning model, extract a first result value as output data of the first machine learning model, use the first result value and the second target data as input data of a second machine learning model, extract a second result value as output data of the second machine learning model, and judge the myopia regression possibility of the subject based on the second result value.

[0009] According to an embodiment of the present application, the first target data can be information related to fundus photos of the subject, and the second target data can include information related to age, gender, refractive power before vision correction surgery (for example, myopia degree, astigmatism value, etc.), vision correction surgery type (excimer laser in situ keratomileusis, excimer laser subepithelial keratoplasty, full femtosecond laser small incision lenticule extraction, etc.), intraocular pressure (IOP) before vision correction surgery, central corneal thickness (CCT) before vision correction surgery, anterior chamber depth (ACD) before vision correction surgery, and expected resection amount during vision correction surgery.

[0010] According to an embodiment of the present application, the processor can collect first training data, process a first training data set based on the first training data, construct the first neural network model based on the first training data set, and judge the performance of the first neural network model at a preset period, and the first training data can include information related to fundus photos of multiple subjects.

[0011] According to an embodiment of the present application, the processor can collect second training data, process a second training data set based on the second training data, construct the second neural network model based on the second training data set, and determine the performance of the second neural network model at a predetermined period. The second training data can include information about the age, gender, pre-operative refractive power (e.g., myopia degree, astigmatism value, etc.), type of vision correction surgery (e.g., PRK, LASIK, femto-LASIK, etc.), pre-operative IOP (Intraocular Pressure), pre-operative CCT (Central Corneal Thickness), pre-operative ACD (Anterior Chamber Depth), and expected resection amount during the vision correction surgery of each of the subjects.

[0012] According to an embodiment of the present application, the processor can provide the myopia regression possibility to an external electronic device.

[0013] The operation method of the myopia regression prediction electronic device according to an embodiment of the present application can include collecting first target data of a subject and second target data of the subject, using the first target data as input data of a first machine learning model, extracting a first result value as output data of the first machine learning model, using the first result value and the second target data as input data of a second machine learning model, extracting a second result value as output data of the second machine learning model, and determining a myopia regression possibility of the subject based on the second result value.

[0014] According to an embodiment of the present application, the first target data can be information about a fundus photograph of the subject, and the second target data can include information about the age, gender, pre-operative refractive power (e.g., myopia degree, astigmatism value, etc.), type of vision correction surgery (e.g., PRK, LASIK, femto-LASIK, etc.), pre-operative IOP (Intraocular Pressure), pre-operative CCT (Central Corneal Thickness), pre-operative ACD (Anterior Chamber Depth), and expected resection amount during the vision correction surgery of the subject.

[0015] According to an embodiment of the present application, the method can further include collecting first training data, processing a first training data set based on the first training data, constructing the first neural network model based on the first training data set, and determining the performance of the first neural network model at a predetermined period, and the first training data can include information about fundus photographs of a plurality of subjects.

[0016] According to an embodiment of the present application, the method can further include collecting second training data, processing a second training data set based on the second training data, constructing the second neural network model based on the second training data set, and determining the performance of the second neural network model at a predetermined period, and the second training data can include information about the age, gender, pre-operative refractive power (e.g., myopia degree, astigmatism value, etc.), type of vision correction surgery (e.g., PRK, LASIK, femto-LASIK, etc.), pre-operative IOP (Intraocular Pressure), pre-operative CCT (Central Corneal Thickness), pre-operative ACD (Anterior Chamber Depth), and expected resection amount during the vision correction surgery of each of the plurality of subjects.

[0017] According to an embodiment of the present application, the method can further include providing the myopia regression possibility to an external electronic device.

[0018] Effects of Invention

[0019] The myopia regression prediction electronic device and the operation method thereof according to the present application can predict myopia regression after vision correction surgery while minimizing the number of examinations for myopia regression observation, thereby saving costs and time and identifying patients at high risk of myopia regression to provide personalized treatment strategies. BRIEF DESCRIPTION OF DRAWINGS

[0020] Brief descriptions of the accompanying drawings are provided to facilitate a more complete understanding of the detailed description of the application.

[0021] Figure 1 FIG. 1 is a diagram illustrating a myopia regression prediction system according to an embodiment of the present application.

[0022] Figure 2 FIG. 2 is a block diagram illustrating an electronic device according to an embodiment of the present application.

[0023] Figure 3a FIG. 3 is a conceptual diagram illustrating an operation of an electronic device according to an embodiment of the present application.

[0024] Figure 3b A flowchart illustrating an operation of an electronic device according to an embodiment of the present disclosure.

[0025] Figure 3c A conceptual diagram illustrating an operation of an electronic device according to an embodiment of the present disclosure.

[0026] Figure 4a A flowchart illustrating an operation of an electronic device according to an embodiment of the present disclosure.

[0027] Figure 4b A flowchart illustrating an operation of an electronic device according to an embodiment of the present disclosure.

[0028] Figure 5 A flowchart illustrating an operation of an electronic device according to an embodiment of the present disclosure.

[0029] Figure 6 A block diagram illustrating a hardware configuration of a myopia regression prediction electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] Hereinafter, various embodiments of the present disclosure will be described with reference to the accompanying drawings. It is to be understood that the present disclosure is not limited to a specific embodiment, but includes various modifications, equivalents, and / or alternatives of the embodiments of the present disclosure. In the description of the drawings, like reference numerals can be used to refer to like elements.

[0031] In this document, "have," "has," "may have," "may include," "including," and "that which" or the like means that there is existence of the corresponding feature (for example, numerical value, function, operation, or part, etc. structural element), and does not exclude the presence of other features.

[0032] In this document, "A or B", "at least one of A or / and B", or "one or more of A or / and B" and the like include all possible combinations of the listed items. For example, "A or B", "at least one of A and B", or "at least one of A or B" can mean 1) including at least one A, 2) including at least one B, or (3) including both at least one A and at least one B.

[0033] The expressions "1st", "2nd", "first", or "second" and the like used herein can modify various structural elements regardless of their order and / or importance, and are used only to distinguish one structural element from other structural elements, and do not limit the structural elements. For example, a first structural element can be referred to as a second structural element, and similarly, a second structural element can also be referred to as a first structural element without departing from the scope of the rights described herein.

[0034] The expression "configured to" used herein can be used interchangeably with, for example, "suitable for", "having the capacity to", "designed to", "adapted to", "made to", or "capable of", according to circumstances. The term "configured to" does not necessarily mean "specifically designed to".

[0035] In this document, for example, "command", "instruction", "control information", "message", "information", "data", "packet", "data packet", "intent", and / or "signal" transmitted and received between the first electronic device(s) and the second electronic device(s) can contain or refer to a human-perceptible thought or a specific electrical representation (e.g., digital code / analog physical quantity), regardless of the expression. The listed exemplary expressions can be interpreted in various ways according to the context of their use, which will be apparent to those of ordinary skill in the art to which the inventions disclosed herein belong. In this document, "A is greater than B" not only simply means "A is greater than B", but also includes "A is equal to or greater than B".

[0036] The terms used herein are used only to describe specific embodiments and are not intended to limit the scope of other embodiments. Unless the context clearly dictates otherwise, a singular expression can include a plural expression. The terms used herein, including technical terms or scientific terms, can have the same meaning as commonly understood by those of ordinary skill in the art to which the technology described herein belongs. In the terms used herein, the terms defined in a general dictionary can be interpreted to have the same or similar meaning as in the relevant technical context, and should not be interpreted in an idealized or overly formal manner unless explicitly defined herein. According to circumstances, even the terms defined herein should not be interpreted as excluding the embodiments herein.

[0037] Figure 1 A diagram for a myopia regression prediction system according to an embodiment of the present invention.

[0038] Referring to Figure 1 The electronic device 150 according to an embodiment of the present invention can communicate with another external electronic device through a network.

[0039] For reference, the electronic device 150 can be implemented as a computer capable of connecting to a remote server or terminal through a network. For example, the computer can include a notebook computer, a desktop, a laptop, etc. equipped with a WEB browser. Also, the electronic device 150 can be implemented as a terminal capable of accessing a remote server or terminal through a network. For example, the electronic device 150 is a wireless communication device that ensures portability and mobility, and can include all types of handheld wireless communication devices, such as a navigation device, a personal communication system (PCS), a global system for mobile communication (GSM), a personal digital cellular network (PDC), a personal handphone system (PHS), a personal digital assistant (PDA), an international mobile telecommunication (IMT)-2000, a code division multiple access (CDMA)-2000, a W-code division multiple access (W-CDMA), a wireless broadband internet (Wibro) terminal, a smartphone, a smartpad, a tablet PC, etc.

[0040] The electronic device 150 of an embodiment of the present application can be implemented as one or more computer devices that provide a command, a code file, content, a service, etc. The electronic device 150 can be implemented in the form of a self-made web page or an application program (APP) made and operated by a person or a company.

[0041] According to an embodiment of the present application, the electronic device 150 can predict myopia regression by receiving input data from the outside. According to an embodiment of the present application, the input data can include the patient's age, gender, refractive power before the vision correction surgery (e.g., myopia degree, astigmatism value, etc.), vision correction surgery type (e.g., LASIK, PRK, Femto-LASIK, etc.), intraocular pressure (IOP) before the vision correction surgery, central corneal thickness (CCT) before the vision correction surgery, anterior chamber depth (ACD) before the vision correction surgery, expected ablation amount in the vision correction surgery, and fundus photograph, etc.

[0042] According to an embodiment of the present application, the electronic device 150 can identify the myopia regression possibility based on the input data.

[0043] Figure 2 For simplicity of illustration Figure 1 a block diagram showing detailed configurations of the server.

[0044] As Figure 2 illustrated, the electronic device 150 can include a bus 210, a display 220, a communication circuit 230, a database 240, a memory 250, an input / output interface 260, and a processor 270. In another embodiment, the electronic device 150 can omit at least one of the structural elements, or further include other structural elements.

[0045] For reference, Figure 2 the structural elements 210, 220, 230, 240, 250, 260, and 270 of the electronic device 150 illustrated are merely exemplary structural elements for describing the document creation support method of an embodiment of the present application. That is, it is obvious that the electronic device 150 of an embodiment of the present application can further include other structural elements in addition to the illustrated structural elements.

[0046] The bus 210 can electrically connect the structural elements 220 to 270 to each other. The bus 210 can include a circuit for communication (e.g., control message and / or data) between the structural elements 220 to 270.

[0047] The display 220 can display texts, images, videos, icons, or symbols constituting various contents. The display 220 can include a touch screen and receive touch, gesture, proximity, or hovering input using an electronic pen or a user's body part.

[0048] For example, the display 220 can include a liquid crystal display (LCD), a light emitting diode (LED) display, an organic LED display, or a microelectromechanical systems (MEMS) display, or an electronic paper display. The display 220 can be implemented in the electronic device 150 or can be implemented separately from the electronic device 150 but can be operatively connected to the electronic device 150.

[0049] The communication circuit 230 can establish a communication channel between the electronic device 150 and an external device. The communication circuit 230 can access a network 280 through wireless or wired communication, thereby communicating with the external device.

[0050] The database 240 can be implemented on the memory 250 or on a separate storage medium. The database 240 can store the contents, details, etc. of all data transmitted and received from the external device. The data stored in the database 240 can be periodically updated according to a predetermined period.

[0051] According to an embodiment of the present application, the database 240 can store various information received from the external device. For example, the database 240 can store the age, gender, pre-operative refractive power (e.g., myopic degree, astigmatism value, etc.), type of vision correction surgery (e.g., epi-LASIK, LASIK, femto-LASIK, etc.), pre-operative intraocular pressure (IOP), pre-operative central corneal thickness (CCT), pre-operative anterior chamber depth (ACD), expected ablation amount in the vision correction surgery, fundus photograph, etc. of the patient.

[0052] According to various embodiments, the data stored in the database 240 can also be stored in a distributed manner on a blockchain network to enhance the security of its use, since these data represent sensitive information of the subject. When the database 240 is stored in a distributed manner in the blockchain network, the history of the information contained in the database 240, such as transmission, modification, deletion, and addition, etc. can be more securely managed in the blockchain network.

[0053] The memory 250 can include a volatile memory and / or a non-volatile memory. The memory 250 can store commands or data related to at least one other element of the electronic device 150. For example, the memory 250 can store instructions that, when executed, cause the processor 270 to perform various operations described herein. As an example, the instructions can be included in a package file of an application program.

[0054] The input / output interface 260 can function to transmit commands or data input by the user or other external devices to other elements of the electronic device 150. The input / output interface 260 can be implemented by hardware or software, and can function as a concept including a user interface (UI) and a terminal for communication with other external devices.

[0055] The processor 270 can include at least one of a central processing unit (CPU), an application processor (AP), or a communication processor (CP). The processor 270 is electrically connected with the memory 250, the display 220, and the communication circuit 230 through the bus 210, and in operation, can perform an operation or data processing related to control and / or communication of other elements according to instructions, programs, or software stored in the memory 250. Accordingly, execution of the instructions, application programs, or software can be understood as an operation of the processor 270.

[0056] According to an embodiment of the present application, the processor 270 can perform data processing, etc., to perform a learning and prediction step of a model related to the following myopia regression prediction. According to an embodiment of the present application, the processor 270 can learn a model related to the following myopia regression prediction, and can generate a myopia regression prediction result using the model related to the myopia regression prediction.

[0057] The network 280 can include at least one of a telecommunications network, a computer network, the Internet, or a telephone network. A wireless communication protocol for accessing the network 280 can use at least one of, for example, Long-Term Evolution (LTE), LTE-A (LTE Advanced), Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), Universal Mobile Telecommunications System (UMTS), Wireless Broadband (WiBro), Global System for Mobile communications (GSM), or a 5G standard communication protocol. However, this is merely an example, and various wired and wireless communication techniques applicable to the technical field can be utilized according to the specific embodiments to which the present application is applied.

[0058] As described above, the electronic device 150 according to an embodiment of the present application can save costs and time by predicting myopia regression after vision correction surgery while minimizing the number of examinations for myopia regression observation, and can identify patients at high risk of myopia regression to provide personalized treatment strategies to the subjects.

[0059] Figure 3a A conceptual diagram illustrating a first machine learning model constructed by the electronic device 150 according to an embodiment of the present application.

[0060] The first machine learning model according to an embodiment of the present application can include a neural network for analyzing a photo taken of the fundus of a subject (hereinafter, fundus photo). According to an embodiment of the present application, the neural network can be composed of a set of interconnected node units. A plurality of nodes represent a plurality of neurons. According to an embodiment of the present application, the nodes constituting the neural network can be connected by one or more links. One or more nodes connected by a link can form a relationship between an input node and an output node.

[0061] In a relationship between an input node and an output node connected by a link, a value of data of the output node can be determined according to data input to the input node. According to an embodiment of the present application, a link connecting an input node and an output node can have a weight. The weight can be variable. Also, the weight of a neural network of an embodiment of the present application can be changed by a user or a predetermined algorithm to predict myopia regression. For example, when more than one input node is connected to one output node by a link, the output node can determine an output node value according to input values of input nodes connected to the corresponding output node and weights set for links corresponding to each input node.

[0062] According to an embodiment of the present application, performance of a neural network can be determined according to the number of nodes and links within the neural network, a connection relationship between the nodes and the links, and a weight assigned to each link. For example, when there are two neural networks having the same number of nodes and links but different weights of the links, the two neural networks can be recognized as different neural networks.

[0063] A neural network can be composed of a set of nodes, and a subset of nodes constituting the neural network can be referred to as a layer. A portion of nodes constituting the neural network can constitute a layer based on their distance from a specific input node. For example, a set of nodes having a distance of "n" from a specific input node can constitute an "n"-layer. The distance from a specific input node refers to the number of links required to reach the node from the specific input node. However, the definition of these layers is merely one embodiment and is not limited thereto.

[0064] According to an embodiment of the present application, the number of nodes included in an input layer of a neural network can be the same as the number of nodes included in an output layer. The number of nodes of an input layer of a neural network of another embodiment of the present application can be different from the number of nodes of an output layer.

[0065] A deep neural network (DNN) is a neural network that includes an input layer, an output layer, and at least one hidden layer. Deep neural networks can be utilized to capture latent structures of data. For example, deep neural networks can be utilized to capture latent structures of fundus photographs (e.g., presence of particular patterns, morphology of retinal nerves, thinness of the pigment layer, etc.). Deep neural networks can include convolutional neural networks, recurrent neural networks, generative adversarial networks, autoencoders, deep belief networks, and the like. These are merely examples and are not limited thereto.

[0066] Neural networks can learn in at least one of supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. Learning of a neural network can be a process of applying information (data) to the neural network to build a particular model for myopia regression prediction.

[0067] A neural network can learn from training data to minimize an output error. The neural network can repeatedly receive training data and calculate an error between an output of the neural network and target data based on the training data. To reduce the calculated error, the neural network can backpropagate the error from an output layer (or output nodes) of the neural network to an input layer (or input nodes) to update weights of each node of the neural network.

[0068] In the case of supervised learning, the neural network can use training data (labeled data) in which each training data point is labeled with a correct value. In the case of unsupervised learning, the neural network can not label each training data set with a correct value. For example, training data applied in supervised learning about data classification can be data labeled with each category. The neural network can receive the labeled data and compare its output with the label of the training data to calculate an error. For another example, for training data applied as input in unsupervised learning about data classification, the neural network can calculate an error by comparing it with the output of the neural network. The calculated error can be backpropagated in the neural network. Through backpropagation, the neural network can update the weight of the connection of each node included in each layer of the neural network. The amount of change in the updated connection weight of each node can be determined according to a learning rate. The calculation of the input data by the neural network and the backpropagation of the error can constitute one learning cycle. Different learning rates can be applied according to the number of iterations of the learning cycle of the neural network.

[0069] In neural network learning, training data is usually a subset of actual data (i.e., fundus photographs, etc. to be processed by the neural network to which learning is applied). Even if the error in the training data is reduced, the error in the actual data can increase. Overfitting is a phenomenon in which over-learning of training data as described above results in an increase in the error of the actual data. Overfitting can be a cause of increasing the error in the machine learning algorithm.

[0070] To prevent overfitting, the electronic device 150 of an embodiment of the present application employs methods such as increasing regularization of training data, dropout in which certain network nodes are deactivated at random during the learning process, a batch normalization layer, etc.

[0071] Figure 3b A flowchart showing the operation of the electronic device 150 of an embodiment of the present application is shown. In particular, Figure 3a The operation of the electronic device 150 of an embodiment of the present application is shown.

[0072] In step S301, the electronic device 150 can collect first training data. According to an embodiment of the present application, the electronic device 150 can receive the first training data from the outside. According to another embodiment of the present application, the electronic device 150 can collect training data by itself. The first training data for constructing the first machine learning model can be fundus photographs corresponding to a plurality of subjects.

[0073] In step S303, the electronic device 150 can process the first training data set based on the first training data. According to an embodiment of the present application, the electronic device 150 can generate the first training data set by dividing the first training data into data having a high possibility of myopia regression (abnormal data) and data having a low possibility of myopia regression (normal data).

[0074] In step S305, the electronic device 150 can construct a first machine learning model for predicting myopia regression using the fundus photograph of the subject based on the first training data set.

[0075] In step S307, the electronic device 150 can judge the performance of the first machine learning model. According to an embodiment of the present application, the electronic device 150 can calculate an error by comparing the data output using the first machine learning model with the first training data. The electronic device 150 can judge the performance of the first machine learning model by periodically calculating the error.

[0076] Although not illustrated, the electronic device 150 can update the first machine learning model in a direction to reduce the calculation error.

[0077] Figure 3c A conceptual diagram to illustrate the operation of the electronic device 150 according to an embodiment of the present application.

[0078] According to an embodiment of the present application, the electronic device 150 can extract a value (first result value) predicting myopia regression using the first machine learning model constructed for the fundus photograph (input data).

[0079] According to an embodiment of the present application, the electronic device 150 can also judge myopia regression of the subject using only the first result value.

[0080] Figure 4a A flowchart to illustrate the operation of the electronic device 150 according to an embodiment of the present application. In particular, Figure 4a An operation of the electronic device 150 constructing a second machine learning model according to an embodiment of the present application is illustrated. Although not illustrated, the second machine learning model can also be applied as is to the contents described above. Figure 3a

[0081] ​The electronic device 150 can collect second training data in step S401. According to an embodiment of the present application, the electronic device 150 can receive the second training data from the outside. According to another embodiment of the present application, the electronic device 150 can collect the second training data by itself. The second training data for constructing the second machine learning model can include information (hereinafter, referred to as "medical data") about the patient age, gender, pre-operation refractive power (e.g., myopia degree, astigmatism value, etc.), vision correction surgery type (e.g., PRK, LASIK, Femto-LASIK, etc.), pre-operation intraocular pressure (IOP), pre-operation central corneal thickness (CCT), pre-operation anterior chamber depth (ACD), and expected resection amount in the vision correction surgery of a plurality of subjects, and the first result value output as the first machine learning model.

[0082] The electronic device 150 can process a second training data set based on the second training data in step S403. According to an embodiment of the present application, the electronic device 150 can generate the second training data set by classifying the second training data into data having a high possibility of myopia regression (abnormal data) and data having a low possibility of myopia regression (normal data). For example, the electronic device 150 can generate the second training data set by reflecting the medical data on the first result value and classifying it into data having a high possibility of myopia regression (abnormal data) and data having a low possibility of myopia regression (normal data).

[0083] The electronic device 150 can construct a second machine learning model for predicting myopia regression using the fundus photograph of the subject and the medical data of the subject based on the second training data set in step S405. In particular, the electronic device 150 can construct the second machine learning model by integrating the medical data and the first result value set output using the first machine learning model to be able to predict myopia regression.

[0084] The electronic device 150 can judge the performance of the second machine learning model in step S407. According to an embodiment of the present application, the electronic device 150 can calculate an error by comparing the data output using the second machine learning model with the second training data. The electronic device 150 can judge the performance of the second machine learning model by periodically calculating the error.

[0085] Although not illustrated, the electronic device 150 can update the second machine learning model in the direction of reducing the calculation error.

[0086] Figure 4bA conceptual diagram to illustrate an operation of the electronic device 150 according to an embodiment of the present disclosure.

[0087] According to an embodiment of the present disclosure, the electronic device 150 can extract a value (a second result value) predicting myopia regression of the subject using a second machine learning model constructed based on the medical data (input data) and the first result value.

[0088] According to an embodiment of the present disclosure, the electronic device 150 can predict a myopia regression possibility of the subject based on the second result value.

[0089] Figure 5 A flowchart to illustrate an operation of the electronic device 150 according to an embodiment of the present disclosure.

[0090] In step S501, the electronic device 150 can collect input data from the outside. The input data can include medical data and a fundus photo. According to various embodiments of the present disclosure, the electronic device 150 can collect the medical data and the fundus photo by itself.

[0091] In step S503, the electronic device 150 can extract a first result value of a fundus photo of the subject using the first machine learning model.

[0092] In step S505, the electronic device 150 can extract a first result value corresponding to the subject and a second result value corresponding to the medical data of the subject using the second machine learning model.

[0093] In step S507, the electronic device 150 can determine a myopia regression possibility of the subject based on the extracted second result value.

[0094] Figure 6 A block diagram to illustrate a hardware configuration of a myopia regression prediction electronic device according to an embodiment of the present disclosure.

[0095] The computing system 1000 of an embodiment disclosed herein can include a micro control unit (MCU) 1010, a memory 1020, an input / output interface (I / F) 1030, and a communication interface (I / F) 1040.

[0096] The micro control unit (MCU) 1010 can be a processor that executes various programs stored in the memory 1020 for predicting myopia regression, processes various data through the programs, and performs the functions of the electronic device 150 shown in the above-described Figure 2

[0097] The memory 1020 can store various programs. Also, the memory 1020 can store various data received from a client.

[0098] ​A plurality of such memories 1020 can also be provided as necessary. The memory 1020 can be a volatile memory or a non-volatile memory. As the volatile memory of the memory 1020, a random access memory (RAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), or the like can be used. As the non-volatile memory of the memory 1020, a read only memory (ROM), a programmable read only memory (PROM), an electrically alterable read only memory (EAROM), an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a flash memory, or the like can be used. The memory 1020 listed is merely an example and is not limited thereto.

[0099] The input / output interface (I / F) 1030 can provide an interface through which data can be transmitted and received by connecting an input device (not shown) such as a keyboard, a mouse, or a touch screen, an output device such as a display (not shown), and the micro control unit (MCU) 1010.

[0100] The communication interface (I / F) 1040 is a structural element capable of transmitting and receiving various data with a server, and can be various devices supporting wired or wireless communication. For example, a program for managing various data or various data, etc. can be transmitted and received from an externally provided server through the communication interface (I / F) 1040.

[0101] As described above, the computer program of the embodiment disclosed herein can be implemented as a module recorded in the memory 1020 and processed by the micro control unit (MCU) 1010, thereby performing, for example Figure 2 the functions illustrated.

[0102] In the foregoing, although all the structural elements constituting the embodiments of the present application have been described as being integrated or operated in combination, the present application is not necessarily limited to these embodiments. That is, all the structural elements can be selectively integrated and operated more than once within the scope of the present application.

[0103] On the other hand, various embodiments described herein can be implemented by hardware, middleware, microcode, software, and / or combinations thereof. For example, various embodiments can be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof.

[0104] Also, for example, various embodiments can be embodied or encoded in a computer readable medium, including a computer readable storage medium encoded with instructions that, when executed in an electronic device, perform a method. Computer readable media includes computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented in a method or technology for storage and / or transmission of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disks (DVDs), other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices.

[0105] Such hardware, software, firmware, etc. can be implemented within the same device or within separate devices to support the various operations and functions described herein. Further, the structural elements, units, modules, and components described in the present disclosure as being "a" or "one" can be implemented together or separately as independent but interoperable logic devices. The description of different features, units, modules, etc. is intended to highlight different functional aspects and does not necessarily imply that they must be realized by separate hardware or software components. Rather, functionality associated with one or more modules or units can be performed by a single hardware or software component, or integrated into common or separate hardware or software components.

[0106] Although the operations are described in a particular, sequential order, it should be understood that other operations can be performed in between described operations, or operations described sequentially can be performed in parallel, or in a different order. In addition, it should be understood that described operations can be performed by different components, or in different components, than those described. In addition, it should be understood that described operations can be performed by the same component or in the same component, as other operations. In addition, it should be understood that different components can perform the same operations, or different components can perform the same operations.

[0107] The electronic device, the server, or the external device of the various embodiments described herein above can include, for example, at least one of a smartphone, a tablet computer, a mobile phone, a video phone, a desktop computer, a laptop computer, a personal digital assistant (PDA), a portable multimedia player (PMP), an MP3 player, a mobile medical device, a camera, or a wearable device.

[0108] According to various embodiments, the wearable device can include at least one of a watch type (e.g., a wrist watch, a ring, a bracelet, an ankle chain, a necklace, glasses, contact lenses, or a head-mounted device (HMD)), a fabric or clothing integrated type (e.g., an electronic clothing), a body-worn type (e.g., a skin pad or a tattoo), or a bio-implant type (e.g., an implantable circuit).

[0109] In certain embodiments, the electronic device or the external device can be a home appliance. The home appliance can include at least one of, for example, a television, a Digital Video Disk player, an audio device, a refrigerator, an air conditioner, a cleaner, an oven, a microwave oven, a washing machine, an air cleaner, a set-top box, a home automation control panel, a security control panel, a TV box, a game console, an electronic dictionary, an electronic key, a camcorder, or an electronic frame.

[0110] In another embodiment, the electronic device, the external device, the wearable device can include at least one of various medical devices (e.g., various portable medical measuring devices (a blood glucose meter, a heart rate meter, a blood pressure meter, or a body temperature meter, etc.), a magnetic resonance angiography (MRA), a magnetic resonance imaging (MRI), a computed tomography (CT), a camera, or an ultrasonic wave device, etc.), a navigation device, a Global Navigation Satellite System (GNSS), an event data recorder (EDR), a flight data recorder (FDR), a car infotainment device, a home robot, or an internet of things (e.g., a light bulb, various sensors, a gas or water meter, a sprinkler, a fire alarm, a thermostat, a street light, a fitness equipment, a hot water tank, a heater, a boiler, etc.).

[0111] The best mode contemplated has been set forth above with the drawings and specification. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation of the scope of the invention set forth in the claims. Accordingly, one of ordinary skill in the art will recognize that modifications can be made by applying common general knowledge without departing from the scope of the application encompassed by the claims. Accordingly, the true technical scope of the present invention should be determined by the technical spirit of the appended claims.

Claims

1. A myopia regression prediction electronic device, characterized in that: include: Memory; as well as a processor connected to the memory and configured to execute the commands contained in the memory, The processor collects the subject's first target data and the subject's second target data, uses the first target data as input data of a first machine learning model, extracts a first result value as output data of the first machine learning model, and determines the possibility of myopia regression of the subject based on the first result value.

2. The myopia regression prediction electronic device according to claim 1, characterized in that: The processor uses the first result value and the second target data as input data of a second machine learning model, extracts the second result value as output data of the second machine learning model, and determines the possibility of myopia regression of the subject based on the second result value.

3. The myopia regression prediction electronic device according to claim 1, characterized in that: The first target data is information related to the fundus photograph of the subject, and the second target data includes information related to the subject's age, gender, refractive power before vision correction surgery such as myopia degree, astigmatism value, type of vision correction surgery such as excimer laser in situ keratomileusis, excimer laser subepithelial keratomileusis, all-femtosecond laser small incision lenticule removal, intraocular pressure before vision correction surgery, central corneal thickness before vision correction surgery, anterior chamber depth before vision correction surgery, and expected resection amount during vision correction surgery.

4. The myopia regression prediction electronic device according to claim 1, characterized in that: The processor collects first training data, processes a first training data set based on the first training data, constructs the first neural network model based on the first training data set, and determines the performance of the first neural network model according to a preset period. The first training data includes information related to fundus photographs of a plurality of subjects.

5. The myopia regression prediction electronic device according to claim 1, characterized in that: The processor collects second training data, processes a second training data set based on the second training data, constructs a second neural network model based on the second training data set, and determines the performance of the second neural network model according to a preset period. The second training data includes information about the age, gender, refractive power before vision correction surgery such as myopia degree, astigmatism value, type of vision correction surgery such as excimer laser in situ keratomileusis, excimer laser subepithelial keratomileusis, all-femtosecond laser small incision lenticule removal, intraocular pressure before vision correction surgery, central corneal thickness before vision correction surgery, anterior chamber depth before vision correction surgery, and expected resection amount in vision correction surgery of each of the multiple subjects.

6. A method for operating an electronic device for myopia regression prediction, characterized in that: The steps include: collecting first target data of a subject and second target data of the subject; Using the first target data as input data of a first machine learning model, and extracting a first result value as output data of the first machine learning model; as well as The possibility of myopia regression of the subject is determined based on the first result value.

7. The method for operating the electronic device for myopia regression prediction according to claim 6, wherein: The following steps are also included: using the first result value and the second target data as input data of a second machine learning model, and extracting a second result value as output data of the second machine learning model; and The possibility of myopia regression of the subject is determined based on the second result value.

8. The method for operating the electronic device for myopia regression prediction according to claim 6, wherein: The first target data is information related to the fundus photograph of the subject, and the second target data includes information related to the subject's age, gender, refractive power before vision correction surgery such as myopia degree, astigmatism value, type of vision correction surgery such as excimer laser in situ keratomileusis, excimer laser subepithelial keratomileusis, all-femtosecond laser small incision lenticule removal, intraocular pressure before vision correction surgery, central corneal thickness before vision correction surgery, anterior chamber depth before vision correction surgery, and expected resection amount during vision correction surgery.

9. The method for operating the electronic device for myopia regression prediction according to claim 6, wherein: The following steps are also included: Collecting first training data; processing a first training data set based on the first training data; Building the first neural network model based on the first training data set; and Determine the performance of the first neural network model according to a preset period, The first training data includes information related to fundus photographs of a plurality of subjects.

10. The method for operating the electronic device for myopia regression prediction according to claim 6, wherein: The following steps are also included: collecting second training data; processing a second training data set based on the second training data; Building the second neural network model based on the second training data set; as well as Determine the performance of the second neural network model according to a preset period, The second training data includes information about the age, gender, refractive power before vision correction surgery such as myopia degree, astigmatism value, type of vision correction surgery such as excimer laser in situ keratomileusis, excimer laser subepithelial keratomileusis, all-femtosecond laser small incision lenticule removal, intraocular pressure before vision correction surgery, central corneal thickness before vision correction surgery, anterior chamber depth before vision correction surgery, and expected resection amount in vision correction surgery of each of the multiple subjects.

Images