Kidney disease diagnosis method and device

Abstract

The invention discloses a kidney disease diagnosis method and device. A control method of a diagnostic apparatus according to one embodiment includes: a step of acquiring a retinal image of a subject; based on the retina image, kidney disease diagnosis information about the object is obtained through a machine learning model, the machine learning model comprises a first model and a second model, the first model is a neural network model, and the second model can be a regression-based machine learning model.

Description

Technical Field

[0001] This invention relates to methods and devices for diagnosing kidney diseases. Background Technology

[0002] Retinal imaging can observe abnormalities in the retina, optic nerve, and macula, and the results can be relatively easily confirmed through imaging, making it a frequently used diagnostic tool in ophthalmology. On the other hand, due to the rapid development of artificial intelligence (AI) technology in recent years, the development of diagnostic AI is actively underway in the field of medical diagnosis, especially in image-based diagnosis. Global companies are also investing heavily in the development of AI to analyze various medical imaging data through collaborations with the medical community, including large-scale data collection. Some companies have even successfully developed AI diagnostic tools that output excellent diagnostic results.

[0003] Retinal imaging allows for non-invasive observation of blood vessels within the body, thus its application in diagnosis should be expanded beyond ophthalmology to include kidney diseases. Summary of the Invention

[0004] Technical issues

[0005] The technical objective of this application is to provide a method for diagnosing kidney diseases by utilizing retinal images and machine learning models to obtain information about kidney diseases with high accuracy.

[0006] The technical subject matter of this application is not limited to the above-mentioned subject matter. Subject matters not mentioned can be clearly understood by those skilled in the art from this specification and the accompanying drawings.

[0007] Solution to the problem

[0008] According to one aspect, a method for controlling a diagnostic device according to one embodiment includes: the step of acquiring a retinal image of a subject; and the step of acquiring diagnostic information about kidney disease of the subject using a machine learning model based on the retinal image. The machine learning model may include a first model and a second model, wherein the first model may be a neural network model and the second model may be a regression-based machine learning model.

[0009] The technical solutions are not limited to the above-mentioned solutions. Any technical solutions not mentioned can be clearly understood by those skilled in the art from this specification and the accompanying drawings.

[0010] Invention Effects

[0011] According to this application, retinal images and machine learning models can be used to obtain information about kidney disease with high accuracy.

[0012] The effects of this invention are not limited to those described above. Effects not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings. Attached Figure Description

[0013] Figure 1 A diagnostic system according to one embodiment is shown.

[0014] Figure 2 This is a block diagram illustrating a learning device according to one embodiment.

[0015] Figure 3 This is a block diagram illustrating a diagnostic apparatus according to one embodiment.

[0016] Figure 4 A diagnostic system according to one embodiment is shown.

[0017] Figure 5 This is a block diagram illustrating a client device according to an embodiment of the present invention.

[0018] Figure 6 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention.

[0019] Figure 7 This is a diagram illustrating the configuration of a learning unit according to an embodiment of the present invention.

[0020] Figure 8 This is a conceptual diagram illustrating an image dataset according to an embodiment of the present invention.

[0021] Figure 9 This is a block diagram illustrating the diagnostic model learning process according to an embodiment of the present invention.

[0022] Figure 10 This is a diagram illustrating the configuration of a diagnostic section according to an embodiment of the present invention.

[0023] Figure 11 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention.

[0024] Figure 12 This is a block diagram illustrating a diagnostic section according to an embodiment of the present invention.

[0025] Figure 13 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention.

[0026] Figure 14 This is a diagram illustrating a diagnostic system according to an embodiment of the present invention.

[0027] Figure 15 This is a diagram used to illustrate a serial diagnostic model according to one embodiment.

[0028] Figure 16 This is a diagram used to illustrate a serial diagnostic model according to another embodiment.

[0029] Figure 17 This is a diagram used to illustrate a serial diagnostic model according to yet another embodiment.

[0030] Figure 18 This is a diagram illustrating a diagnostic method using a diagnostic model according to one embodiment.

[0031] Figure 19 This is a diagram illustrating a method for diagnosing kidney disease according to one embodiment.

[0032] Figure 20 A diagnostic model for obtaining diagnostic information of kidney disease according to one embodiment is described.

[0033] Figure 21 The clinical characteristics of the subjects according to Example 1 are shown.

[0034] Figure 22a and Figure 22b This is a graph used to illustrate the incidence of kidney disease-related events based on kidney disease diagnostic information according to Example 1.

[0035] Figure 23 This is a graph used to illustrate the predictive performance of kidney disease-related events over 5 years based on kidney disease diagnostic information from Example 1.

[0036] Figure 24 This is a diagram illustrating a method for diagnosing kidney disease according to another embodiment.

[0037] Figure 25 This is a diagram illustrating a method for providing guidance information on the diagnosis of kidney disease according to one embodiment.

[0038] Figure 26 This is a diagram illustrating a method for providing guidance information using existing prescription information and kidney disease diagnostic information according to one embodiment.

[0039] Figure 27 This is a diagram illustrating a method for predicting the rate of progression of kidney disease using diagnostic information of kidney disease, according to one embodiment.

[0040] Figure 28 This is a diagram used to illustrate the clinical characteristics of the object according to Example 2.

[0041] Figure 29a and Figure 29b This is a graph used to illustrate the incidence of kidney disease-related events based on KDIGO classification and kidney disease diagnostic information according to Example 2. Detailed Implementation

[0042] The best way to implement the invention

[0043] A control method for a diagnostic device according to one embodiment includes: acquiring a retinal image of a subject; and acquiring diagnostic information about kidney disease of the subject using a machine learning model based on the retinal image, wherein the machine learning model may include a first model and a second model, the first model being a neural network model and the second model being a regression-based machine learning model.

[0044] Methods for implementing the invention

[0045] The above-described objects, features, and advantages of the present invention will become more apparent from the following detailed description in conjunction with the accompanying drawings. However, the present invention can have various modifications and embodiments; therefore, specific embodiments will be illustrated and described in detail below.

[0046] In the accompanying drawings, the thickness of layers and regions is exaggerated for clarity, and when an element or layer is referred to as "on" or "above" another element or layer, it includes cases where not only is it directly on the other element or layer, but other layers or other elements are interposed in between. Throughout the specification, the same reference numerals generally denote the same constituent elements. Furthermore, the same reference numerals are used to describe the same constituent elements that function within the same conceptual framework and appear in the drawings of the various embodiments.

[0047] If it is determined that a detailed description of a known function or constituent element related to this invention may unnecessarily obscure the gist of the invention, then such detailed description shall be omitted. Furthermore, the numbers used in the description (e.g., first, second, etc.) are merely identification symbols used to distinguish one constituent element from another.

[0048] Furthermore, the suffixes “module” and “part” used for constituent elements in the following description are merely assigned or used interchangeably for the convenience of writing the instruction manual, and do not in themselves have a distinguishing meaning or function.

[0049] The method according to the embodiments can be implemented and recorded on a computer-readable medium in the form of program instructions executable by various computer means. The computer-readable medium may include program instructions, data files, data structures, etc., individually or in combination. The program instructions recorded on the medium may be specifically designed and configured for the embodiments, or may be those known and usable by those skilled in the art. 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; magneto-optical media, such as floppy disks; and hardware devices specifically configured for storing and executing program instructions, such as ROMs, RAMs, and flash memory. Examples of program instructions include machine language code generated by a compiler, and high-level language code executable by a computer using an interpreter or similar means. The hardware device may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

[0050] 1. Diagnosis using retinal images

[0051] 1.1. Diagnostic System and Process

[0052] 1.1.1. Purpose and Definition

[0053] The following describes a diagnostic system and method that assists medical personnel in determining the presence of a disease or the presence of abnormalities that can be used as a basis for judgment, based on ocular images. In this specification, the term "diagnosis" may mean diagnostic assistance in aiding disease diagnosis, rather than direct diagnosis of a disease. The term "diagnosis" is used below for ease of explanation, but it can mean diagnostic assistance.

[0054] In particular, this document will explain a diagnostic method that utilizes deep learning technology to construct machine learning models for diagnosing diseases, and uses these models to assist in detecting the presence or abnormalities of diseases. In this specification, an eye image is an image including the subject's eye, and may include various images such as retinal images and / or fundus images. For ease of explanation, this specification will focus on retinal images in its description, but it is not limited to this, and the description can certainly be applied to other eye images.

[0055] The machine learning models described in this specification can be designed based on various machine learning libraries. For example, machine learning models can refer to various forms of models designed based on supervised, unsupervised, semi-supervised, or reinforcement learning artificial intelligence algorithms, such as decision trees, random forests, stochastic gradient descent, neural networks, k-nearest neighbors, linear regression, logistic regression, Cox proportional hazards model (survival model) (based on regression), support vector machines, k-means, hierarchical cluster analysis (HCA), expectation maximization, principal component analysis (PCA), kernel PCA, locally-linear embedding (LLE), t-SNE (t-distributed stochastic neighbor embedding), Apriori, Eclat, etc.

[0056] Unless otherwise specified, the machine learning model is primarily described in the following text using the neural network model. However, it does not necessarily refer only to models based on neural network algorithms. Within the scope of the functions and purposes of the invention as described in this specification, it can be replaced by models based on other algorithms, which is obvious.

[0057] According to one embodiment of the present invention, a diagnostic system or method based on retinal image-assisted diagnosis of at least one ophthalmic disease, cardiovascular disease (and / or cerebrovascular disease), kidney disease, or other systemic disease can be provided.

[0058] For example, in this specification, ophthalmic diseases may include at least one of the following: cataract, glaucoma, macular degeneration, diabetic retinopathy, epiretinal membrane, macular hole, high / degenerative myopia, melanoma, retinal detachment, dry eye syndrome, presbyopia, and astigmatism.

[0059] In addition, cardiovascular diseases can include at least one of the following: coronary artery disease (CAD), aortic stenosis, hypertension, arterial hypertension, heart failure, arrhythmia, atrial fibrillation, valvular heart disease, cardiomyopathy, peripheral vascular disease, peripheral artery disease (PAD), heart attack, aneurysm, aortic aneurysm (e.g., abdominal aortic aneurysm, thoracic aortic aneurysm), thrombotic disease (e.g., deep vein thrombosis, pulmonary embolism), myocardial infarction, and cardiovascular and cerebrovascular diseases. Furthermore, cardiovascular and cerebrovascular diseases can include at least one of the following: stroke, ischemic stroke, cerebral infarction, cerebral hemorrhage, subarachnoid hemorrhage, transient ischemic attack, and death due to cardiovascular disease. Cardiovascular diseases can also include complications. In addition, complications may include aspiration pneumonia, dysphagia, decreased motor function, decreased language function, decreased cognitive function, sleep disorders, mood disorders, neuralgia, urinary tract infection, malnutrition, deep vein thrombosis, bedsores, falls, pain, seizures, depression, etc.

[0060] In addition, kidney disease can include at least one of the following: chronic kidney disease (CKD), acute kidney injury (AKI), kidney stones, nephrotic syndrome, glomerulonephritis, polycystic kidney disease (PKD), kidney cancer, and pyelonephritis. Furthermore, kidney disease can include complications. These complications can include dialysis-related side effects, low blood pressure, muscle cramps, nausea and vomiting, headache, dialysis disequilibrium syndrome, itching, and cardiac arrhythmias.

[0061] In addition, other systemic diseases may include at least one of diabetes, hypertension, hypotension, Alzheimer's disease, cytomegalovirus, and arteriosclerosis.

[0062] Furthermore, according to another embodiment of the present invention, various parameters of the subject can be predicted based on retinal images. For example, the parameters may include at least one of the parameters representing the subject's physical information, such as biological age, sex, height, weight, BMI, body mass, body fat percentage, body muscle mass, etc. In addition, the parameters may include at least one of the diagnostic numerical parameters, such as hematocrit, red blood cell count, white blood cell count, hemoglobin level, platelet count, total iron-binding capacity (TIBC), iron level, ferritin (storage ferritin) level, total protein level, albumin level, aspartate aminotransferase (AST) level, aminotransferase level, γ-GTP, γ-GT, alkaline phosphatase (ALP) level, globulin level, hepatitis antigen level, hepatitis antibody level, glycated hemoglobin level (HbA1c), blood urea nitrogen (BUN) level, creatinine level, uric acid level, total cholesterol level, high-density lipoprotein (HDL) level, low-density lipoprotein (LDL) level, triglyceride (TG) level, bicarbonate level, systolic blood pressure (SBP), diastolic blood pressure (DBP), etc.

[0063] According to another embodiment of the present invention, a diagnostic system or method may be provided for detecting abnormal retinal features that can be used to diagnose ophthalmic diseases or other conditions. For example, a diagnostic system or method may be provided to obtain information on the following: abnormal color of the entire retina, lens opacity, abnormal cup-to-disc ratio of the optic disc, macular abnormalities (e.g., macular hole), abnormalities in vessel diameter and course, abnormalities in retinal artery diameter, retinal hemorrhage, microaneurysms, hard exudates, epiretinal membrane, myelinated nerve fibers, chorioretinal atrophy, retinal nerve fiber layer defect (RNFL defect), exudation, drusen, cataract, glaucoma, diabetic retinopathy, tessellated fundus, large optic cup, retinal vein occlusion (RVO), branch retinal vein occlusion (BRVO), central retinal vein occlusion (CRVO), retinal artery occlusion (RAO), and rhegmatogenous retinal detachment. RD, posterior serous / exudative retinal detachment, central serous chorioretinopathy (CSCR), VKH disease, macular disease, epiretinal membrane (ERM), macular hole (MH),Macular Hole, Pathological Myopia, Optic Nerve Degeneration, Optic Atrophy, Severe Hypertensive Retinopathy, Disc Swelling and Elevation, Dragged Disc, Pigmentary Degeneration, Congenital Disc Abnormality, Retinitis Pigmentosa, Bietti Crystalline Dystrophy, Peripheral Retinal Degeneration and Break, Myelinated Nerve Fiber, Vitreous Particles, Fundus Neoplasm, Massive Hard Exudates, Yellow-White Spots / Flecks, Cotton-Wool Spots, Vessel Tortuosity Tortuosity, choroidal retinal atrophy / defect, preretinal hemorrhage, fibrosis, laser spots, intraocular silicone oil, blurred fundus, blurred fundus without PDR, and blurred fundus with suspected PDR, etc.

[0064] In this specification, diagnostic information can be understood to include diagnostic information based on the determination of the presence or absence of a disease, or the observed information on which it is based.

[0065] 1.1.2. Diagnostic System Composition

[0066] According to one embodiment of the present invention, a diagnostic system can be provided.

[0067] Figure 1 A diagnostic system according to an embodiment of the present invention is illustrated. (Refer to...) Figure 1 The diagnostic system 1 may include: a learning device 10 for training a diagnostic model, a diagnostic device 20 for performing diagnoses using the diagnostic model, and a client device 30 for obtaining diagnostic requests. The diagnostic system 1 may include multiple learning devices, multiple diagnostic devices, or multiple client devices.

[0068] The learning device 10 may include a learning unit 100. The learning unit 100 can perform training of a diagnostic model. For example, the learning unit 100 can acquire a retinal image dataset and perform training on a diagnostic model for detecting diseases or abnormalities seen in retinal images. The learning unit 100 may be included in the processor of the learning device 10, as described later, meaning that the processor learns functional expressions of the processing.

[0069] The diagnostic device 20 may include a diagnostic unit 200. The diagnostic unit 200 may utilize a diagnostic model to perform disease diagnosis or acquire auxiliary information for diagnosis. For example, the diagnostic unit 200 may utilize a diagnostic model trained by a learning unit to acquire diagnostic information. The diagnostic unit 200 may be included in the processor of the diagnostic device 20 (described later), signifying a functional expression of the processor's diagnostic processing.

[0070] The client device 30 may include an imaging unit 300. The imaging unit 300 can capture retinal images. The client device may be an ophthalmic retinal imaging device. Alternatively, the client device 30 may be a handheld device such as a smartphone or tablet.

[0071] In the diagnostic system 1 of this embodiment, the learning device 10 can determine the diagnostic model for diagnosis by acquiring a dataset and performing learning of the diagnostic model. When the diagnostic device obtains an information request from the client, it uses the determined diagnostic model to obtain diagnostic information corresponding to the diagnostic target image. The client device can request information from the diagnostic device and obtain the diagnostic information transmitted in response.

[0072] A diagnostic system according to another embodiment may include: a diagnostic apparatus for performing diagnostic model learning and using the model to perform diagnosis, and a client apparatus. A diagnostic system according to yet another embodiment may include: a diagnostic apparatus for performing diagnostic model learning, obtaining a diagnostic request, and performing a diagnosis. A diagnostic system according to yet another embodiment may include: a learning apparatus for performing diagnostic model learning, and a diagnostic apparatus for obtaining a diagnostic request and performing a diagnosis.

[0073] The diagnostic system disclosed in this specification is not limited to the above embodiments and can be implemented in any form, including: a learning unit that performs model learning, a diagnostic unit that acquires diagnostic information based on the learned model, and an imaging unit that acquires diagnostic target images.

[0074] The following will describe some embodiments of the devices that constitute the system.

[0075] 1.1.2.1. Learning Device

[0076] The learning device according to one embodiment of the present invention can perform training of a diagnostic model for assisted diagnosis.

[0077] Figure 2 This is a block diagram illustrating a learning apparatus according to an embodiment of the present invention. (Refer to...) Figure 2 The learning device 10 may include a processor 12 and a storage module 11.

[0078] The learning device 10 may include a processor 12. The processor 12 can control the operation of the learning device 10.

[0079] Processor 12 may include one or more of a central processing unit (CPU), random access memory (RAM), graphics processing unit (GPU), one or more microprocessors, and other electronic components that process input data according to predetermined logic. Furthermore, processor 12 may be one or more components.

[0080] The processor 12 can read system programs and various processing programs stored in the storage module 11. For example, the processor 12 can expand the procedures and methods for performing the diagnostics described later onto RAM and perform various processing according to the expanded program. The processor 12 can perform the learning of the diagnostic model described later.

[0081] The learning device 10 may include a storage module 11. The storage module 11 may store the data and learning models required for learning.

[0082] The storage module 11 can be implemented in the form of non-volatile semiconductor memory, hard disk, flash memory, RAM, read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM) or other tangible non-volatile recording media.

[0083] Storage module 11 can store various processing programs, parameters for executing program processing, or data resulting from such processing. For example, storage module 11 can store data processing flow programs for performing the diagnostics described later, diagnostic flow programs, parameters for executing each program, and data obtained by executing such programs (e.g., processed data or diagnostic result values). Furthermore, storage module 11 can store various diagnostic models described later.

[0084] The learning device 10 may include a separate learning unit. The learning unit can perform the learning of the diagnostic model.

[0085] The learning unit may be included in the processor 12. The learning unit may be stored in the storage module 11. The learning unit may be implemented by a portion of the processor 12 and the storage module 11. For example, the learning unit may be stored in the storage module 11 and driven by the processor 12.

[0086] The learning device 10 may further include a communication module 13. The communication module 13 can communicate with external devices. For example, the communication module 13 can communicate with a diagnostic device, a server device, or a client device, as described later. The communication module 13 can perform wired or wireless communication. The communication module 13 can perform bidirectional or unidirectional communication.

[0087] 1.1.2.2. Diagnostic Device

[0088] Diagnostic devices can use diagnostic models to obtain diagnostic information.

[0089] Figure 3 This is a block diagram illustrating a diagnostic apparatus according to an embodiment of the present invention. (Refer to...) Figure 3 The diagnostic device 20 may include a processor 22 and a storage module 21.

[0090] Processor 22 may include one or more of a central processing unit (CPU), random access memory (RAM), graphics processing unit (GPU), one or more microprocessors, and other electronic components that process input data according to predetermined logic. Furthermore, processor 22 may be one or more components.

[0091] Processor 22 can read system programs and various processing programs stored in storage module 21. For example, processor 22 can expand the procedures and methods for performing the diagnosis described later onto RAM and execute various processes according to the expanded program. Processor 22 can generate diagnostic information using a diagnostic model. Processor 22 can acquire diagnostic data for diagnosis (e.g., retinal data of the subject) and use a learned diagnostic model to obtain diagnostic information predicted from the diagnostic data.

[0092] Storage module 21 can store the diagnostic model. Storage module 21 can store the parameters, variables, etc. of the diagnostic model.

[0093] The storage module 21 can be implemented in the form of non-volatile semiconductor memory, hard disk, flash memory, RAM, read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM) or other tangible non-volatile recording media.

[0094] Storage module 21 can store various processing programs, parameters used to execute program processing, or data resulting from such processing. For example, storage module 21 can store data processing flow programs, diagnostic flow programs, parameters used to execute each program, and data obtained by executing such programs (e.g., processed data or diagnostic result values). Furthermore, storage module 21 can store various diagnostic models, as described later.

[0095] Although not shown, the diagnostic device 20 may further include an input module. The input module can acquire user input. For example, the input module can acquire user input requesting diagnostic information. Furthermore, the input module can acquire at least one of the following: height, weight, age, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), and cholesterol value. The processor 22 can acquire the information input through the input module.

[0096] The diagnostic device 20 may further include a communication module 23. The communication module 23 can communicate with the learning device and / or the client device. For example, the diagnostic device 20 may be configured as a server communicating with the client device. This will be described in more detail below.

[0097] 1.1.2.3. Server Device

[0098] According to one embodiment of the present invention, the diagnostic system may include a server device. The diagnostic system according to one embodiment of the present invention may also include multiple server devices.

[0099] The server device can store and / or drive the diagnostic model. The server device can store the weight values ​​that constitute the learned diagnostic model. The server device can collect or store data used for diagnosis.

[0100] The server device can output the results of the diagnostic process using the diagnostic model to the client device. The server device can also receive feedback from the client device. The server device can operate similarly to the diagnostic device described above.

[0101] Figure 4 A diagnostic system according to an embodiment of the present invention is illustrated. (Refer to...) Figure 4 According to one embodiment of the present invention, the diagnostic system 20 may include a diagnostic server 40, a learning device, and a client device.

[0102] The diagnostic server 40, i.e., the server device, can communicate with multiple learning devices or multiple diagnostic devices. (See reference...) Figure 6 The diagnostic server 40 can communicate with the first learning device 10a and the second learning device 10b. (See reference...) Figure 4The diagnostic server 40 can communicate with the first client device 30a and the second client device 30b.

[0103] For example, the diagnostic server 40 can communicate with a first learning device 10a for training a first diagnostic model to obtain first diagnostic information and a second learning device 10b for training a second diagnostic model to obtain second diagnostic information.

[0104] The diagnostic server 40 can store a first diagnostic model for obtaining first diagnostic information and a second diagnostic model for obtaining second diagnostic information. In response to a diagnostic information acquisition request from a first client device 30a or a second client device 30b, it acquires diagnostic information and transmits the acquired diagnostic information to the first client device 30a or the second client device 30b.

[0105] Alternatively, the diagnostic server 40 may also communicate with a first client device 30a that requests first diagnostic information and a second client device 30b that requests second diagnostic information.

[0106] 1.1.2.4. Client Device

[0107] The client device can request diagnostic information from the diagnostic device or the server device. The client device can obtain the data required for diagnosis and transmit the obtained data to the diagnostic device.

[0108] Figure 5 This is a block diagram illustrating a client device according to an embodiment of the present invention. (Refer to...) Figure 5 According to one embodiment of the present invention, the client device 30 may include an imaging module 31, a processor 32 and a communication module 33.

[0109] Imaging module 31 can acquire image or video data. Imaging module 31 can acquire retinal images. However, client device 30 can also be replaced by a data acquisition unit other than imaging module 31.

[0110] The communication module 33 can communicate with external devices, such as diagnostic devices or server devices. The communication module 33 can perform wired or wireless communication.

[0111] The processor 32 can control the imaging module 31 to acquire images or data. The processor 32 can control the imaging module 31 to acquire retinal images. The processor 32 can transmit the acquired retinal images to a diagnostic device. The processor 32 can transmit the images acquired by the imaging module 31 to a server device via the communication module 33, and obtain diagnostic information generated based on this information.

[0112] Although not shown, the client device may further include an output module. The output module may include a display for outputting video or images, or a speaker for outputting sound. The output module may output video or image data acquired by the imaging unit. The output module may output diagnostic information acquired from the diagnostic device.

[0113] Although not shown, the client device may further include an input module. The input module may acquire user input. For example, the input module may acquire user input requesting diagnostic information. The input module may acquire user information used to evaluate diagnostic information obtained from the diagnostic device. Furthermore, the input module may acquire at least one of the following: height, weight, age, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), and cholesterol level.

[0114] Furthermore, although not shown, the client device may further include a storage module. The storage module can store images acquired by the imaging unit.

[0115] 1.1.3. Overview of the Diagnostic Process

[0116] The diagnostic system or diagnostic device disclosed in this specification can perform diagnostic procedures. Diagnostic procedures can be broadly divided into a learning procedure that learns a diagnostic model for diagnosis and a diagnostic procedure that utilizes the diagnostic model.

[0117] Figure 6 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention. (Refer to...) Figure 6 According to an embodiment of the present invention, the diagnostic process may include: a learning process of acquiring and processing data S11, learning a diagnostic model S12, and acquiring parameters of the learned diagnostic model S13; and a diagnostic process of acquiring diagnostic target data S21 and acquiring diagnostic information S23 using a diagnostic model learned based on the diagnostic target data S22.

[0118] More specifically, the learning process may include: a data processing process that transforms the input learning image data into states suitable for model training, and a learning process that trains the model using the processed data. The learning process may be executed by the aforementioned learning device.

[0119] The diagnostic process may include: a data processing process that processes the input image data of the target object into a state that can be used for diagnosis using a diagnostic model, and a diagnostic process that performs the diagnosis using the processed data. The diagnostic process may be executed by the aforementioned diagnostic device or server device.

[0120] The following will explain each process.

[0121] 1.2. Learning Process

[0122] According to one embodiment of the present invention, a process for training a diagnostic model can be provided. Specifically, a training process for a diagnostic model that performs or assists in diagnosis based on retinal images can be disclosed.

[0123] The learning process described below can be executed by the learning device described above.

[0124] 1.2.1. Academic Department

[0125] According to one embodiment of the present invention, the learning process can be executed by a learning unit. The learning unit can be located within the aforementioned learning device.

[0126] Figure 7 This is a diagram illustrating the configuration of a learning unit according to an embodiment of the present invention. (Refer to...) Figure 7 The learning unit 100 may include a data processing module 110, a queue module 130, a learning module 150, and a learning result acquisition module 170. Each module can execute the various steps of the data processing flow and learning flow described later. However, Figure 7 Not all of the constituent elements and the functions performed by each element described herein are necessary. Depending on the learning format, some elements may be added or omitted.

[0127] 1.2.2. Data Processing Flow

[0128] 1.2.2.1. Image Data Acquisition

[0129] According to one embodiment of the present invention, a dataset can be acquired. According to one embodiment of the present invention, a data processing module can acquire a dataset.

[0130] The dataset can be an image dataset.

[0131] For example, it could be a retinal image dataset. Retinal image datasets can be acquired using a general non-mydriatic retinal camera. Retinal images can be panoramic retinal images or wide-field retinal images. Retinal images can be red-free images. Retinal images can be images captured using infrared light. Retinal images can be images captured using autofluorescence. Image data can be acquired in any of the following formats: JPG, PNG, DCM (DICOM), BMP, GIF, or TIFF.

[0132] For example, the dataset can be an image dataset, including any one of optical coherence tomography (OCT) images, OCT angiography images, or retinal angiography images. In this case, a diagnostic model learned using the dataset including OCT images, OCT angiography images, or retinal angiography images can predict or output diagnostic information (or labels) based on the target OCT image, target OCT angiography image, or target retinal angiography image.

[0133] A dataset can include a training dataset. A dataset can include a test dataset. A dataset can include a validation dataset. In other words, a dataset can be assigned to at least one of the training dataset, the test dataset, and the validation dataset.

[0134] The dataset can be determined based on the diagnostic information to be obtained by the diagnostic model learned from that dataset. For example, if the goal is to train a diagnostic model that acquires diagnostic information related to cataracts, the dataset could be an infrared retinal image dataset. Alternatively, if the goal is to train a diagnostic model that acquires diagnostic information related to macular degeneration, the dataset could be an autofluorescence retinal image dataset.

[0135] Individual data points included in a dataset may include labels. There may be multiple labels. In other words, individual data points included in a dataset may be labeled for at least one feature. For example, a dataset may be a retinal image dataset that includes multiple retinal image data points, each of which may include labels for diagnostic information (e.g., presence or absence of a specific disease) and / or visible information (e.g., presence or absence of abnormalities in a specific area) corresponding to that image.

[0136] For example, the dataset is a retinal image dataset, and each retinal image data can include peripheral information labels about the image. For example, each retinal image data can include peripheral information labels such as: information about whether the retinal image is a left or right eye image, gender information about whether it is a female or male retinal image, and age information about the subject who took the retinal image.

[0137] Figure 8 This is a conceptual diagram illustrating an image dataset according to an embodiment of the present invention. (Refer to...) Figure 8 According to one embodiment of the present invention, an image dataset DS may include multiple image data (IDs). Each image data (ID) may include an image (I) and a label (L) assigned to the image. See reference... Figure 10The image dataset DS may include a first image data ID1 and a second image data ID2. The first image data ID1 may include a first image I1 and a first label L1 corresponding to the first image.

[0138] exist Figure 8 The illustration is based on the case where an image data includes one label, but as mentioned above, an image data can include multiple labels.

[0139] 1.2.2.2. Image Preprocessing

[0140] According to one embodiment of the present invention, image preprocessing can be performed. If the image is input as is for learning, overfitting may occur due to learning unnecessary features, and the learning efficiency will also decrease.

[0141] To prevent this, the data processing module can perform appropriate preprocessing on the image data to meet the learning objectives, thereby improving learning efficiency and performance.

[0142] In one embodiment, the data processing module may utilize or appropriately combine various techniques to perform image preprocessing, such as image resizing, grayscale conversion, histogram equalization, normalization, feature enhancement, image enhancement, denoising, edge detection, segmentation, morphological operations, color space conversion, etc. Image preprocessing using some of these techniques will be described below.

[0143] 1.2.2.2.1. Image Resizing

[0144] According to one embodiment of the present invention, the size of the acquired image data can be adjusted. That is, the image can be resized. According to one embodiment of the present invention, the image resizing can be performed by the data processing module of the learning unit described above.

[0145] You can adjust the size or aspect ratio of an image. Multiple images can be resized to maintain a constant size. Alternatively, an image can be resized to maintain a constant aspect ratio. Resizing an image can also be achieved by applying an image transformation filter.

[0146] If the size or file size of a single image is too large or too small, its size or file size can be adjusted to be of an appropriate size. Alternatively, if the sizes or files of individual images are different, their sizes or file sizes can be standardized by resizing.

[0147] According to one embodiment, the image size can be adjusted. For example, if the image size exceeds an appropriate range, it can be reduced by downsampling. Alternatively, if the image size does not meet an appropriate range, it can be enlarged by upsampling or interpolating.

[0148] According to another embodiment, the size or aspect ratio of an image can be adjusted by cropping the image or adding pixels to the acquired image. For example, if the image contains regions that are not necessary for learning, a portion of the image can be cropped to remove these regions. Alternatively, if cropping a portion of the image results in an aspect ratio mismatch, the image aspect ratio can be adjusted by adding columns or rows. In other words, the aspect ratio can be adjusted by adding margins or padding to the image.

[0149] According to yet another embodiment, the image size and aspect ratio can be adjusted together. For example, if the image size is large, it can be reduced by downsampling the image and converted into appropriate image data by cropping out unnecessary regions contained in the reduced image.

[0150] Furthermore, according to another embodiment of the present invention, the orientation of the image data can also be changed.

[0151] Specifically, when using a retinal image dataset as the dataset, each retinal image can have its size or volume adjusted. Cropping can be performed to remove blank areas from the retinal image excluding the retinal portion, or padding can be performed to fill in the cropped portions of the retinal image to adjust the aspect ratio.

[0152] 1.2.2.2.2. Feature Enhancement

[0153] According to one embodiment of the present invention, the data processing module may perform preprocessing to highlight features of the retinal image. For example, the data processing module may perform preprocessing on the retinal image to more easily detect abnormal signs of ophthalmic diseases, or to highlight changes in retinal vessels or blood flow.

[0154] For example, image preprocessing can be performed on an image that has already undergone the resizing process described above. However, the invention disclosed in this specification is not limited to this, and image preprocessing can also be performed without omitting the resizing process. Image preprocessing can involve applying a preprocessing filter to the image.

[0155] According to one embodiment, a blur filter can be applied to an image. A Gaussian filter can be applied to an image. A Gaussian blur filter can also be applied to an image. Alternatively, a deblur filter can be applied to an image to make it sharper.

[0156] According to another embodiment, filters that adjust or modulate the colors of an image can be applied. For example, filters that change the values ​​of some components in the RGB values ​​that make up the image, or filters that binarize the image, can be applied.

[0157] According to yet another embodiment, filters that highlight specific elements can be applied. For example, preprocessing can be performed on retinal image data to highlight vascular elements in each image. In this case, the preprocessing to highlight vascular elements can involve applying one or more filters sequentially or in combination.

[0158] Furthermore, according to one embodiment of the present invention, image preprocessing can be performed based on the characteristics of the diagnostic information to be acquired. For example, if it is desired to acquire diagnostic information related to findings such as retinal hemorrhage, drusen, microaneurysms, and exudates, preprocessing can be performed to convert the acquired retinal image into a red-light-free retinal image format.

[0159] 1.2.2.2.3. Image Enhancement

[0160] According to one embodiment of the present invention, an image can be enhanced or expanded. Image enhancement can be performed by the data processing module of the learning unit described above.

[0161] Enhanced images can be used to improve the training performance of diagnostic models. For example, if the amount of data used to train a diagnostic model is insufficient, the amount of data used for training can be increased by modulating the existing training image data, and by using modulated (or altered) images and the original images to increase the amount of training image data. Therefore, overfitting can be suppressed, deeper model layers can be formed, and prediction accuracy can be improved.

[0162] For example, image data expansion can be performed by horizontally flipping the image, cropping a portion of the image, correcting the image's color values, or adding artificial noise. Specifically, cropping a portion of the image can be done by cropping a portion of the elements that make up the image or by randomly cropping a portion of the image. Furthermore, image data can be expanded by horizontally flipping, vertically flipping, proportionally resizing, cropping, filling, color adjustment, or brightness adjustment.

[0163] Furthermore, in one embodiment, the data processing module can enhance the image by rotating the retinal image. Since the retina is circular, there is no data loss even when the retina is rotated. Therefore, when image enhancement is performed by rotating the retinal image, multiple retinal images can be acquired without data loss. For example, the data processing module can rotate the retinal image based on a predetermined angle (e.g., 15 degrees, 30 degrees, 60 degrees, etc.) to acquire multiple retinal images.

[0164] Furthermore, in one embodiment, the enhancement or expansion of the image data described above can typically be applied to the training dataset. However, it can also be applied to other datasets, such as a test dataset, i.e., a dataset used to test a model that has completed learning using the training data and validation using the validation data.

[0165] Specifically, when using a set of retinal images as a dataset, one or more of the following processes can be randomly applied: horizontal flipping, cropping, adding noise, or changing colors, to increase the amount of data and thus obtain an enhanced retinal image dataset.

[0166] 1.2.2.3. Image Serialization

[0167] According to one embodiment of the present invention, image data can be linearized. The image can be linearized by the data processing module of the learning unit described above. The linearization module can serialize the preprocessed image data and transmit it to the queue module.

[0168] If image data is used for learning as is, it needs to be decoded because the image data is in formats such as JPG, PNB, and DCM. If decoding is performed every time before learning, the model's learning performance may degrade. Therefore, instead of using image files for learning as is, image data can be serialized before learning. This serialization of image data improves learning performance and speed. The serialized image data can be image data that has undergone one or more of the image resizing and preprocessing steps described above, or it can be raw, unprocessed image data.

[0169] Each image data point in the image dataset can be converted to string format. Image data can also be converted to binary data format. Specifically, image data can be converted to a format suitable for use in diagnostic model learning. For example, image data can be converted to TFRecord format for use in diagnostic model learning with TensorFlow (tensorflow).

[0170] Specifically, when using a set of retinal images as the dataset, the acquired retinal image set can be converted into TFRecord format and used for learning the diagnostic model.

[0171] 1.2.2.4. Queue

[0172] Queues can be used to address data bottlenecks. The queue module in the learning unit described above can store image data in a queue and transmit it to the learning model module.

[0173] Especially when using both the central processing unit (CPU) and the graphics processing unit (GPU) to execute the learning process, the use of queues can minimize the bottleneck between the CPU and GPU, make database access smoother, and improve memory usage efficiency.

[0174] Queues can store data used for diagnostic model learning. Queues can store image data. Image data stored in queues can be image data that has undergone at least one of the above data processing steps (i.e., resizing, preprocessing, and enhancement), or it can be the acquired raw image.

[0175] The queue can store image data, preferably serialized image data as described above. The queue can store image data and feed it to the diagnostic model. The queue can transmit image data to the diagnostic model in batches.

[0176] Queues can provide image data. Queues can also supply data to the learning module, which will be discussed later. As data is extracted from the learning module, the amount of data accumulating in the queue may decrease.

[0177] As the diagnostic model learns, when the amount of data stored in the queue decreases below the baseline, the queue can request additional data. The queue can request additional data of specific types. When additional data is requested, the learning unit can add data to the queue.

[0178] Queues can be set up in the system memory of the learning device. For example, queues can be formed in the random access memory (RAM) of the central processing unit (CPU). In this case, the size of the queue, i.e., its capacity, can be determined based on the RAM capacity of the CPU. As queues, first-in-first-out (FIFO) queues, primary queues, or random queues can be used.

[0179] 1.2.3. Learning Process

[0180] According to one embodiment of the present invention, the learning process of a diagnostic model can be disclosed.

[0181] According to one embodiment of the present invention, the learning of the diagnostic model can be performed by the aforementioned learning device. The learning process can be performed by the processor of the aforementioned learning device. The learning process can be performed by the learning module of the aforementioned learning unit.

[0182] Figure 9 This is a block diagram illustrating the diagnostic model learning process according to an embodiment of the present invention. (Refer to...) Figure 9 According to an embodiment of the present invention, the diagnostic model learning process can be performed through the following steps: acquiring data S31, learning the diagnostic model S32, verifying the learned model S33, and acquiring the variables of the learned model S34.

[0183] 1.2.3.1. Data Input

[0184] You can obtain datasets for learning diagnostic models.

[0185] The acquired data can be an image dataset that has undergone the data processing steps described above. For example, the dataset may include resized, pre-processed, augmented, and subsequently serialized retinal image data.

[0186] During the learning phase of the diagnostic model, a training dataset can be acquired and used. During the validation phase, a validation dataset can be acquired and used. During the testing phase, a test dataset can be acquired and used. Each dataset may include retinal images and labels.

[0187] Datasets can be retrieved from a queue. Datasets can be retrieved from the queue in batch sizes. For example, if the batch size is specified as 60, the dataset can be retrieved from the queue in units of 60. The batch size can be limited by the GPU's RAM capacity.

[0188] The dataset can be randomly selected from the queue to receive learning modules. Alternatively, the dataset can be selected in the order it accumulates in the queue.

[0189] The learning module can specify the composition of the dataset retrieved from the queue for extraction. For example, the learning module can extract retinal image data with left-eye labels and right-eye labels for a specific subject for use in learning.

[0190] The learning module can retrieve datasets with specific labels from a queue. For example, it can retrieve a dataset of retinal images labeled "abnormal" based on diagnostic information. The learning module can also retrieve datasets from the queue in a ratio corresponding to a specified label. For instance, it can retrieve a retinal image dataset such that the ratio of retinal images labeled "abnormal" to those labeled "normal" is 1:1.

[0191] 1.2.3.2. Model Design

[0192] The diagnostic model can be designed as a neural network model, a machine learning model, or various other models. In one embodiment, if the diagnostic model includes a neural network model, the neural network model may include multiple layers or levels.

[0193] Neural network models can be implemented as classifiers that generate diagnostic information. Classifiers can perform binary or multi-class classification. For example, a neural network model can be a binary classification model that classifies input data into normal or abnormal categories for target diagnostic information such as a specific disease or abnormal sign. Alternatively, a neural network model can be a multi-class classification model that classifies input data into multiple hierarchical categories for a specific characteristic (e.g., the degree of disease progression). Alternatively, a neural network model can also be implemented as a regression model that outputs a specific numerical value associated with a specific disease.

[0194] Neural network models can include convolutional neural networks (CNNs). As a CNN architecture, at least one of AlexNet, LENET, NIN, VGGNet, ResNet, WideResNet, GoogleNet, FractaNet, DenseNet, FitNet, RitResNet, HighwayNet, MobileNet, and DeeplySupervisedNet can be used. A neural network model can be implemented using multiple CNN architectures.

[0195] For example, a neural network model can be implemented as comprising multiple VGGNet blocks. More specifically, a neural network model can be prepared by combining the following structure: a first structure that sequentially combines a CNN layer, a Batch Normalization (BN) layer, and a ReLU layer with 64 filters of size 3x3, and a second block that sequentially combines a CNN layer, a ReLU layer, and a BN layer with 128 filters of size 3x3.

[0196] Neural network models can include a max pooling layer after each CNN block, and at the end, a global average pooling (GAP) layer, a fully connected (FC) layer, and an activation layer (e.g., Sigmoid, Softmax, etc.).

[0197] In another embodiment, if the diagnostic model includes a machine learning model, the machine learning model may include a linear regression model, a Cox proportional hazards model, etc.

[0198] 1.2.3.3. Model Learning

[0199] Diagnostic models can learn using training datasets.

[0200] Diagnostic models can learn using labeled datasets. However, the learning process for diagnostic models described in this specification is not limited to this; diagnostic models can also learn in an unsupervised manner using unlabeled data.

[0201] The learning of the diagnostic model can be performed as follows: based on training image data, a diagnostic model with arbitrary weights is used to obtain result values; the obtained result values ​​are compared with the label values ​​of the training data; and backpropagation is performed based on the error to optimize the weight values. Furthermore, the learning of the diagnostic model can be influenced by model validation results, test results, and / or feedback from the diagnostic phase (described later).

[0202] The learning of the diagnostic model described above can be performed using TensorFlow. However, this invention is not limited to this; frameworks such as Theano, Keras, Caffe, Torch, and CNTK (Microsoft Cognitive Toolkit) can also be used for learning the diagnostic model.

[0203] 1.2.3.4. Model Validation

[0204] Diagnostic models can be validated using validation datasets. Validation of a diagnostic model can be performed by taking the result values ​​from the validation dataset from the learned diagnostic model and comparing those results with the labels on the validation dataset. Validation can be performed by measuring the accuracy of the result values. Based on the validation results, the parameters (e.g., weights and / or biases) or hyperparameters (e.g., learning rate) of the diagnostic model can be adjusted.

[0205] For example, a learning device according to an embodiment of the present invention can train a diagnostic model based on retinal image prediction diagnostic information, and perform the verification of the diagnostic model by comparing the diagnostic information of the learned model's verification retinal image with the verification label corresponding to the verification retinal image.

[0206] For validating a diagnostic model, a separate validation set can be used, which is a dataset containing discriminative factors not included in the training dataset. For example, a separate validation set could be a dataset that differs from the training dataset in terms of factors such as race, environment, age, and gender.

[0207] 1.2.3.5. Model Testing

[0208] The diagnostic model can be tested using a test dataset.

[0209] According to one embodiment of the present invention, the learning process can be used to test the diagnostic model using a test dataset that is separate from the training and validation datasets. Based on the test results, the parameters (e.g., weights and / or biases) or hyperparameters (e.g., learning rate) of the diagnostic model can be adjusted.

[0210] For example, a learning device according to an embodiment of the present invention can obtain result values ​​from a diagnostic model that has been learned to predict diagnostic information based on retinal images, with test retinal image data not used for training and validation as input, in order to perform a test on the learned and validated diagnostic model.

[0211] For testing diagnostic models, a separately prepared external data set can be used, which is a dataset with factors that are distinct from the training and / or validation data.

[0212] 1.2.3.6. Output of Results

[0213] As a result of diagnostic model learning, optimized parameter values ​​can be obtained. As mentioned above, by repeatedly learning the model using the test dataset, more suitable parameter (or variable) values ​​can be obtained. Once the learning is complete, optimized values ​​for the weights and / or biases can be obtained.

[0214] According to one embodiment of the present invention, the learned diagnostic model and / or the parameters or variables of the learned diagnostic model can be stored in a learning device and / or a diagnostic device (or server). The learned diagnostic model can be used by the diagnostic device and / or client device, etc., for predicting diagnostic information. Furthermore, the parameters or variables of the learned diagnostic model can also be updated through feedback obtained from the diagnostic device or client device.

[0215] 1.2.3.7. Model Ensemble

[0216] According to one embodiment of the present invention, multiple sub-models can be learned simultaneously during the learning of a diagnostic model. These multiple sub-models can have different hierarchical structures.

[0217] In this case, the diagnostic model according to an embodiment of the present invention can be implemented by combining multiple sub-diagnostic models. In other words, the learning of the diagnostic model can be performed using an ensemble technique that combines multiple sub-diagnoses.

[0218] When diagnostic models are constructed by forming an ensemble, predictions can be performed by integrating the results predicted from various forms of sub-diagnostic models, thereby further improving the accuracy of the predictions.

[0219] 1.3. Diagnostic Process

[0220] According to one embodiment of the present invention, a process (or diagnostic procedure) for obtaining diagnostic information using a diagnostic model can be provided. Specifically, through the diagnostic procedure, diagnostic information (e.g., diagnostic information or seen information) can be predicted using retinal images and a learned diagnostic model.

[0221] The diagnostic procedure described below can be performed by a diagnostic device.

[0222] 1.3.1. Diagnostic Department

[0223] According to one embodiment of the present invention, the diagnostic process can be executed by the processor of the aforementioned diagnostic device. The processor can be disposed within the aforementioned diagnostic device.

[0224] Figure 10 This is a diagram illustrating the configuration of a diagnostic unit according to an embodiment of the present invention. (Refer to...) Figure 10 The diagnostic unit 200 may include a diagnostic request acquisition module 210, a data processing module 230, a diagnostic module 250, and an output module 270.

[0225] Each module can execute the various steps of the data processing and learning processes described later. However, Figure 10 The constituent elements and functions described herein are not all necessary; depending on the diagnostic aspect, some elements may be added or omitted.

[0226] 1.3.2. Data Acquisition and Diagnostic Requests

[0227] A diagnostic apparatus according to an embodiment of the present invention can acquire diagnostic target data and obtain diagnostic information based thereon. The diagnostic target data may be image data. Data acquisition and diagnostic request acquisition can be performed by the diagnostic request acquisition module of the aforementioned diagnostic unit.

[0228] For example, diagnostic target data (TD) may include diagnostic target images (TI) and diagnostic target patient information (PI).

[0229] A diagnostic target image (TI) can be an image used to obtain diagnostic information about a target object. For example, a diagnostic target image can be a retinal image and / or a retinal image. Diagnostic targets (TIs) can be in any of the following formats: JPG, PNG, DCM (DICOM), BMP, GIF, or TIFF.

[0230] Diagnostic subject information (PI) can be information used to identify the target subject for diagnosis. Alternatively, diagnostic subject information (PI) can be characteristic information of the target subject or image. For example, diagnostic subject information (PI) may include the time the image was captured, the imaging device, the identification number, ID, name, gender, age, weight, ethnicity, smoking status, blood pressure (hypertension or diabetes), etc. If the target image is a retinal image, the diagnostic subject information (PI) may further include eye-related information, such as binocular information indicating whether it is a left or right eye image.

[0231] The diagnostic device can acquire diagnostic requests. The diagnostic device can acquire diagnostic target data along with the diagnostic request. When acquiring a diagnostic request, the diagnostic device can utilize a learned diagnostic model to obtain diagnostic information. The diagnostic device can acquire diagnostic requests from client devices. Alternatively, the diagnostic device can acquire diagnostic requests from the user through a separately configured input method.

[0232] 1.3.3. Data Processing Flow

[0233] The acquired data can be processed. Data processing can be performed by the data processing module of the diagnostic department mentioned above.

[0234] The data processing flow can typically be executed similarly to the data processing flow in the learning flow described above. The following section will focus on the differences between the data processing flow in the learning flow and the data processing flow in the diagnostic flow.

[0235] In the diagnostic process, the diagnostic device can acquire data in the same way as in the learning process. The acquired data can then have the same format as the data acquired in the learning process. For example, if the learning device trained a diagnostic model using DCM format image data in the learning process, the diagnostic device can acquire DCM images and use the learned diagnostic model to obtain diagnostic information.

[0236] In the diagnostic process, the acquired diagnostic target images can be resized, just like the image data used in the learning process. The diagnostic target images can be reshaped to have appropriate capacity, size, and / or aspect ratio in order to effectively perform diagnostic information predictions through the learned diagnostic model.

[0237] For example, if the target image for diagnosis is a retinal image, then in order to predict diagnostic information based on the retinal image, resizing can be performed, such as cropping unnecessary parts of the image or reducing its size.

[0238] In the diagnostic process, preprocessing filters can be applied to the acquired diagnostic target images, similar to the image data used in the learning process. Appropriate filters can be applied to the diagnostic target images to further improve the accuracy of diagnostic information predictions made by the learned diagnostic model.

[0239] For example, if the target image for diagnosis is a retinal image, preprocessing that helps predict diagnostic information can be applied to the target image, such as image preprocessing that emphasizes blood vessels or image preprocessing that emphasizes or weakens specific colors.

[0240] In the diagnostic process, the acquired diagnostic target images can be serialized, just like the image data used in the learning process. Diagnostic target images can be transformed or serialized into a form that facilitates driving the diagnostic model within a specific working framework.

[0241] The serialization of the diagnostic target image can be omitted. This is because, unlike the learning phase, the amount of data processed by the processor at one time is small in the diagnostic phase, thus placing a relatively small burden on data processing speed.

[0242] In the diagnostic process, the acquired diagnostic target images can be stored in a queue, just like the image data used in the learning process. However, since the amount of data processed in the diagnostic process is smaller than that in the learning process, the step of storing the data in a queue can be omitted.

[0243] On the other hand, in the diagnostic process, since no increase in data volume is required, data augmentation or image enhancement processes can be omitted in order to obtain accurate diagnostic information, unlike the learning process.

[0244] 1.3.4. Diagnostic Process

[0245] According to one embodiment of the present invention, a process for performing diagnosis using a learned diagnostic model can be disclosed. The diagnostic process can be executed in the aforementioned diagnostic apparatus. The diagnostic process can be executed in the aforementioned diagnostic server. The diagnostic process can be executed in the control unit of the aforementioned diagnostic apparatus. The diagnostic process can be executed by the diagnostic module of the aforementioned diagnostic unit.

[0246] Figure 11 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention. (Refer to...) Figure 11 The diagnostic process can be performed through the following steps: acquiring diagnostic target data S31, utilizing the learned diagnostic model S42, and obtaining the results corresponding to the acquired diagnostic target data S43. However, data processing can be performed selectively.

[0247] The following will refer to Figure 11 This explains each step of the diagnostic process.

[0248] 1.3.4.1. Data Input

[0249] According to one embodiment of the present invention, the diagnostic module can acquire diagnostic target data. The acquired data may be data that has been processed as described above. For example, the acquired data may be retinal image data of a subject that has been resized and preprocessed to highlight prominent blood vessels. According to one embodiment of the present invention, left-eye and right-eye images of a subject may be input together as diagnostic target data.

[0250] 1.3.4.2. Data Classification

[0251] A diagnostic model configured as a classifier can classify an input diagnostic target image into a positive or negative category based on a predetermined label.

[0252] A learned diagnostic model can receive diagnostic target data as input and output predicted labels. A learned diagnostic model can output predicted values ​​of diagnostic information. Diagnostic information can be obtained using a learned diagnostic model. Diagnostic information can be determined based on the predicted labels.

[0253] For example, a diagnostic model can predict diagnostic information (i.e., information about the presence or absence of a disease) or observed information (i.e., information about the presence or absence of abnormalities) for an examinee's ophthalmological or systemic diseases. In this case, the diagnostic or observed information can be output in probabilistic form. For example, the probability that the examinee has a specific disease or the probability that a specific abnormality is seen in the examinee's retinal image can be output. When using a diagnostic model set up as a classifier, the predicted label can be determined by considering whether the output probability value (or prediction score) exceeds a threshold.

[0254] Specifically, the diagnostic model can use the retinal image of the subject as the diagnostic target image and output the probability value of whether the subject has diabetic retinopathy. When using a diagnostic model with a classifier where normal is 1, the subject's retinal photograph can be input into the model, and the probability value of whether the subject has diabetic retinopathy can be obtained in the form of normal:abnormal = 0.74:0.26, etc.

[0255] This explanation is based on the case of using a diagnostic model in the form of a classifier to classify data, but the invention is not limited to this. Diagnostic models implemented in the form of regression models can also be used to predict specific diagnostic values ​​(e.g., blood pressure).

[0256] According to another embodiment of the present invention, applicability information of an image can be obtained. This applicability information can indicate whether a diagnostic target image is suitable for obtaining diagnostic information using a diagnostic model.

[0257] Image suitability information can be image quality information. Quality information or suitability information can indicate whether a diagnostic target image meets a baseline level.

[0258] For example, if the target image for diagnosis is defective due to limitations of the imaging equipment or lighting conditions during capture, an "inapplicable" result can be output as applicability information. If the noise contained in the target image exceeds a certain level, the target image can be determined to be inapplicable.

[0259] Applicability information can be obtained by using values ​​predicted by a diagnostic model. Alternatively, applicability information can be obtained through a separate image analysis process.

[0260] According to one embodiment, diagnostic information based on inapplicable images can be obtained even if the image is classified as inapplicable.

[0261] According to one embodiment, images classified as inapplicable can be re-examined by a diagnostic model.

[0262] At this point, the diagnostic model used for the re-examination can be different from the diagnostic model used for the initial examination. For example, the diagnostic device can store a first diagnostic model and a second diagnostic model, and images classified as inapplicable by the first diagnostic model can be examined using the second diagnostic model.

[0263] According to another embodiment of the present invention, a map can be obtained from a learned diagnostic model. Diagnostic information may include the map. The map may be obtained together with other diagnostic information. For example, the map may include a saliency map, a class activation map (CAM), a heat map, etc. Furthermore, the CAM can be selectively obtained. For example, the CAM can be extracted and / or output when the diagnostic information or observed information obtained by the diagnostic model is classified as an abnormal category.

[0264] 1.3.5. Output of Diagnostic Information

[0265] Diagnostic information can be determined based on the results output by the diagnostic model. The output of diagnostic information can be executed by the output module of the aforementioned diagnostic unit. Diagnostic information can be output from the diagnostic device to the client device. Diagnostic information can be output from the diagnostic device to the server device. Diagnostic information can be stored in the diagnostic device or the diagnostic server. Diagnostic information can also be stored in a separately configured server device, etc.

[0266] Diagnostic information can be databased and managed. For example, acquired diagnostic information can be stored and managed along with the corresponding diagnostic target image of the examinee, based on the examinee's identification number. In this case, the examinee's diagnostic target image and diagnostic information can be managed in chronological order. By managing diagnostic information and diagnostic target images in a time-series manner, the history of an individual's diagnostic information can be easily tracked and managed.

[0267] Diagnostic information can be provided to the user. This information can be provided through the output of a diagnostic device or client device. The diagnostic information can be output through visual or auditory means provided in the diagnostic device or client device for the user to perceive.

[0268] According to one embodiment of the present invention, an interface for effectively providing diagnostic information to a user can be provided.

[0269] Furthermore, if an image is classified as inapplicable, its applicability information can be provided together. For example, if an image is classified as inapplicable, the diagnostic information and inapplicability determination information corresponding to that image can be provided together.

[0270] Diagnostic target images deemed inapplicable can also be classified as target images to be re-captured. In this case, guidance can be provided, along with applicability information, for re-capture of the objects in the images classified as re-capture targets. On the other hand, in response to providing diagnostic information obtained through the diagnostic model, feedback related to the learning of the diagnostic model can be obtained. For example, feedback for adjusting parameters or hyperparameters related to the learning of the diagnostic model can be obtained. This feedback can be obtained through an input module set in the diagnostic device or client device.

[0271] According to one embodiment of the present invention, the diagnostic information corresponding to the diagnostic target image may include grading information. The grading information can be selected from multiple gradations. The grading information can be determined based on diagnostic information and / or seen information obtained through a diagnostic model. The grading information can be determined taking into account the applicability or quality information of the diagnostic target image. If the diagnostic model is a classifier model that performs multi-class classification, the grading information can be determined considering the category into which the diagnostic target image is classified by the diagnostic model. If the diagnostic model is a regression model that outputs a numerical value related to a specific disease, the grading information can be determined considering the output numerical value. Furthermore, the grading information can be determined by applying a predetermined cutoff value to the score corresponding to the diagnostic information.

[0272] For example, the diagnostic information corresponding to the target image can include any level of information selected from either first-level or second-level information. If anomaly observation or diagnostic information is obtained through the diagnostic model, the first-level information can be selected. If no anomaly observation or diagnostic information is obtained through the diagnostic model, the second-level information can be selected. Alternatively, if the value obtained through the diagnostic model exceeds a baseline value, the first-level information can be selected; and if the obtained value does not meet the baseline value, the second-level information can also be selected. First-level information can indicate a stronger anomaly in the target image compared to second-level information.

[0273] On the other hand, if image analysis or a diagnostic model determines that the quality of the target image is below the baseline, the grading information can be selected as third-grade information. Alternatively, the diagnostic information may include third-grade information as well as first or second-grade information.

[0274] 1.4. Diagnostic systems using multiple diagnostic models

[0275] According to one embodiment of the present invention, diagnostic information can be output using a diagnostic model as described above. The diagnostic model can consist of one diagnostic model or multiple diagnostic models. Furthermore, if the diagnostic model consists of multiple diagnostic models, these models can be configured in parallel or sequentially. Parallel diagnostic models and sequential diagnostic models will be described below.

[0276] 1.4.1.1. Composition of Parallel Diagnostic System

[0277] According to one embodiment of the present invention, a parallel diagnostic system for acquiring multiple diagnostic information can be provided. The parallel diagnostic system can train multiple diagnostic models for acquiring multiple diagnostic information and utilize the learned diagnostic models to acquire multiple diagnostic information.

[0278] For example, a parallel diagnostic system can train a first diagnostic model to obtain information related to the presence or absence of ophthalmic diseases in a subject and a second diagnostic model to obtain information related to the presence or absence of systemic diseases in a subject, based on retinal images, and use the learned first and second diagnostic models to output diagnostic information about the presence or absence of ophthalmic diseases and systemic diseases in the subject.

[0279] Multiple diagnostic models can be learned in parallel and / or independently. By learning models in this way through multiple diagnostic models predicting different labels, the prediction accuracy for each label can be improved, and the efficiency of the prediction operation can be increased. Since the matters described above can be applied to the learning of multiple diagnostic models, detailed explanations are omitted.

[0280] 1.4.1.2. Parallel Diagnostic Process

[0281] According to one embodiment of the present invention, a diagnostic process for acquiring multiple diagnostic information can be provided. This diagnostic process for acquiring multiple diagnostic information can be implemented as a parallel diagnostic process comprising multiple independent sub-diagnostic processes.

[0282] According to one embodiment of the present invention, the diagnostic process can be executed by multiple diagnostic modules. Each diagnostic process can be executed independently.

[0283] Figure 12 This is a block diagram illustrating a diagnostic section according to an embodiment of the present invention.

[0284] Reference Figure 12 According to one embodiment of the present invention, the diagnostic unit 200 may include a diagnostic request acquisition module 211, a data processing module 231, a first diagnostic module 251, a second diagnostic module 253, and an output module 271. Each module of the diagnostic unit 200, unless otherwise specified, may be connected to... Figure 10 The diagnostic module of the diagnostic department shown in the image operates similarly.

[0285] exist Figure 12 In this embodiment, even though the diagnostic unit 200 includes multiple diagnostic modules, and the diagnostic request acquisition module 211, data processing module 231, and output module 271 are shown as common, the present invention is not limited to this configuration. Multiple diagnostic request acquisition modules, data processing modules, and / or output modules can also be configured. Multiple diagnostic request acquisition modules, data processing modules, and / or output modules can also operate in parallel.

[0286] For example, the diagnostic unit 200 may include a first data processing module that performs a first processing on the input diagnostic target image and a second processing module that performs a second data processing on the diagnostic target image. The first diagnostic module can obtain first diagnostic information based on the first-processed diagnostic target image, and the second diagnostic module can obtain second diagnostic information based on the second-processed diagnostic target image. The first processing and / or the second processing may be any one of image resizing, image color modulation, applying a blur filter, vascular emphasis processing, red light removal conversion, partial region cropping, and partial element extraction.

[0287] Multiple diagnostic modules can acquire different diagnostic information. These modules can utilize different diagnostic models to obtain diagnostic information. For example, the first diagnostic module can use a first diagnostic model to predict whether a subject has an ophthalmological disease to acquire first diagnostic information related to the presence or absence of such a disease; the second diagnostic module can use a second diagnostic model to predict whether a subject has a systemic disease to acquire second diagnostic information related to the presence or absence of such a disease.

[0288] More specifically, the first diagnostic module can use a first diagnostic model based on retinal images to predict whether the subject has diabetic retinopathy to obtain first diagnostic information about whether the subject has diabetic retinopathy, and the second diagnostic module can use a second diagnostic model based on retinal images to predict whether the subject has hypertension to obtain second diagnostic information related to whether the subject has hypertension.

[0289] Furthermore, a diagnostic process according to an embodiment of the present invention may include multiple sub-diagnostic processes. Each sub-diagnostic process may be executed using a different diagnostic model. Each sub-diagnostic process may be executed within a different diagnostic process. For example, a first diagnostic module may execute a first sub-diagnostic process for obtaining first diagnostic information using a first diagnostic model. Alternatively, a second diagnostic module may execute a second sub-diagnostic process for obtaining second diagnostic information using a second diagnostic model.

[0290] Multiple learned diagnostic models can receive diagnostic target data as input and output predicted labels or probabilities. Each diagnostic model can be configured as a classifier, classifying the input diagnostic target data against predetermined labels. In this case, multiple diagnostic models can be configured as classifiers that learn for different characteristics.

[0291] On the other hand, maps can be obtained from each diagnostic model. Maps can be selectively obtained. Maps can be extracted when predetermined conditions are met. For example, if the first diagnostic information indicates that the subject is abnormal for a first characteristic, a first map can be obtained from the first diagnostic model.

[0292] Figure 13 This is a diagram illustrating a diagnostic process according to an embodiment of the present invention.

[0293] Reference Figure 13 According to one embodiment of the present invention, the diagnostic process may include: acquiring diagnostic target data S51, and acquiring diagnostic information corresponding to the diagnostic target data using a first diagnostic model and a second diagnostic model (S51a, S51b). The diagnostic target data may be processed data.

[0294] A diagnostic process according to an embodiment of the present invention may include: obtaining first diagnostic information through a learned first diagnostic model, and obtaining second diagnostic information through a learned second diagnostic model. The first diagnostic model and the second diagnostic model may obtain the first diagnostic information and the second diagnostic information respectively based on the same diagnostic target data.

[0295] For example, the first diagnostic model and the second diagnostic model can obtain first diagnostic information about the subject's ophthalmological disease and second diagnostic information about the presence or absence of the subject's kidney disease, respectively, based on the retinal image of the diagnostic target.

[0296] In addition, unless otherwise specified, with Figure 13 The relevant diagnostic process can be referred to above. Figure 11 The diagnostic process described is implemented similarly.

[0297] 1.4.1.3. Output of Diagnostic Information

[0298] According to one embodiment of the present invention, diagnostic information obtained through a parallel diagnostic process can be acquired. The acquired diagnostic information can be stored in a diagnostic device, a server device, and / or a client device. The acquired diagnostic information can be transmitted to an external device.

[0299] Multiple diagnostic information entries can each indicate multiple labels predicted by multiple diagnostic models. Multiple diagnostic information entries can each correspond to multiple labels predicted by multiple diagnostic models. Alternatively, diagnostic information can be information determined based on multiple labels predicted by multiple diagnostic models. Diagnostic information can correspond to multiple labels predicted by multiple diagnostic models.

[0300] In other words, the first diagnostic information can be diagnostic information corresponding to a first label predicted by a first diagnostic model. Alternatively, the first diagnostic information can be diagnostic information determined by combining the first label predicted by the first diagnostic model and the second label predicted by the second diagnostic model.

[0301] On the other hand, it can output images of maps obtained from multiple diagnostic models. Map images can be output when predetermined conditions are met. For example, if either the first diagnostic information indicates that the subject is abnormal regarding a first characteristic, or the second diagnostic information indicates that the subject is abnormal regarding a second characteristic, a map image obtained from the diagnostic model that outputs diagnostic information indicating the abnormality can be output.

[0302] Multiple diagnostic information items and / or map images can be provided to the user. These diagnostic information items can be provided to the user through the output of the diagnostic device or client device.

[0303] According to one embodiment of the present invention, the diagnostic information corresponding to the diagnostic target image may include grading information. The grading information can be selected from multiple gradations. The grading information can be determined based on multiple diagnostic information and / or seen information obtained through a diagnostic model. The grading information can be determined by considering the applicability or quality information of the diagnostic target image. The grading information can be determined by considering the category in which the diagnostic target image is classified by multiple diagnostic models. The grading information can be determined by considering the numerical values ​​output from multiple diagnostic models.

[0304] For example, the diagnostic information corresponding to the target image may include any level of information selected from either first-level or second-level information. If at least one abnormality observation or abnormality diagnosis information is obtained from the diagnostic information acquired through multiple diagnostic models, the first-level information may be selected. If no abnormality observation or abnormality diagnosis information is obtained from the diagnostic information acquired through the diagnostic models, the second-level information may be selected.

[0305] If at least one of the values ​​obtained through the diagnostic model exceeds a baseline value, the grading information can be selected as first-level information. If none of the obtained values ​​meet the baseline value, the grading information can be selected as second-level information. First-level information indicates a stronger abnormality in the diagnostic target image compared to second-level information.

[0306] If image analysis or a diagnostic model determines that the quality of the target image is below a baseline, the grading information can be selected as third-grade information. Alternatively, the diagnostic information may include third-grade information as well as first or second-grade information.

[0307] Furthermore, a diagnostic system according to an embodiment of the present invention may include a retinal image acquisition unit, a first processing unit, a second processing unit, a third processing unit, and a diagnostic information output unit.

[0308] According to one embodiment of the present invention, a diagnostic system may include a diagnostic device. The diagnostic device may include a retinal image acquisition unit, a first processing unit, a second processing unit, a third processing unit, and / or a diagnostic information output unit. However, the present invention is not limited thereto, and the various units included in the diagnostic system may be located at appropriate positions on a learning device, a diagnostic device, a learning diagnostic server, and / or a client device. For convenience, the following description will be based on the case where the diagnostic device of the diagnostic system includes a retinal image acquisition unit, a first processing unit, a second processing unit, a third processing unit, and a diagnostic information output unit.

[0309] Figure 14 This is a diagram illustrating a diagnostic system according to an embodiment of the present invention. (Refer to...) Figure 14 The diagnostic system includes a diagnostic device, which may include a retinal image acquisition unit, a first processing unit, a second processing unit, a third processing unit, and a diagnostic information output unit.

[0310] According to one embodiment of the present invention, a diagnostic system for assisting in the diagnosis of multiple diseases, based on retinal images, may include: a retinal image acquisition unit for acquiring a target retinal image as the basis for acquiring diagnostic information about a subject; a first processing unit for acquiring a first result about the subject using a first diagnostic model (the first diagnostic model performs machine learning based on a first set of retinal images) on the target retinal image; a second processing unit for acquiring a second result about the subject using a second diagnostic model (the second diagnostic model performs machine learning based on a second set of retinal images that is at least partially different from the first set of retinal images) on the target retinal image; a third processing unit for determining diagnostic information about the subject based on the first and second results; and a diagnostic information output unit for providing the determined diagnostic information to a user.

[0311] The third processing unit can combine the first and second results to determine whether the diagnostic information corresponding to the target retinal image is normal or abnormal.

[0312] The third processing unit can improve diagnostic accuracy by prioritizing abnormal results to determine diagnostic information about the subject.

[0313] When both the first and second results are normal, the third processing unit can determine the diagnostic information as normal; when either the first or second result is abnormal, the diagnostic information can be determined as abnormal.

[0314] The first and second results can be for the same disease or for different diseases.

[0315] At least one map related to the first / second result can be obtained through the first processing unit and / or the first / second diagnostic model, and the diagnostic information output unit can output an image of at least one map.

[0316] When the diagnostic information acquired by the third processing unit is abnormal diagnostic information, the diagnostic information output unit can output at least one map image.

[0317] The diagnostic system may further include a fourth processing unit that acquires quality information of a target retinal image, and a diagnostic information output unit that can output the quality information of the target retinal image acquired by the fourth processing unit.

[0318] If the fourth processing unit determines that the quality information of the target retinal image is lower than a predetermined quality level, the diagnostic information output unit can provide the user with the determined diagnostic information and information indicating that the quality information of the target retinal image is lower than the predetermined quality level.

[0319] 1.4.2.1. Composition of a Serial Diagnostic System

[0320] According to one embodiment, a serial diagnostic system comprising multiple diagnostic models connected in a serial manner can be provided. Examples of serial-type diagnostic models will be given below.

[0321] Figure 15 This is a diagram illustrating a serial diagnostic model according to one embodiment. (See reference...) Figure 15 The diagnostic model 1000 may include a first serial model 1100 and a second serial model 1200.

[0322] The first diagnostic model 1100 can acquire input data including retinal images and acquire a first output (or intermediate output).

[0323] The first diagnostic model 1100 can acquire input data including retinal images and / or other medical diagnostic images and / or non-visual diagnostic data. Input data may include retinal images, OCT images, iris images, ophthalmic angiography images, lung CT images, cardiac CT images, lung X-ray images, cardiac X-ray images, kidney X-ray images, other computed tomographic images, MRI images, or X-ray images. Input data may include information such as the subject's age, height, sex, smoking status, and family history.

[0324] The first output can be a value obtained from the output layer of the first diagnostic model 1100. For example, if the first diagnostic model 1100 is a classifier model, the first output can include the output values ​​at multiple nodes of the output layer of the first diagnostic model 1100. As another example, if the first diagnostic model 1100 is a regression model, the first output can include the numerical values ​​obtained by the first diagnostic model 1100.

[0325] The first output can be a value provided by a portion of the layers of the first diagnostic model 1100. For example, the first output can be a value obtained based on the values ​​of the output layer of the first diagnostic model 1100. Alternatively, the first output can also be a value obtained based on the values ​​of the hidden layers of the first diagnostic model 1100.

[0326] According to one embodiment, the first output may be a value obtained by an activation function in the output layer of the first diagnostic model 1100. If the output layer of the first diagnostic model 1100 includes multiple nodes (or neurons), the first output may include the respective output values ​​corresponding to the multiple nodes or values ​​obtained by a predetermined function (e.g., summation) based on the respective output values.

[0327] The activation function can be any one of the following: Sigmoid function, hyperbolic tangent function, Rectified Linear Unit (ReLU) function, PReLU, Leaky ReLU function, Identity Function, Exponential Linear Unit (ELU) function, or Maxout function.

[0328] The first output can be a feature map or feature value related to the target disease. The first output can be a probability map, significance map, heatmap, etc., related to the target disease. The second diagnostic model 1200 can be set to obtain diagnostic information based on the feature map or feature value related to the target disease.

[0329] The first output can be a probabilistic phenotype related to the target disease. For example, if the target disease is coronary artery disease, and the diagnostic information is numerical information related to the target coronary artery disease, then the first output can be the probability that the subject has the target coronary artery disease based on the eye image. The second diagnostic model 1200 can be set to obtain the diagnostic information of the target disease based on the probabilistic expression related to the target disease.

[0330] The second diagnostic model 1200 can obtain a second output (or diagnostic information) based on the first output. The second diagnostic model 1200 can be a diagnostic model 1000 that has been learned to obtain a second output with the first output as input. The second output can be diagnostic information in various forms as described in this specification.

[0331] Figure 16 This is a diagram illustrating a serial diagnostic model according to another embodiment. (Refer to...) Figure 16 The diagnostic model 1000 may include a first diagnostic model 1100 and a second diagnostic model 1200.

[0332] The first diagnostic model 1100 can acquire input data including an eye image and obtain a first output (intermediate output). The second diagnostic model 1200 can obtain diagnostic information based on the first output and the second input.

[0333] The second input can be the same input data as the first input. For example, the first diagnostic model 1100 can obtain a first output based on an eye image, and the second diagnostic model 1200 can obtain diagnostic information based on the first output and the eye image.

[0334] The second input can be input data obtained based on the first input. For example, the second input can be eye image data obtained after performing image processing. For example, the second input can be a black and white processed eye image, an eye image with blood vessels emphasized, a blood vessel image extracted from an eye image, or an eye image with blood vessels removed.

[0335] The second input can be input data that is at least partially different from the first input.

[0336] The second input can be image data different from the first input. The second input may include retinal images, OCT images, iris images, ophthalmic angiography images, lung CT images, lung CT images, cardiac CT images, lung X-ray images, cardiac X-ray images, kidney X-ray images, other tomographic images, MRI images, or X-ray images.

[0337] The second input may include non-visual information about the subject. For example, the first diagnostic model 1100 may obtain a first output about the target disease based on eye images, and the second diagnostic model 1200 may obtain diagnostic information based on the first output and the second input (e.g., the subject's physical information (e.g., age, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), cholesterol value, family history, etc.)).

[0338] Figure 17 This is a diagram illustrating a serial diagnostic model according to yet another embodiment. (Refer to...) Figure 17 The diagnostic model 1000 may include a first diagnostic model 1100 and a second diagnostic model 1200.

[0339] and Figure 16In comparison, diagnostic model 1000 can further acquire the first diagnostic information obtained by the first diagnostic model 1100. Diagnostic model 1000 can acquire intermediate diagnostic information and secondary diagnostic information based on the intermediate diagnostic information.

[0340] Diagnostic model 1000 can acquire first diagnostic information and second diagnostic information. First diagnostic model 1100 can acquire input data including an eye image and acquire a first output (first diagnostic information or intermediate output). Diagnostic model 1000 can acquire first diagnostic information based on the first output. Second diagnostic model 1200 can acquire second diagnostic information at least partially based on the first output. Diagnostic model 1000 can acquire second diagnostic information at least partially based on the first output, taking into account other information extracted from the eye image.

[0341] The primary and secondary diagnostic information can be diagnostic information for the same target disease. The primary diagnostic information can be diagnostic information that can be obtained from eye images (clinically or through machine learning models), such as the probability expression of the presence or absence of eye diseases, vascular abnormalities, kidney diseases, etc.

[0342] Second diagnostic information can represent more detailed diagnostic information than the first diagnostic information. For example, second diagnostic information may include rating information indicating the degree of risk of the target disease for the same target disease as the first diagnostic information, or score information indicating the score related to the target disease.

[0343] The second diagnostic information may be associated with the first diagnostic information, but may include diagnostic information that takes into account information obtained beyond the images, such as the rate of disease progression, guidance information related to the diagnostic information, etc.

[0344] On the other hand, the primary and secondary diagnostic information can be diagnostic information for different diseases. Alternatively, they can be diagnostic information for different diseases belonging to the same group. For example, the primary diagnostic information could be related to glaucoma (belonging to the ophthalmology group), and the secondary diagnostic information could be related to macular degeneration (belonging to the ophthalmology group). Similarly, the primary diagnostic information could be related to drusen (belonging to the ophthalmology group), and the secondary diagnostic information could be related to diabetic retinopathy (belonging to the ophthalmology group).

[0345] The primary and secondary diagnostic information can be diagnostic information for different diseases belonging to different groups. For example, the primary diagnostic information could be related to macular degeneration or drusen, which belong to the ophthalmology group, while the secondary diagnostic information could be related to hyperlipidemia, which belongs to the cardiovascular group.

[0346] In the above embodiments, the description is based on the case where the diagnostic model 1000 has a first diagnostic model 1100 and a second diagnostic model 1200, but the diagnostic model 1000 may also include a greater number of diagnostic models. Furthermore, each diagnostic model can be connected via the parallel or serial connection described above.

[0347] 1.4.2.2. Diagnosis using a serial diagnostic model

[0348] According to one embodiment of the invention described in this specification, a diagnostic method utilizing a diagnostic model including serially connected sub-models can be provided.

[0349] Figure 18 This is a diagram illustrating a diagnostic method using a diagnostic model according to one embodiment.

[0350] Reference Figure 18 A diagnostic method according to one embodiment may include: acquiring input data S61, acquiring first diagnostic information S62, and acquiring second diagnostic information S63.

[0351] The step of acquiring input data S61 may include acquiring retinal images of the subject. The step of acquiring input data S61 may further include acquiring medical images of body parts other than the subject's eyes. The step of acquiring input data S61 may further include acquiring non-visual information about the subject (e.g., body information of the subject). The step of acquiring input data S61 may further include performing preprocessing on the subject's eye images required to obtain diagnostic information.

[0352] The step of obtaining first diagnostic information S62 may include obtaining first diagnostic information about the subject based on an eye image and through a first diagnostic model. Obtaining the first diagnostic information may involve obtaining diagnostic information related to a first disease. For example, obtaining the first diagnostic information may include obtaining first diagnostic information based on an eye image that represents the probability that the subject's coronary artery calcium score is greater than or equal to 0.

[0353] The step of obtaining second diagnostic information S63 may include obtaining second diagnostic information about the subject based on the first diagnostic information and through the second diagnostic model.

[0354] Obtaining second diagnostic information may include obtaining diagnostic information that is related to the first disease and is different from the first diagnostic information. For example, obtaining first diagnostic information may include obtaining first diagnostic information that indicates the probability that the subject's coronary artery calcium score is greater than or equal to 0, and obtaining second diagnostic information may include, based on the subject's first diagnostic information, obtaining second diagnostic information that indicates a score related to the subject having the target cardiovascular disease (e.g., the probability of cardiovascular disease-related events occurring within 10 years).

[0355] Alternatively, obtaining second diagnostic information may include obtaining diagnostic information related to a second disease that is different from the first disease. For example, obtaining first diagnostic information may include obtaining first diagnostic information indicating that the subject has a target ophthalmological disease, and obtaining second diagnostic information may include obtaining second diagnostic information indicating that the subject has a cardiovascular or cerebrovascular disease.

[0356] The diagnosis using the above diagnostic model will be explained in detail below through specific examples.

[0357] 2. Diagnostic methods for kidney diseases

[0358] 2.1.1. Biomarkers of Kidney Disease

[0359] In one embodiment, kidney disease biomarkers used to predict, measure, and confirm the presence, progression, etc., of kidney disease can be used in kidney disease diagnostic methods. In this specification, kidney disease biomarkers may include kidney disease risk assessment tools.

[0360] For example, biomarkers for kidney disease may include eGFR (Estimated Glomerular Filtration Rate) values, albuminuria values, cystatin C values, KDIGO (Kidney Disease: Improving Global Outcomes) classifications, etc.

[0361] eGFR values ​​can be calculated based on serum creatinine concentration, age, sex, and ethnicity, and are used as an indicator of kidney filtration function. For example, based on eGFR values, the risk of kidney disease can be categorized as low risk (90 mL / min / 1.73 mcg / mL). 2 (or higher), medium risk (60-89 mL / min / 1.73m) 2 Medium to high risk (45-59 mL / min / 1.73m) 2 Extremely high risk (30-44 mL / min / 1.73m) 2 Highest risk (29 mL / min / 1.73m) 2 (or below).

[0362] Albuminuria is the proportion of albumin in urine and is a primary indicator for assessing kidney disease. It can be expressed as the ratio of albumin to creatinine. For example, based on albuminuria levels, the risk of kidney disease can be categorized as low risk (less than 30 mg / g), intermediate risk (30-300 mg / g), and high risk (300 mg / g or higher).

[0363] Cystatin C levels are a biomarker of kidney function measured in the blood. Higher cystatin C levels indicate higher risk for kidney function. For example, based on cystatin C levels, the risk of kidney disease can be categorized into a low-risk group (1.0 mg / L) and a high-risk group (above 1.0 mg / L). Alternatively, in another example, the risk of kidney disease can be categorized into a low-risk group (0.62–1.15 mg / L) and a high-risk group (above 1.15 mg / L) based on cystatin C levels. Furthermore, in yet another example, the risk of kidney disease can be categorized into a low-risk group (0.9 mg / L or below), a low-to-intermediate-risk group (0.9–1.2 mg / L), an intermediate-risk group (1.2–1.9 mg / L), an intermediate-to-high-risk group (1.9–3.0 mg / L), and a high-risk group (above 3.0 mg / L) based on cystatin C levels.

[0364] The KDIGO classification can be a renal function biomarker determined by combining eGFR values ​​(or classifications) and albuminuria values ​​(or classifications). For example, the KDIGO classification can be divided into low-risk, intermediate-risk, and high-risk groups based on eGFR values ​​(or classifications) and albuminuria values ​​(or classifications).

[0365] In addition, various biomarkers for kidney disease can be used in kidney disease diagnostic methods.

[0366] 2.1.2. Diagnostic methods for kidney diseases

[0367] The kidney disease diagnostic method according to one embodiment of this specification can be executed using at least one of the above-described diagnostic model, parallel diagnostic model, or serial diagnostic model. For ease of explanation, the following description will focus on executing the kidney disease diagnostic method using a serial diagnostic model.

[0368] Figure 19 This is a diagram illustrating a method for diagnosing kidney disease according to one embodiment.

[0369] Reference Figure 19 The processor of the diagnostic device may include a step of acquiring retinal images (S100) and a step of acquiring diagnostic information for kidney disease (S200).

[0370] In step S100, the processor of the diagnostic device can acquire a retinal image. Furthermore, according to an embodiment, the processor of the diagnostic device can perform preprocessing, enhancement, serialization, etc., on the acquired retinal image. Since the above can be applied here, detailed descriptions are omitted.

[0371] Furthermore, in step S200, the processor of the diagnostic device can acquire kidney disease diagnostic information. In this specification, kidney disease diagnostic information may be represented as Reti-CKD. The following will refer to... Figure 20 Explain the diagnostic methods for kidney diseases.

[0372] Figure 20 A diagnostic model for obtaining diagnostic information of kidney disease according to one embodiment is described.

[0373] Reference Figure 20 The diagnostic model 1000 may be included in the processor and / or storage module of the diagnostic device. Furthermore, Figures 15 to 18 The description of the diagnostic model can be applied to Diagnostic Model 1000.

[0374] The diagnostic model 1000 may include a first diagnostic model 1100 and a second diagnostic model 1200. Both the first diagnostic model 1100 and the second diagnostic model 1200 can be machine learning models. Furthermore, the first diagnostic model 1100 and the second diagnostic model 1200 can be models based on the same algorithm or models based on different algorithms. For example, the first diagnostic model 1100 and the second diagnostic model 1200 can be neural network models. Alternatively, the first diagnostic model 1100 can be a neural network model, and the second diagnostic model 1200 can be a non-neural network machine learning model. In the following description, for ease of explanation, the case where the first diagnostic model 1100 is a neural network model and the second diagnostic model 1200 is a non-neural network machine learning model will be emphasized, but the description in this specification is not limited to this.

[0375] The processor of the diagnostic device can input retinal images (or preprocessed retinal images, serialized retinal images) into the first diagnostic model 1100. Furthermore, the processor can obtain from the first diagnostic model 1100 the probability value and / or classification of the subject's current high risk of kidney disease based on the input retinal image. For example, the processor can obtain from the first diagnostic model 1100 that the subject's eGFR value in the current retinal image is less than or equal to 60 mL / min / 1.73 m 2 The probability value and / or grading of the probability of the presence of albuminuria in the current subject's urine (or the probability of an albuminuria value greater than or equal to 300 mg / g). Here, the probability value can be a number between 0 and 1.

[0376] Specifically, the first diagnostic model 1100 can be a neural network model with a CNN structure.

[0377] In one embodiment, during the training process of the first diagnostic model 1100, the target retinal image for training may include retinal images of patients with kidney disease and retinal images of normal individuals who are not patients with kidney disease.

[0378] Furthermore, in the training process of the first diagnostic model 1100, each monocular image of the target retinal image (left and right eyes) can be trained independently. During the training process, the ground truth can be the presence or absence of kidney disease. Additionally, predetermined information can be labeled on each retinal image. For example, each retinal image can be labeled with whether the eGFR value is less than or equal to 60 mL / min / 1.73 m 2 And / or the presence of albuminuria in the subject's urine (e.g., whether the albuminuria value is greater than or equal to 300 mg / g), a binary variable. For example, if the eGFR value is less than or equal to 60 mL / min / 1.73 m 2 If albuminuria is present in the patient's urine, the patient in the retinal image is marked as having kidney disease; if the eGFR value is greater than 60 mL / min / 1.73 m 2 Furthermore, if the subject's urine does not contain albuminuria, the subject in that retinal image can be labeled as not having kidney disease. Of course, the eGFR and albuminuria values ​​in the above example can be modified according to the embodiments. Additionally, for example, the subject's eGFR and albuminuria values ​​can be labeled on each retinal image.

[0379] Furthermore, during the evaluation of the first diagnostic model 1100, probability values ​​were assigned to the left and right retinal images, and the average of the probability values ​​can be regarded as the output of the test results.

[0380] Furthermore, the last fully connected layer of the first diagnostic model 1100 can perform a one-logit probabilistic prediction. The logit from the last fully connected layer can be converted into a probability using a sigmoid function. Additionally, the first diagnostic model 1100 can be trained to minimize the loss between the target and the prediction. Furthermore, in one embodiment, the first diagnostic model 1100 can be trained for 50 epochs using a learning rate of 0.0002, a cosine learning rate schedule, leveraging the AdamW Optimizer.

[0381] Furthermore, for data augmentation of the target retinal image during training, at least one of mixup, cutmix, randaugment, an enhancing contrast module, and random cropping can be used. Additionally, focal loss and exponential moving average are used during the training of the first diagnostic model 1100, and the size of the training target retinal image can be set to 384x384.

[0382] Furthermore, the processor of the diagnostic device can input the output value of the first diagnostic model 1100 and the subject's physical information (at least one of age, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), and cholesterol value) into the second diagnostic model 1200. The processor can then obtain from the second diagnostic model 1200 a probability value and / or classification of the probability of kidney disease-related events occurring within 5 years for the subject. Here, kidney disease-related events within 5 years can include fatal kidney disease, non-fatal kidney disease, and various events caused by kidney disease (hospitalization, surgery, procedures, death, etc.) occurring within 5 years from the time the retinal image was taken. Additionally, in some cases, the processor can obtain the probability value of the probability of kidney disease-related events occurring within 5 years for the subject from the second diagnostic model 1200 and apply a predetermined cutoff value to the obtained probability value to obtain a classification corresponding to the obtained probability value. Furthermore, although this specification focuses on kidney disease-related events within 5 years, it is not limited to this. According to the embodiments, kidney disease-related events within various periods such as 3 years and 10 years can of course be applied to the description of this specification.

[0383] Therefore, the processor of the diagnostic device can output a probability value (score) and / or grade regarding the probability of kidney disease-related events occurring in the subject within 5 years as diagnostic information for kidney disease.

[0384] Furthermore, according to an embodiment, the processor of the diagnostic device can obtain a probability value and / or classification of the current kidney disease risk from the second diagnostic model 1200. Additionally, according to an embodiment, the processor of the diagnostic device can obtain a probability value and / or classification of the probability of kidney disease-related events occurring within the subject over a 5-year period from the second diagnostic model 1200, and based on the obtained probability value and / or classification of the probability of kidney disease-related events occurring within the subject over a 5-year period, obtain a probability value and / or classification of the current kidney disease risk. For example, the processor of the diagnostic device can compare the probability value and / or classification of the probability of kidney disease-related events occurring within the subject over a 5-year period obtained from the second diagnostic model 1200 with at least one predetermined baseline probability value and / or classification, and based on the comparison result, obtain a probability value and / or classification of the current kidney disease risk.

[0385] For example, a high risk corresponding to an existing biomarker can be predefined (e.g., an eGFR value less than or equal to 44 mL / min / 1.73 m). 2The diagnostic device's processor can then set the first score and / or first grade of kidney disease diagnostic information (albuminuria value greater than or equal to 300 mg / g and / or KDIGO classification of high risk group or higher). When the probability value and / or grade of the probability of kidney disease-related events occurring within 5 years of the subject, obtained from the second diagnostic model 1200, is greater than or equal to a predetermined first threshold and / or first threshold grade, the processor of the diagnostic device can determine that the current risk of kidney disease is high. Furthermore, although one threshold score and / or threshold grade is described as one in the above example, it is not limited to this; multiple or more threshold scores and / or threshold grades can be set. For example, the threshold score and / or threshold grade may include a first threshold and / or first threshold grade corresponding to high risk of the existing biomarker, and a second threshold and / or second threshold grade corresponding to intermediate risk of the existing biomarker.

[0386] Furthermore, in one embodiment, the second diagnostic model 1200 can be trained based on training target data. Here, the training of the second diagnostic model 1200 can include the meaning of the fitting of the second diagnostic model 1200. Additionally, the second diagnostic model 1200 can be constructed using a Cox proportional hazards model based on linear regression.

[0387] For example, the training target data may include: the physical information of the subject in the training target retinal image of the first diagnostic model 1100, the tracking observation results of kidney disease events of the subject within 5 years, and the probability value of the first diagnostic model 1100 corresponding to the training target retinal image. Furthermore, according to an embodiment, the subject in the training target retinal image of the first diagnostic model 1100 and the subject in the training target data of the second diagnostic model 1200 may be the same or at least partially different. For example, in the training target data of the subject in the training target retinal image of the first diagnostic model 1100, the training target data of subjects who have experienced kidney disease events or have kidney disease at the baseline time point can be excluded from the training of the second diagnostic model 1200. This is because by using the training target data of normal individuals who have not experienced kidney disease events and do not have kidney disease to train the second diagnostic model 1200, the accuracy of the probability of kidney disease events occurring within 5 years output by the second diagnostic model 1200 can be improved.

[0388] Furthermore, in one embodiment, during the training process of the second diagnostic model 1200, the coefficients of the covariates of the Cox proportional hazards model can be set by fitting the Cox proportional hazards model to the UK Biobank cohort. In the UK Biobank cohort, the 5-year survival probability of kidney disease can be 0.9980896. The second diagnostic model 1200 can then be modeled as the 5-year failure probability. Example 1 uses the UK Biobank cohort as an example, but this specification is not limited to this; the second diagnostic model can be fitted or calibrated based on cohorts other than the UK Biobank cohort.

[0389] Furthermore, in one embodiment, the second diagnostic model 1200 may include age, sex, presence of hypertension, presence of diabetes, and probability values ​​of current kidney disease risk output from the first diagnostic model 1100 as variables. Since these variables can be obtained from the subject using non-invasive methods, they can be used in the second diagnostic model 1200.

[0390] For example, the second diagnostic model 1200 can obtain the probability value of the occurrence of kidney disease-related events in the subject within 5 years based on the following mathematical formula 1. For example, the Cox proportional hazards model of the second diagnostic model 1200 can be constructed based on mathematical formula 1 and mathematical formula 2.

[0391] [Mathematical Expression 1]

[0392] Predicted risk = 1 - SP^(Exp[LP])

[0393] Here, Predicted risk is the probability value of the occurrence of kidney disease-related events within 5 years, and SP (Survival Probability) is the probability of survival from kidney disease within 5 years, which could be 0.9980896 in the UK Biobank example above. Furthermore, LP (Linear Predictor) will be explained in the following mathematical formula 2.

[0394] [Mathematical Expression 2]

[0395] LP = a1*RPS*100 + a2*Age + a3*Female + a4*Hypertension + a5*Diabetes

[0396] Here, RPS (Retinal photograph-based Prediction Score) is the probability value of the current risk of kidney disease output from the first diagnostic model 1100, where Age is the age and Female indicates the subject's sex. For example, Female can be an indicator function, reflecting a value of 1 when the subject is female and 0 when the subject is male. Hypertension reflects a value of 1 when the subject has hypertension and 0 when the subject does not. For example, for a user taking antihypertensive medication, it could be set to have hypertension. Furthermore, Diabetes reflects a value of 1 when the subject has diabetes (or prediabetes) and 0 when the subject does not have diabetes (or prediabetes).

[0397] Furthermore, a1 to a5 are the coefficients of the covariates of each variable. For example, through fitting the second diagnostic model 1200, a1 to a5 can be set to 0.0546351, 0.0357466, -0.3317242, -0.1207903, and 0.4679712, respectively. In addition, the processor of the diagnostic device can, based on (or applying) the mathematical formulas 1 and 2, provide the risk of kidney disease over 3, 5, or 10 years. For example, the processor of the diagnostic device can add or replace factors such as race, smoking status, and cholesterol levels in the covariates of mathematical formula 2, and set the coefficients of each covariate through fitting the second diagnostic model 1200.

[0398] Furthermore, the processor of the diagnostic device can obtain probability values ​​of the occurrence probability of kidney disease-related events within 5 years from the second diagnostic model 1200, and apply predetermined cutoff values ​​to the probability values ​​to obtain a classification corresponding to the probability values. Of course, in some cases, predetermined cutoff values ​​can also be applied to the second diagnostic model 1200 in advance to obtain a classification corresponding to the probability values ​​from the second diagnostic model 1200. In this case, there can be one or more cutoff values. For example, if there is one cutoff value, there can be two classifications corresponding to the probability values; if there are two cutoff values, there can be three classifications corresponding to the probability values. This specification applies to various numbers and types of cutoff values.

[0399] In one embodiment, the processor of the diagnostic device can set predetermined cutoff values ​​based on the population distribution (i.e., incidence rate) of each group, which is divided into multiple tiers, according to the follow-up observation results of a predetermined population. For example, the cutoff values ​​can be determined based on the results of follow-up observations of the predetermined population. For instance, if, based on the entire follow-up observation subjects, the follow-up observation results show that 0-n1% is a low-risk group, n1-n2% is a medium-risk group, n2-n3% is a high-risk group, and n3-100% is a highest-risk group, then n1%, n2%, and n3% can be set as cutoff values ​​respectively. Furthermore, even if n1%, n2%, and n3% are not cutoff values, cutoff values ​​can still be set so that the population distribution of each risk group in the kidney disease diagnostic information is similar to the follow-up observation results of the predetermined population.

[0400] For example, if the target population is the UK Biobank cohort and the Korean Diabetic cohort, cutoff values ​​can be set based on the population distribution of low-risk, intermediate-risk, high-risk, and highest-risk groups in the UK Biobank and Korean Diabetic cohorts, so that the population distribution of each risk group in the kidney disease diagnosis information is similar to that of each risk group in the UK Biobank and Korean Diabetic cohorts.

[0401] Furthermore, the diagnostic device's processor can set cutoff values ​​based on the incidence of kidney disease biomarkers. For example, for eGFR, the risk of kidney disease can be categorized as low risk (90 mL / min / 1.73 mcg) based on the eGFR value. 2 (or higher), medium risk (60-89 mL / min / 1.73m) 2 Medium to high risk (45-59 mL / min / 1.73m) 2 Extremely high risk (30-44 mL / min / 1.73m) 2 Highest risk (29 mL / min / 1.73m) 2 (or below). Furthermore, if the incidence rate at the eGFR value, corresponding to the proportions of the population at low risk, intermediate risk, medium-high risk, very high risk, and highest risk, are x1%, x2%, x3%, x4%, and x5%, respectively, then a cutoff value can be set so that the proportions of the population included in each grade of the kidney disease diagnostic information output by the processor of the diagnostic device match the stated x1%, x2%, x3%, x4%, and x5%. This can be applied not only to eGFR values ​​but also to biomarkers such as albuminuria values, cystatin C values, and KDIGO classifications.

[0402] In addition, the incidence rates based on the Kidney Failure Risk Equation and the Risk Prediction Equation can also be used to set the above cutoff values.

[0403] In addition, the cutoff value can be an optimized value to most effectively detect groups corresponding to high risk for each biomarker.

[0404] Specifically, the processor of the diagnostic device can set a cutoff value based on the proportion of high-risk and low-risk groups classified by the 5-year kidney risk calculator, according to ethnicity or applicable clinical guidelines for calculating 5-year kidney risk. This improves the predictive performance of kidney disease diagnostic information for high-risk groups. For example, to maximize the diagnostic performance of kidney disease diagnostic information for high-risk groups within eGFR values, a cutoff value can be set such that the proportions of all subjects in the low-risk and high-risk groups of the kidney disease diagnostic information are similar to the proportions of subjects in the low-risk and high-risk groups within the eGFR values. Alternatively, the processor can set a cutoff value such that the proportions of all subjects in the low-risk and high-risk groups of the kidney disease diagnostic information are within a predetermined range of the proportions of subjects in the low-risk and high-risk groups within the eGFR values. In this case, the processor of the diagnostic device can accurately identify subjects corresponding to the low-risk and high-risk groups within the eGFR values.

[0405] Furthermore, the biomarkers used to train the first diagnostic model 1100 and the biomarkers used to set the cutoff values ​​for the second diagnostic model 1200 can be the same or different.

[0406] Furthermore, when cutoff values ​​are set based on each biomarker, the grading of kidney disease diagnostic information output by the processor of the diagnostic device can be similar to the actual results of the biomarkers for which the cutoff values ​​are set. For example, if the cutoff value corresponding to the eGFR value biomarker is applied to the probability value of the occurrence of kidney disease-related events within 5 years output from the second diagnostic model 1200, and the processor of the diagnostic device outputs the grade of the subject, then the grade of the subject output by the processor of the diagnostic device can match the grade determined after the subject actually undergoes eGFR-related examinations. If examinations such as eGFR, albuminuria, and cystatin C are actually performed, the subject may experience inconvenience due to blood or urine tests. However, according to the kidney disease diagnostic method of this specification, since highly accurate kidney disease diagnostic information can be obtained using only the subject's retinal images, user convenience can be improved.

[0407] Furthermore, in one embodiment, when kidney disease diagnostic information is applied to the scores and / or classifications of existing biomarkers, the incidence of kidney disease-related events corresponding to the scores and / or classifications of the existing biomarkers can be differentiated or stratified according to the groups of kidney disease diagnostic information (e.g., low-risk, intermediate-risk, high-risk, and highest-risk groups). Moreover, the cutoff value for the kidney disease diagnostic information can be set such that, when the kidney disease diagnostic information is applied to the scores and / or classifications of existing biomarkers, the incidence of kidney disease-related events corresponding to the scores and / or classifications of the existing biomarkers is explicitly stratified according to the groups of the kidney disease diagnostic information.

[0408] The following will describe in detail embodiments of the kidney disease diagnosis method according to this specification.

[0409] 2.1.3. Example 1

[0410] Example 1 illustrates experimental results of a kidney disease diagnostic method according to this specification, using clinical data and retinal images from the UK Biobank cohort and the Korean diabetes cohort.

[0411] First, in Example 1, the first diagnostic model 1100 can be based on 158,216 retinal images (79,108 people) and clinical data from a health checkup center in South Korea (whether the eGFR value of each retinal image is less than or equal to 60 mL / min / 1.73 m). 2 Training is conducted to determine whether albuminuria is present in the urine of the examinee.

[0412] Furthermore, in Example 1, the second diagnostic model 1200 can be trained (or fitted) using the UK Biobank cohort (30,477 participants) and the Korean diabetes cohort (5,014 participants). Participants with chronic kidney disease (CKD) and eGFR values ​​below 90 mL / min / 1.73 m... 2 Data from participants who tested positive for albuminuria (or whose albuminuria levels were greater than or equal to 30 mg / Cr) can be excluded from the training of the second diagnostic model 1200. In this case, the processor of the diagnostic device can provide a probability value and / or classification with high accuracy for the probability of kidney disease-related events occurring within 5 years in healthy individuals who do not currently have kidney disease.

[0413] besides, Figure 20 The content described herein can be applied to the first diagnostic model 1100 and the second diagnostic model 1200 in Example 1.

[0414] The results of Example 1 will be described below. In the figures below, the Reti-CKD score, as kidney disease diagnostic information, represents the probability value of the occurrence of kidney disease-related events within 5 years, and each level of the Reti-CKD score can be used as kidney disease diagnostic information to represent the level corresponding to the respective probability value.

[0415] Figure 21 The clinical characteristics of the subjects corresponding to the kidney disease diagnostic information according to Example 1 are shown.

[0416] Reference Figure 21 The health screening data from South Korea used to train the first diagnostic model 1100 consisted of data from 79,108 individuals with a mean age of 49.5 years (standard deviation (SD): 11.8) and a mean eGFR of 100.3 mL / min / 1.73 m³. 2 Furthermore, the UK Biobank cohort data used to train the second diagnostic model 1200 comprised data from 30,477 participants, of whom 720 (2.4%) were diagnosed with chronic kidney disease during a mean follow-up period of 10.8 years (interquartile range (IQR): 10.7–11.0). Additionally, the Korean diabetes cohort data used to train the second diagnostic model 1200 comprised data from 5,014 participants, of whom 206 (4.1%) were diagnosed with chronic kidney disease during a mean follow-up period of 6.1 years (interquartile range (IQR): 4.0–8.4). Furthermore, the mean eGFR value in the UK Biobank cohort was 99.4 (standard deviation (SD): 6.6) mL / min / 1.73 m 2 The mean eGFR value in the Korean diabetes cohort was 102.5 mL / min / 1.73 m (standard deviation (SD): 9.1). 2 .

[0417] Figure 22a and Figure 22b This is a graph used to illustrate the incidence of kidney disease-related events based on kidney disease diagnostic information according to Example 1.

[0418] Reference Figure 22a and Figure 22b ,exist Figure 22a and Figure 22b In the chart, the x-axis represents time (years), and the y-axis represents the incidence of chronic kidney disease-related events.

[0419] in addition, Figure 22aKaplan-Meier curves are shown for the results of Kaplan-Meier survival analysis performed on four groups of subjects in the UK Biobank cohort, divided into four grades based on kidney disease diagnostic information. Figure 22b Kaplan-Meier curves can be shown for the results of Kaplan-Meier survival analysis performed on four groups of subjects in a Korean diabetes cohort, divided into four grades based on kidney disease diagnostic information. Figure 22a In the UK Biobank cohort, 321,317 person-years were surveyed over an average follow-up period of 10.8 years. Figure 22b In the Korean diabetes cohort, 30,122 person-years were investigated over a mean follow-up period of 6.1 years.

[0420] like Figure 22a and Figure 22b As shown, kidney disease diagnostic information can be based on four groups, clearly distinguishing kidney disease risk in the UK Biobank cohort and the Korean diabetes cohort. Figure 22a and Figure 22b In China, the hazard ratio (HR) for the development of chronic kidney disease can show a dose-dependent association based on four levels. The adjusted HRs per 1 standard deviation increment of kidney disease diagnostic information are... Figure 22a In the UK Biobank cohort, the value could be 1.34 (95% confidence interval: 1.27–1.41). Figure 22b In the Korean diabetes cohort, the value could be 1.94 (95% confidence interval: 1.63–2.31).

[0421] Furthermore, kidney disease diagnostic information can even clearly distinguish kidney disease risk with significant risk ratios within subgroups defined by gender, age, presence or absence of hypertension, and presence or absence of diabetes. For example, in Figure 22aIn the UK Biobank cohort, the trends in hazard ratios for chronic kidney disease were as follows: overall 1.52 (95% confidence interval: 1.42–1.63), males 1.47 (95% confidence interval: 1.31–1.65), females 1.51 (95% confidence interval: 1.36–1.67), under 55 years of age 1.35 (95% confidence interval: 1.20–1.51), over 55 years of age 1.46 (95% confidence interval: 1.29–1.65), and without hypertension 1.59. (95% confidence interval: 1.47-1.74), with hypertension: 1.32 (95% confidence interval: 1.12-1.56), without diabetes: 1.50 (95% confidence interval: 1.39-1.61), with diabetes: 1.35 (95% confidence interval: 1.07-1.72), eGFR greater than 100: 1.53 (95% confidence interval: 1.34-1.73), eGFR less than or equal to 100: 1.41 (95% confidence interval: 1.30-1.53). Furthermore, in the Korean diabetes cohort in (b), the trends in hazard ratios for chronic kidney disease could be: overall 1.99 (95% confidence interval: 1.72–2.31), males 2.17 (95% confidence interval: 1.73–2.70), females 1.85 (95% confidence interval: 1.51–2.28), under 55 years of age 1.75 (95% confidence interval: 1.28–2.39), and over 55 years of age 1. 0.97 (95% confidence interval: 1.58-2.46), no hypertension 2.13 (95% confidence interval: 1.76-2.57), hypertension 1.71 (95% confidence interval: 1.37-2.12), eGFR greater than 100 2.10 (95% confidence interval: 1.62-2.72), eGFR less than or equal to 100 1.49 (95% confidence interval: 1.28-1.78).

[0422] Figure 23 This is a graph used to illustrate the predictive performance of kidney disease-related events over 5 years based on kidney disease diagnostic information from Example 1.

[0423] Reference Figure 23 , Figure 23 The table compares the performance of the kidney disease diagnostic information corresponding to the Reti-CKD score and the eGFR-based kidney disease score in the UK Biobank cohort and the Korean diabetes cohort. The eGFR-based kidney disease score can be used as a biomarker to assess the risk of existing kidney disease.

[0424] exist Figure 23In the table, the NRI (Net Reclassification Index) can be used to measure how much better the new model is compared to the old model. For the NRI, it is 0.109 in the UK Biobank cohort (95% confidence interval (CI): 0.44–0.156) and 0.179 in the Korean diabetes cohort (95% confidence interval: 0.017–0.292), showing similar results. Furthermore, when comparing scores corresponding to kidney disease diagnostic information and eGFR-based kidney disease scores, the difference in C-statistics is 0.020 in the UK Biobank cohort (95% confidence interval (CI): 0.011–0.029) and 0.024 in the Korean diabetes cohort (95% confidence interval: 0.002–0.046), indicating that scores corresponding to kidney disease diagnostic information show significantly higher performance than eGFR-based kidney disease scores. Therefore, the scores corresponding to kidney disease diagnostic information show superior performance in terms of accuracy when compared with existing eGFR-based kidney disease scores.

[0425] In addition, although Figure 23 The value is not shown in the table, but if a sensitivity analysis is performed, it is not limited to an eGFR value greater than or equal to 90 mL / min / 1.73 m. 2 Instead of focusing on a single object, this study included all objects. When comparing scores corresponding to kidney disease diagnostic information and eGFR-based kidney disease scores, the C-statistic showed a difference of 0.008 (95% confidence interval (CI): 0.001–0.016) in the UK Biobank cohort and 0.057 (95% confidence interval: 0.048–0.067) in the Korean diabetes cohort. Therefore, scores corresponding to kidney disease diagnostic information demonstrated superior performance with higher sensitivity when compared to existing eGFR-based kidney disease scores. Thus, even without eGFR testing via blood tests (e.g., even if the eGFR value is unknown as greater than or equal to 90 mL / min / 1.73 m...), the study found that scores based on kidney disease diagnostic information exhibited significantly higher performance than eGFR-based scores. 2 Kidney disease diagnostic information can also stratify the risk of future kidney disease in the general population and patients with diabetes better than eGFR.

[0426] In addition, although Figure 23 The table is not shown, but the sensitivity analysis can be performed in groups with hypertension and / or diabetes and groups without hypertension and diabetes.

[0427] In the group without hypertension and diabetes, when comparing scores corresponding to kidney disease diagnostic information and eGFR-based kidney disease scores, the difference in C-statistics was 0.025 (95% confidence interval (CI): 0.002–0.048) in the UK Biobank cohort, indicating that scores corresponding to kidney disease diagnostic information showed significantly higher performance than eGFR-based kidney disease scores. Therefore, scores corresponding to kidney disease diagnostic information demonstrate superior performance with higher sensitivity than existing eGFR-based kidney disease scores for individuals with normal renal function without hypertension and diabetes.

[0428] Furthermore, in the group with hypertension and / or diabetes, when comparing scores corresponding to kidney disease diagnostic information and eGFR-based kidney disease scores, the difference in C-statistics was 0.019 (95% confidence interval (CI): 0.008–0.030) in the UK Biobank cohort, with scores corresponding to kidney disease diagnostic information showing significantly higher performance than eGFR-based kidney disease scores. Therefore, scores corresponding to kidney disease diagnostic information demonstrate superior performance with higher sensitivity than existing eGFR-based kidney disease scores, regardless of the presence of hypertension and / or diabetes. This suggests that the method is applicable not only to patients with hypertension and diabetes, the most important risk factors for kidney disease, but also to kidney disease caused by problems with the kidneys themselves, unrelated to hypertension / diabetes.

[0429] 2.1.4. Obtain information on kidney disease risk based on existing biomarkers

[0430] Figure 24 This is a diagram illustrating a method for diagnosing kidney disease according to another embodiment.

[0431] Reference Figure 24 A kidney disease diagnosis method according to another embodiment may include: step S300 of acquiring retinal images and step S400 of acquiring kidney disease risk information based on existing biomarkers as kidney disease diagnosis information.

[0432] In step S300, the processor of the diagnostic device can acquire a retinal image. Because Figure 19 The above description applies here, therefore detailed explanation is omitted.

[0433] Furthermore, in step S400, the processor of the diagnostic device can acquire information on the risk of kidney disease based on existing biomarkers, which serves as diagnostic information for kidney disease. The aforementioned content regarding the diagnostic model can be applied to step S400.

[0434] Here, the information on kidney disease risk based on existing biomarkers refers to the kidney disease risk calculated based on existing biomarkers. The processor of the diagnostic device can acquire this information using retinal images. For example, according to step S400, the processor of the diagnostic device can, based on the retinal images, use a diagnostic model to acquire information about the eGFR value and / or the corresponding grade of the eGFR value, the albuminuria value and / or the corresponding grade of the albuminuria value, and the cystatin C value and / or the corresponding grade of the cystatin C value, and output the acquired information, or output information about the final score and / or the corresponding grade of the final score based on the acquired information.

[0435] In one embodiment, the processor of the diagnostic device can utilize a diagnostic model to obtain information on the risk of kidney disease based on existing biomarkers, which serves as diagnostic information for kidney disease.

[0436] For example, learning can be performed using retinal images and information about kidney disease risk based on existing biomarkers labeled on the retinal images. For instance, at least one of the following can be labeled on each retinal image: eGFR value (and / or corresponding grade), albuminuria value (and / or corresponding grade), or cystatin C value (and / or corresponding grade). The diagnostic model can then learn using the retinal images and their corresponding labels. Furthermore, the labels on the retinal images can include information about at least one of the following: age, sex, ethnicity, smoking status, blood pressure (hypertension or diabetes).

[0437] The processor of a diagnostic device can obtain information about the risk of kidney disease based on existing biomarkers by inputting retinal images into a single, learned diagnostic model. For example, if the diagnostic model is learned based on eGFR values, the processor can input retinal images into the diagnostic model and obtain eGFR values ​​and / or grading information based on eGFR values ​​from the model.

[0438] Furthermore, according to an embodiment, the processor of the diagnostic device can also input information about at least one of the subject's age, sex, ethnicity, smoking status, blood pressure (hypertension), diabetes status, or cholesterol level, along with the retinal image, into the diagnostic model, thereby obtaining eGFR values ​​and / or classification information based on eGFR values ​​from the diagnostic model.

[0439] Furthermore, in one embodiment, the diagnostic model can be built in parallel. For example, the descriptions of the diagnostic model in sections 1.4.1.1 to 1.4.1.3 can be applied to the diagnostic model.

[0440] Specifically, a diagnostic model can include multiple diagnostic models. For example, a diagnostic model can include a first diagnostic model and a second diagnostic model. The first and second diagnostic models can learn using kidney disease risk information corresponding to different biomarkers. For instance, the first diagnostic model can learn using retinal images labeled with eGFR values ​​(and / or corresponding grades), and the second diagnostic model can learn using retinal images labeled with albuminuria values ​​(and / or corresponding grades). Of course, this is not limited to this; the diagnostic model can include additional diagnostic models, such as a third / fourth diagnostic model in conjunction with the first / second diagnostic models. Furthermore, the labels on the retinal images can include information about at least one of the following: age, sex, ethnicity, smoking status, blood pressure (hypertension or diabetes).

[0441] Furthermore, the processor of the diagnostic device can acquire kidney disease risk information corresponding to different biomarkers by inputting retinal images into a first diagnostic model and a second diagnostic model. For example, the processor can acquire the subject's eGFR value (and / or corresponding grade) from the first diagnostic model and the subject's albuminuria value (and / or corresponding grade) from the second diagnostic model, and output the acquired information. Additionally, the processor can output information about the final score and / or the grade corresponding to the final score based on the information acquired from each diagnostic model. For example, it can acquire and output information about the final score and / or the grade corresponding to the final score based on the subject's eGFR value (and / or corresponding grade) acquired from the first diagnostic model and the subject's albuminuria value (and / or corresponding grade) acquired from the second diagnostic model. Therefore, the processor of the diagnostic device can non-invasively acquire kidney disease risk information based on existing biomarkers, serving as diagnostic information for kidney disease, using retinal images without using invasive methods such as blood sampling.

[0442] Furthermore, in one embodiment, the diagnostic model can be constructed sequentially. For example, the descriptions of the diagnostic model in 1.4.2.1, 1.4.2.2, and 2.1.2 can be applied to the diagnostic model.

[0443] For example, diagnostic models can be like Figure 20 The diagnostic device also includes a first diagnostic model and a second diagnostic model. For example, the processor of the diagnostic device can input retinal images into the first diagnostic model and obtain from the first diagnostic model the probability value that the subject is currently at high risk of kidney disease (subject's eGFR value is less than or equal to 60 mL / min / 1.73 m). 2 The probability of having albuminuria in the current urine of the subject and / or the probability of having an eGFR value less than or equal to 60 mL / min / 1.73m 2The probability of albuminuria in the current urine sample and / or the probability of albuminuria in the current urine sample are considered. Furthermore, the processor of the diagnostic device can input the output value of the first diagnostic model and the subject's physical information (at least one of age, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), and cholesterol value). At this time, the second diagnostic model can be set with cutoff values ​​based on various biomarkers (such as eGFR value, albuminuria value, cystatin C value). For example, the cutoff value of the second diagnostic model can be set to be similar to the intermediate-high risk group of eGFR values ​​(45-59 mL / min / 1.73 mcg). 2 The proportion of the population with albuminuria at high risk (30-300 mg / g) and / or the proportion of the population with similar albuminuria levels (intermediate risk group). Therefore, kidney disease diagnostic information can predict with high accuracy the intermediate-to-high risk group (or intermediate risk group) corresponding to various biomarkers.

[0444] 2.1.5. Provide guidance information on the diagnosis of kidney disease.

[0445] Figure 25 This is a diagram illustrating a method for providing guidance information on the diagnosis of kidney disease according to one embodiment.

[0446] Reference Figure 25 A method for providing guidance information according to one embodiment may include: step S500 of obtaining kidney disease diagnostic information and step S600 of providing guidance information about the diagnostic information.

[0447] In step S500, the processor of the diagnostic device can acquire diagnostic information. Since the above description applies to step S500, a detailed explanation is omitted.

[0448] Furthermore, in step S600, the processor of the diagnostic device can provide guidance information regarding the diagnostic information. For ease of explanation, steps S500 and S600 are described primarily using kidney disease as an example, but are not limited to this; the guidance information in this specification can also be applied to various diseases such as ophthalmological diseases and cardiovascular diseases. Therefore, steps S500 and S600 will focus on the guidance information corresponding to kidney diagnostic information.

[0449] In one embodiment, guidance information may refer to medical / non-medical treatment recommendations given to the subject based on kidney disease diagnosis information. For example, guidance information may include prescription information, treatment information, and management information.

[0450] Prescription information can refer to information about medications (e.g., prescription drugs) recommended to the examinee to maintain or improve the risk of kidney disease corresponding to the diagnostic information of kidney disease. Furthermore, prescription information may include information about the prescribed medication, the time of administration, and the dosage. For example, prescription information may include information about one or more of the following: ACE inhibitors (Angiotensin-Converting Enzyme Inhibitors) (e.g., Lisinopril, Enalapril), Hydrochlorothiazide), cholesterol-lowering drugs (e.g., statins, Atorvastatin, Rosuvastatin), phosphate binders (e.g., Calcitriol, Sevelamer), and anticoagulants (e.g., Warfarin). In addition, prescription information may include SGLT2 inhibitors (Sodium-Glucose Co-Transporter 2 Inhibitors) (e.g., Dapagliflozin, Empagliflozin, Canagliflozin) as combination drugs for diabetes prescriptions, ARBs (Angiotensin II Receptor Blockers) (e.g., Losartan, Valsartan) as combination drugs for hypertension prescriptions, and GLP1 (Glucagon-Like Peptide-1) agonists and / or diuretics (e.g., Furosemide) as combination drugs for diabetes and obesity prescriptions.

[0451] In addition, the processor of the diagnostic device can provide prescription information that may be necessary for patients with kidney disease to take medications. For example, medications requiring caution when administering to patients with the aforementioned diseases include analgesics and antipyretics (nonsteroidal anti-inflammatory drugs (NSAIDs), high-dose aspirin), antibiotics (aminoglycosides, amphotericin B, cephalosporins, penicillins, beta-lactamase inhibitors, quinolones, rifampin, sulfonamides, vancomycin), antiviral drugs (acyclovir, adefovir, ganciclovir, atazanavir, indinavir, tenofovir), and bisphosphonates (pamidronate, zoledronic acid). (acid), calcineurin inhibitors (cyclosporine, tacrolimus), anticancer drugs (alkylating agents, cisplatin, methotrexate, mitomycin, interferon-alpha, proteasome inhibitors, vascular endothelial growth factor (VEGF) inhibitors, checkpoint inhibitors), iodine contrast agents used in CT and angiography, diuretics (loop diuretics, thiazides, triamterene), proton pump inhibitors.Inhibitors (Dexlansoprazole, Esomeprazole, Lansoprazole, Omeprazole, Pantoprazole, Rabeprazole), other drugs (Allopurinol, Gold sodium thiomalate, Lithium, Quinine, Sodium phosphate), traditional Chinese medicines (Aristolochic acid, Cats claw, Licorice root)

[0452] Furthermore, treatment information can refer to the follow-up treatment information recommended to the examinee in order to maintain or improve the risk of kidney disease corresponding to the diagnostic information of kidney disease. For example, supplementary examination information may include information about secondary diagnosis or medical treatment for the examinee. For example, supplementary examination information may include additional required examinations, information on hospitals / medical personnel that can perform additional examinations, and information on recommended procedures / surgeries.

[0453] In addition, management information may include non-medical treatment recommendations given to the examinee to maintain or improve the kidney disease risk corresponding to the kidney disease diagnosis information. For example, management information may include information on lifestyle habits, dietary habits, exercise, and over-the-counter medications such as nutritional supplements to reduce the risk of kidney disease.

[0454] Furthermore, in one embodiment, the diagnostic device can be linked to an external monitoring device. Here, the monitoring device can refer to a device that monitors the subject's lifestyle habits or behaviors. For example, the monitoring device can include portable devices, wearable devices, health measurement devices, etc. Additionally, the monitoring device can be the aforementioned client device. For example, the monitoring device can include an imaging unit, which captures images of the inside and outside of the eyeball to obtain images of the inside and outside of the eye.

[0455] Furthermore, the monitoring device can monitor various information, such as the subject's activity level, exercise method, exercise time, food intake, intake amount, health supplement intake information, sleep time, sleep habits, heart rate, blood pressure, blood sugar level, body water content, oxygen level, body temperature, oxygen saturation, pulse wave, medical visit status, whether examinations were performed, whether surgery was performed, eye images, eGFR value, albumin urine value, cystatin C value, etc.

[0456] The processor of the diagnostic device can communicate with the monitoring device via wired or wireless communication through the communication module.

[0457] The processor of the diagnostic device can provide guidance information to the monitoring device. Furthermore, the monitoring device can provide various information to the subject based on the guidance information and the monitored information. For example, the monitoring device can obtain management information (e.g., lifestyle, dietary, and exercise information) from the diagnostic device as guidance information, compare the management information with the monitored information, determine whether the monitored information matches the management information, and provide the determination result and / or corresponding additional information.

[0458] For example, if the monitored exercise time is less than the exercise time specified in the management information, the monitoring device can provide information to the subject, requesting them to exercise according to the management information. Furthermore, if the food intake in the monitored information corresponds to the food intake in the management information, the monitoring device can provide information to the subject, informing them that they are properly consuming food according to the management information.

[0459] Furthermore, the processor of the diagnostic device can acquire monitored information from the monitoring device. The processor can provide various information to the subject based on guidance information and the monitored information. For example, the processor can compare the monitored information with guidance information, determine whether the monitored information matches management information, and provide the determination result and / or corresponding additional information. The above example of the monitoring device can be applied to the operation of the processor of the diagnostic device.

[0460] Furthermore, the processor of the diagnostic device can generate guidance information based on monitoring information received from the monitoring device. For example, the processor of the diagnostic device can obtain the subject's status information (exercise status, lifestyle status, dietary habits, etc.) based on the monitoring information, and modify the guidance information identified as diagnostic information for kidney disease according to the subject's status information to make it suitable for the subject.

[0461] In one embodiment, the processor of the diagnostic device may utilize a predetermined database to provide guidance information. For example, the diagnostic device may include a database that matches scores and / or classifications of kidney disease diagnostic information with guidance information. For instance, if the kidney disease diagnostic information is represented by three classifications, the database may include: guidance information matching a low-risk classification (e.g., prescription information - none, treatment information - information about the next appointment, management information - providing dietary information, providing exercise information), guidance information matching a medium-risk classification (e.g., prescription information - none, treatment information - providing additional examination information, management information - providing dietary information, providing exercise information, providing over-the-counter medication information), and guidance information matching a high-risk classification (e.g., prescription information - providing statin prescription information, treatment information - additional examination information, providing recommended procedures / surgeries, management information - providing dietary information, providing exercise information, providing over-the-counter medication information). The processor of the diagnostic device may provide guidance information matching the kidney disease diagnostic information based on the database.

[0462] In another embodiment, the processor of the diagnostic device can utilize a machine learning model to provide guidance information. For example, the diagnostic device may include a guidance information model based on a machine learning model or a neural network model. The guidance information model can learn based on scores and / or classifications of kidney disease diagnostic information and guidance information. Additionally, the guidance information model can also be learned in conjunction with at least one of various physical information of the subject (e.g., height, weight, sex, race, smoking status, blood pressure (e.g., blood pressure value, hypertension status), diabetes status (or blood glucose value), and cholesterol levels). Therefore, the processor of the diagnostic device can input scores and / or classifications of kidney disease diagnostic information and the subject's physical information into the guidance information model to obtain guidance information about the subject. Furthermore, according to embodiments, the guidance information model can be included in the diagnostic model or constructed independently of the diagnostic model.

[0463] 2.1.6. Provide guidance information using existing prescription information and ancillary information about kidney disease.

[0464] Figure 26 This is a diagram illustrating a method for providing guidance information using existing prescription information and kidney disease diagnostic information according to one embodiment.

[0465] Reference Figure 26 A method for providing guidance information according to one embodiment may include: step S710 of obtaining existing prescription information of the subject, step S720 of obtaining kidney disease diagnosis information, and step S730 of providing guidance information based on existing prescription information and kidney disease diagnosis information.

[0466] In some cases, kidney disease often progresses alongside underlying conditions such as diabetes, hypertension, and / or obesity. Furthermore, some prescription medications for diabetes, hypertension, and / or obesity are also effective in treating kidney disease. For example, SGLT2 inhibitors are prescription medications for diabetes but are also effective in treating kidney disease; ARBs are prescription medications for hypertension but are also effective in treating kidney disease. Additionally, GLP-1 (Glucagon-Like Peptide-1) agonists are prescription medications for diabetes and / or obesity but are also effective in treating kidney disease.

[0467] If the subject is taking such combination medications, depending on their kidney disease risk, an additional prescription for kidney disease medication may not be necessary, or medications that would reduce the effectiveness of treatment when taken with the combination medication may need to be excluded. Prescription information should be provided to the subject. For example, some studies have indicated that combining ARBs (prescription drugs for hypertension) and ACE inhibitors (prescription drugs for kidney disease) is not recommended. Thus, when providing guidance, especially prescription information, to the subject, the medications they are currently taking must be considered to provide accurate prescription information.

[0468] Therefore, the processor of the diagnostic device can provide guidance information based on the patient's existing prescription information and kidney disease diagnosis information, especially the prescription information.

[0469] According to step S710, the processor of the diagnostic device can acquire the subject's existing prescription information. For example, the processor of the diagnostic device can acquire existing prescription information from an external device or the input module of the diagnostic device. In addition, the processor of the diagnostic device can acquire information about the subject's underlying diseases (e.g., whether they have hypertension, blood pressure value, diabetes, blood sugar value, obesity, BMI value, cholesterol value, etc.) to replace or acquire together with the existing prescription information.

[0470] Furthermore, in step S720, the processor of the diagnostic device can acquire kidney disease diagnostic information. Here, kidney disease diagnostic information may refer to kidney disease diagnostic information based on retinal images of the same subject as the subject acquired in step S710 with existing prescription information (and / or underlying disease information). Since the above description can be applied to step S720, a detailed description is omitted.

[0471] Furthermore, in step S730, the processor of the diagnostic device can provide guidance information based on existing prescription information and kidney disease diagnostic information.

[0472] The processor of the diagnostic device can provide guidance information based on kidney disease diagnostic information, as described in 2.1.5 above. However, in step S730, guidance information is provided by considering not only kidney disease diagnostic information but also existing prescription information (and / or underlying disease information).

[0473] In one embodiment, if the score and / or classification of the kidney disease diagnostic information falls into the low-risk group, the processor of the diagnostic device may not provide prescription information in the guidance information. Furthermore, the processor of the diagnostic device can determine from existing prescription information (and / or underlying disease information) whether the subject is receiving a prescription for treatment of an underlying disease, or if the underlying disease risk is high, and can provide information that is helpful for the treatment and even management of the underlying disease when providing treatment and / or management information.

[0474] Furthermore, if the score and / or classification of the kidney disease diagnostic information falls into the intermediate-risk and / or high-risk groups, the processor of the diagnostic device can provide prescribing information from the guidance information. If the existing prescribing information includes combination drugs such as SGLT2 inhibitors, ARBs, and GLP1 agonists that can treat both the underlying disease and kidney disease, the risk of the kidney disease diagnostic information can be considered when providing prescribing information. For example, if the risk of the kidney disease diagnostic aid is intermediate-risk, the processor of the diagnostic device can provide prescribing information instructing against prescribing new kidney disease treatments and instead maintaining the existing combination drugs. Furthermore, if the risk of the kidney disease diagnostic aid is high-risk, the processor of the diagnostic device can provide prescribing information about kidney disease treatments that can be taken with the combination drugs in the existing prescribing information.

[0475] In addition, if the existing prescription information does not include medications for the underlying disease, such as combination drugs, or if the risk of the underlying disease is low and the risk of assisting in the diagnosis of kidney disease is moderate or high, the processor of the diagnostic device can provide prescription information on medications (or combination drugs) for the treatment of kidney disease.

[0476] However, these examples are merely one example among many embodiments of this specification. Not limited to these examples, the processor of the diagnostic device may provide guidance information using a predetermined database and / or guidance information model, as described in 2.1.5. In this case, the database may store guidance information corresponding to existing prescription information and kidney disease diagnosis information, and the guidance information model may also learn based on the guidance information corresponding to existing prescription information and kidney disease diagnosis information. Therefore, the processor of the diagnostic device may obtain guidance information from the database using existing prescription information and kidney disease diagnosis information, or it may input existing prescription information (or existing prescription information and kidney disease diagnosis information) into the guidance information model to obtain guidance information.

[0477] 2.1.7. Predicting the rate of progression of kidney disease using diagnostic information for kidney disease.

[0478] Figure 27 This is a diagram illustrating a method for predicting the rate of progression of kidney disease using diagnostic information of kidney disease, according to one embodiment.

[0479] Reference Figure 27 A method for predicting the rate of progression of kidney disease according to one embodiment may include: step S810 of obtaining the result value corresponding to a biomarker, step S820 of obtaining kidney disease diagnostic information, and step S830 of determining the rate of progression of kidney disease using the result value corresponding to the biomarker and the kidney disease diagnostic information.

[0480] In step S810, the processor of the diagnostic device can acquire the result value corresponding to the biomarker. For example, the processor of the diagnostic device can acquire the score and / or grade of the biomarker from an external device or the input module of the diagnostic device. Here, the biomarker may include the aforementioned eGFR value (and / or corresponding grade), albuminuria value (and / or corresponding grade), or cystatin C value (and / or corresponding grade), etc. In addition, the biomarker may include other kidney disease biomarkers or kidney disease risk assessment tools.

[0481] Furthermore, in step S820, the processor of the diagnostic device can acquire kidney disease diagnostic information. Here, kidney disease diagnostic information may refer to kidney disease diagnostic information based on retinal images of the same subject as the subject whose biomarker corresponding result value was acquired in step S810. Since the above description can be applied to step S820, a detailed description is omitted.

[0482] In addition, in step S830, the rate of progression of kidney disease can be determined by using the result values ​​corresponding to the biomarkers and the diagnostic information of kidney disease.

[0483] Specifically, even if the corresponding values ​​for biomarkers are the same, the rate of progression of kidney disease may differ. For example, even if the eGFR value is 50 mL / min / 1.73 m... 2 Even within the intermediate-to-high-risk group, the scores on kidney disease diagnostic information may differ. However, even with the same eGFR value, individuals with higher kidney disease diagnostic information scores may experience faster kidney disease progression than those with lower scores.

[0484] This will be explained in detail below with reference to Example 2.

[0485] 2.1.7.1. Example 2

[0486] Example 2 is to use KDIGO classification and kidney disease diagnostic information to predict the rate of kidney disease progression.

[0487] In Example 2, data from 5,346 diabetic patients from two tertiary hospitals in South Korea were used. This may include data from individuals with an eGFR <90 ml / min / 1.73m2 or those with albuminuria. Furthermore, data from individuals lacking retinal images, serum creatinine, or albuminuria information were excluded.

[0488] Figure 28 This is a diagram used to illustrate the clinical characteristics of the object according to Example 2.

[0489] Reference Figure 28 The KDIGO classification can be divided into three groups (low-risk, intermediate-risk, and high-risk) based on eGFR (eGFR) value (or classification) and albuminuria value (or classification). Figure 28 The table categorizes kidney disease risk into three groups based on albuminuria levels: low-risk (A1, less than 30 mg / g), intermediate-risk (A2, 30-300 mg / g), and high-risk (A3, 300 mg / g or higher). Furthermore, based on eGFR values, kidney disease risk is categorized into three groups: low-risk (G1, 90 mL / min / 1.73 mcg) and high-risk (A3, 300 mg / g or higher). 2 (or higher), intermediate-risk group (G2, 60-89 mL / min / 1.73m) 2 ), medium-to-high risk group (G3a, 45-59 mL / min / 1.73m 2 ), extremely high risk group (G3b, 30-44 mL / min / 1.73m), 2 The highest risk group (G4, 29 mL / min / 1.73m) 2 (or below).

[0490] In the KIDGO classification, the low-risk group includes 3,135 people, comprising low-risk based on albuminuria value, low-risk based on eGFR value, and a portion of the intermediate-risk group; the intermediate-risk group includes 1,814 people, comprising low-risk and intermediate-risk based on albuminuria value, low-risk based on eGFR value, intermediate-risk based on eGFR value, and a portion of the intermediate-high-risk group; the high-risk group may include 397 people, comprising all risk groups based on albuminuria value and a portion of all risk groups based on eGFR value.

[0491] Furthermore, the mean age of the participants was 62.4 (+ / - 11.4) years, and 60.6% were male. The mean eGFR was 86.6 (+ / - 15.3 mL / min per 1.73 m2), and 46.9% had albuminuria. During the 5.0-year follow-up period (interquartile range: 2.5–7.8), 1,379 (25.8%) participants experienced kidney disease-related events.

[0492] Figure 29a and Figure 29b This is a graph used to illustrate the incidence of kidney disease-related events based on KDIGO classification and kidney disease diagnostic information according to Example 2.

[0493] Reference Figure 29a and Figure 29b ,exist Figure 29a and Figure 29b In the chart, the x-axis represents time (years), and the y-axis represents the incidence of chronic kidney disease-related events.

[0494] Figure 29a The chart shows the incidence of kidney disease-related events according to the KDIGO classification (low-risk group a1, intermediate-risk group a2, high-risk group a3), with the incidence of kidney-related events likely being higher in the high-risk group a3.

[0495] Figure 29b The chart shows the incidence of kidney disease-related events when the scores corresponding to kidney disease diagnostic information are reflected in the KDIGO classification.

[0496] The first group, b1, represents individuals in the low-risk KDIGO group whose kidney disease diagnostic information score is less than 20; the second group, b2, represents individuals in the low-risk KDIGO group whose kidney disease diagnostic information score is greater than or equal to 20; the third group, b3, represents individuals in the intermediate-risk KDIGO group whose kidney disease diagnostic information score is less than 20; the fourth group, b4, represents individuals in the intermediate-risk KDIGO group whose kidney disease diagnostic information score is greater than or equal to 20; and the fifth group, b1, can represent the high-risk KDIGO group.

[0497] Specifically, Figure 29a In the chart, the incidence of kidney disease-related events in low-risk group a1 can be differentiated or stratified in the same way as the incidence of kidney disease-related events in group 1 b1 and group 2 b2 in (b). Here, the incidence of kidney disease-related events in group 1 b1 in (b) is lower than Figure 29a The chart shows the incidence of kidney disease-related events in the low-risk group a1. Figure 29b The incidence of kidney disease-related events in group b2 of the chart may be higher than that in the second group. Figure 29aThe chart shows the incidence of kidney disease-related events in the low-risk group a1. That is, even among individuals belonging to the same KDIGO low-risk group, the incidence of kidney disease-related events may be differentiated or stratified based on their kidney disease diagnostic information scores. In other words, even among individuals belonging to the same KDIGO low-risk group, those with higher kidney disease diagnostic information scores may have a higher probability of experiencing kidney disease-related events than those with lower scores. This may mean that, even among individuals belonging to the same KDIGO low-risk group, those with higher kidney disease diagnostic information scores may experience faster kidney disease progression than those with lower scores.

[0498] also, Figure 29a The incidence of kidney disease-related events in the intermediate-risk group a2 in the chart can be like... Figure 29b The graphs show that the incidence of kidney disease-related events is the same in group 3 (b3) and group 4 (b4), allowing for differentiation or stratification. Figure 29b The incidence of kidney disease-related events in group b3 of the chart is lower than that in group b3. Figure 29a The incidence of kidney disease-related events in the intermediate-risk group a2 is shown in the chart. Figure 29b The incidence of kidney disease-related events in group 4b4 of the chart may be higher than that in the other group. Figure 29a The graph shows the incidence of kidney disease-related events in the intermediate-risk group a2. Based on these results, as mentioned above, even among individuals belonging to the same KDIGO intermediate-risk group, those with higher kidney disease diagnostic information scores may experience faster kidney disease progression than those with lower scores.

[0499] Furthermore, such as Figure 29a As shown in the chart, if only the results corresponding to existing biomarkers are considered, the incidence of kidney disease-related events in the low-risk group a1 of the KDIGO classification may be higher than that in the intermediate-risk group a2 of the KDIGO classification.

[0500] However, as Figure 29b As shown in the chart, the incidence of kidney disease-related events in Group 2b2, where individuals in the low-risk KDIGO group had a score of 20 or higher for kidney disease diagnostic information, was likely higher than the incidence in Group 3b3, where individuals in the intermediate-risk KDIGO group had a score of less than 20 for kidney disease diagnostic information. Therefore, even among individuals with the same low KDIGO classification, a higher score for kidney disease diagnostic information compared to those with a high KDIGO classification may indicate a higher risk of kidney disease and a faster progression of kidney disease.

[0501] Furthermore, the C-statistic when the scores corresponding to kidney disease diagnostic information are reflected in the KDIGO classification is 0.04 higher than the C-statistic when only the KDIGO classification is used (95% confidence interval (CI): 0.02-0.04). This means that reflecting the scores corresponding to kidney disease diagnostic information in the KDIGO classification can predict the risk of kidney disease more accurately than using the KDIGO classification alone.

[0502] To reiterate Figure 27 In one embodiment, the processor of the diagnostic device can determine whether the score of the kidney disease diagnostic information is greater than or equal to a predetermined threshold.

[0503] Furthermore, if the score of the kidney disease diagnostic information is greater than or equal to a predetermined threshold, the processor of the diagnostic device can determine that the kidney disease is progressing rapidly; if the score of the kidney disease diagnostic information is less than the predetermined threshold, it can determine that the kidney disease is progressing slowly.

[0504] For example, as shown in Example 2, even if the first and second subjects have the same KDIGO classification, eGFR value, albuminuria value, and cystatin C value, if the kidney disease diagnosis information score of the first subject is greater than or equal to a predetermined threshold (e.g., kidney disease diagnosis information score 20), while the kidney disease diagnosis information score of the second subject is less than the predetermined threshold, the processor of the diagnostic device can determine that the kidney disease progression rate of the first subject is faster than that of the second subject.

[0505] Furthermore, in one embodiment, in step S830, the processor of the diagnostic device can predict acute kidney injury (AKI) after cardiac surgery. Specifically, after cardiac surgery, kidney injury may or may not occur, depending on the patient. For example, after cardiac surgery, 5% of patients develop chronic kidney disease, and 1-2% of patients experience complete kidney function impairment, potentially requiring dialysis or a kidney transplant.

[0506] While the aforementioned existing biomarkers can be used to predict acute kidney injury after cardiac surgery, their accuracy may be limited. Because the kidney disease diagnostic information in this specification can predict the risk of kidney injury more accurately than existing biomarkers, it can also predict the risk of acute kidney injury after cardiac surgery that existing biomarkers cannot predict.

[0507] For example, the processor of the diagnostic device can determine whether the score of kidney disease diagnostic information is greater than or equal to a predetermined threshold. This predetermined threshold may be the same as or different from a threshold used to predict the rate of progression of kidney disease. Furthermore, if the score of the kidney disease diagnostic information is greater than or equal to the predetermined threshold, the processor of the diagnostic device can determine that the likelihood of acute kidney injury after cardiac surgery is high; if the score of the kidney disease diagnostic information is less than the predetermined threshold, it can determine that the likelihood of acute kidney injury after cardiac surgery is low.

[0508] Furthermore, the processor of the diagnostic device can provide information about the rate of progression of kidney disease (or information about acute kidney injury after cardiac surgery) and / or corresponding guidance. In this case, because the rate of progression of kidney disease (or the likelihood of acute kidney injury after cardiac surgery) differs between the first and second subjects, the processor of the diagnostic device can provide different guidance to the first and second subjects.

[0509] For example, even if the first and second subjects belong to the same risk group based on existing biomarkers and / or kidney disease diagnostic information, different guidance information can be provided based on the rate of kidney disease progression in the first and second subjects. For instance, if the first subject's kidney disease progression is predicted to be rapid and the first subject belongs to a low-risk group, the processor of the diagnostic device can provide guidance information aimed at improving the management of blood pressure, diabetes, etc. For example, the processor of the diagnostic device can provide treatment information including information on additional examinations (e.g., instructions for additional examinations, dates of additional examinations, information on hospitals / medical personnel where additional examinations can be performed, etc.) and / or management information including recommended lifestyle modification goals, recommended dietary information, etc., as guidance information.

[0510] Furthermore, if the first subject's kidney disease is predicted to progress rapidly and the first subject belongs to a high-risk group, the processor of the diagnostic device can increase the prescription dose for the first subject as prescription information.

[0511] Furthermore, the processor of the diagnostic device can communicate with the aforementioned monitoring device. The processor can provide guidance information determined based on the rate of kidney disease progression to the monitoring device. For example, if the processor determines that the kidney disease of the first subject in a low-risk group (asymptomatic and without kidney risk factors) is progressing rapidly, it can determine appropriate treatment information (e.g., additional examination information, etc.) and / or management information (e.g., controlling food intake, target exercise levels, etc.) as guidance information and provide this guidance information to the monitoring device. The monitoring device can then provide the guidance information obtained from the diagnostic device to the subject and determine whether the monitored information matches the guidance information. If the result is a match, the monitoring device can provide information informing the subject that they have followed the guidance information well; if there is a mismatch, the monitoring device can issue a warning to the subject, requiring them to follow the guidance information.

[0512] Furthermore, the processor of the diagnostic device can provide guidance based on the likelihood of acute kidney injury after cardiac surgery. For example, if the likelihood of acute kidney injury after cardiac surgery is determined to be high, the processor of the diagnostic device can provide recommended guidance to reduce the likelihood of acute kidney injury after cardiac surgery.

[0513] Furthermore, the processor of the diagnostic device can provide guidance information using a predetermined database and / or guidance information model, as described in 2.1.5. In this case, the database may store guidance information matching kidney disease diagnostic information with the rate of kidney disease progression (or information on acute kidney injury after cardiac surgery), and the guidance information model can also learn based on the guidance information corresponding to the kidney disease diagnostic information and the rate of kidney disease progression (or the likelihood of acute kidney injury after cardiac surgery). Therefore, the processor of the diagnostic device can obtain guidance information from the database and / or guidance information model using kidney disease diagnostic information and the rate of kidney disease progression (or the likelihood of acute kidney injury after cardiac surgery).

[0514] For example, if the eGFR value is in the low-risk group and the score of the kidney disease diagnostic information is greater than or equal to a predetermined threshold, the processor of the diagnostic device can determine that the subject's kidney risk is low, but the subject's kidney disease is progressing rapidly compared to other low-risk groups. Therefore, the processor of the diagnostic device can provide information about the rapid progression of kidney disease and / or corresponding guidance information (e.g., guidance information provided when the kidney disease diagnostic information is determined to be of intermediate risk).

[0515] Furthermore, if the eGFR value is in the low-risk group and the score of the kidney disease diagnostic information is less than a predetermined threshold, the processor of the diagnostic device can determine that the subject's kidney risk is also low, but the subject's kidney disease progression is slow compared to other low-risk groups. Therefore, the processor of the diagnostic device can provide information about the slow progression of kidney disease and / or corresponding guidance information (e.g., guidance information provided when the kidney disease diagnostic information is determined to be low-risk).

[0516] Furthermore, according to the embodiments, different thresholds can be applied to kidney disease diagnostic information based on the result values ​​corresponding to the biomarkers. For example, different thresholds can be set in the diagnostic device when the KDIGO classification is low-risk, intermediate-risk, or high-risk.

[0517] Furthermore, according to embodiments, the threshold can be set based on various benchmarks. For example, the threshold can be set such that the incidence of kidney disease events is explicitly stratified when kidney disease diagnostic information is applied to the outcome values ​​corresponding to biomarkers.

[0518] Various embodiments of this specification can be implemented as software, including instructions recorded on a machine-readable storage medium (e.g., a computer). A machine is an apparatus that can invoke stored instructions from the storage medium and operate according to the invoked instructions, and may include an electronic device according to the disclosed embodiments. When the instructions are executed by a processor, the processor may perform the function corresponding to the instructions directly or under the control of the processor using other constituent elements. Instructions may include code generated or executed by a compiler or interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory storage medium" means that it does not include signals and is tangible, but does not distinguish whether data is permanently or temporarily stored in the storage medium. For example, a "non-transitory storage medium" may include a buffer that temporarily stores data.

[0519] According to one embodiment, the methods of various embodiments disclosed in this specification can be included in a computer program product. The computer program product can be traded as a commodity between a seller and a buyer. The computer program product can be in the form of a machine-readable storage medium (e.g., a read-only optical disc, CD-ROM) or distributed online through an app store (e.g., the Play Store™). In the case of online distribution, at least a portion of the computer program product (e.g., a downloadable application) can be at least temporarily stored or temporarily generated in a storage medium such as the memory of a manufacturer's server, an app store's server, or a relay server.

[0520] As described above, although the embodiments have been illustrated with limited examples and drawings, those skilled in the art can make various modifications and variations from the foregoing description. For example, suitable results can be achieved even if the described techniques are performed in a different order than the described methods, and / or the constituent elements of the described systems, structures, devices, circuits, etc. are combined or integrated in a different form than the described methods, or are replaced or substituted by other constituent elements or equivalents.

[0521] Therefore, other implementations, other embodiments, and things equivalent to the scope of the claims also fall within the scope of the claims.

Claims

1. A control method for a diagnostic device, characterized in that, include: Steps for acquiring retinal images of a subject; as well as The step of obtaining diagnostic information about kidney disease in the subject using a machine learning model based on the retinal image. The machine learning model includes a first model and a second model, wherein the first model is a neural network model and the second model is a regression-based machine learning model.

2. The control method for the diagnostic device according to claim 1, characterized in that, The first model is trained based on first training data and corresponding results from a first biomarker. The first training data includes multiple retinal images. In the step of obtaining diagnostic information about kidney disease in the subject, The retinal image is input into the first model, and a first result value is obtained from the first model. The first score and the physical information of the subject are input into the second model, and the second result value is obtained from the second model. The kidney disease diagnostic information is obtained based on the second result value.

3. The control method for the diagnostic device according to claim 2, characterized in that, The probability that the result of the first biomarker is equal to or less than the first value of eGFR (Estimated Glomerular Filtration Rate), or information about the presence of albuminuria in the urine of the subject, is at least one of the following: The first result value represents at least one of the following: the probability that the eGFR value of the subject is equal to or less than the first value, or the probability that albuminuria is present in the urine of the subject.

4. The control method for the diagnostic device according to claim 3, characterized in that, The second machine learning model is trained based on second training data, which includes kidney disease event tracking results of objects in a predetermined population.

5. The control method for the diagnostic device according to claim 4, characterized in that, In the predetermined population of objects, data concerning objects whose eGFR values ​​are less than a second value are excluded from the second training data, where the second value is higher than the first value.

6. The control method for the diagnostic device according to claim 1, characterized in that, The first result value includes information about the subject's current risk of kidney disease. The second result value includes information about the probability of future kidney disease-related events occurring in the subject.

7. The control method for the diagnostic device according to claim 6, characterized in that, The second result value includes information about the probability of the subject experiencing kidney disease-related events within 5 years.

8. The control method for the diagnostic device according to claim 1, characterized in that, In the step of obtaining diagnostic information about kidney disease in the subject, A predetermined cutoff value is applied to the second result value as the diagnostic information for the kidney disease, and any grade corresponding to the subject is obtained from multiple grades.

9. The control method for the diagnostic device according to claim 8, characterized in that, The cutoff value is set based on the population distribution in multiple hierarchical groups divided according to the tracking and observation results of a predetermined population.

10. The control method for the diagnostic device according to claim 1, characterized in that, It also includes the step of providing guidance information to the subject based on the kidney disease diagnosis information.

11. The control method for the diagnostic device according to claim 10, characterized in that, In the step of providing guidance information to the subject based on the kidney disease diagnostic information Using a pre-stored database or machine learning model, guidance information corresponding to the diagnostic information of the kidney disease can be provided.

12. The control method for the diagnostic device according to claim 10, characterized in that, It also includes the step of determining the rate of progression of the subject's kidney disease based on the kidney disease diagnostic information. In the step of determining the rate of progression of kidney disease in the subject based on the aforementioned kidney disease diagnostic information... Obtain the biomarker results of the subject. The rate of kidney disease progression in the subject is determined by comparing the results of the biomarkers with the second result value.

13. The control method for the diagnostic device according to claim 12, characterized in that, When the second result value is greater than or equal to the predetermined value, it is determined that the rate of progression of kidney disease is faster than the rate of progression of kidney disease expected based on the result value of the biomarker. When the second result value is less than a predetermined value, it is determined that the progression of kidney disease is slower than the rate of kidney disease progression expected based on the result value of the biomarker.

14. The control method for the diagnostic device according to claim 12, characterized in that, In the step of providing guidance information to the subject based on the kidney disease diagnostic information The subject is given guidance information based on the rate of progression of kidney disease and the diagnostic information of the kidney disease.

15. A recording medium having a program for performing the method according to any one of claims 1 to 14.

16. A diagnostic device, characterized in that, include: Storage module; as well as At least one processor, The at least one processor is used for: Acquire retinal images of the subject; Based on the retinal image, the machine learning model stored in the storage module is used to obtain diagnostic information about the kidney disease of the subject. The machine learning model includes a first model and a second model. The first model is a neural network model. The second model is a regression-based machine learning model.

Images