Ophthalmic device, program, and recording medium

Abstract

The glaucoma screening method according to an embodiment includes the following steps. In a calculation model preparation step, a calculation model is prepared, the calculation model including a linear combination of at least one papilla parameter that represents the thickness of a retinal tissue in an optic papilla area and a linear combination of at least one macula parameter that represents the thickness of a retinal tissue in a macula area. In an OCT data generation step, the ocular fundus of an eye to be tested is subjected to OCT scanning to generate OCT data. In a measurement value calculation step, a plurality of measurement values including a papilla measurement value corresponding to each papilla parameter and a macula measurement value corresponding to each macular parameter of the calculation model are calculated on the basis of the OCT data. In a measurement value input step, the calculated plurality of measurement values are input into the calculation model. In a calculation result provision step, a calculation result output from the calculation model in accordance with the inputting of the plurality of measurement values is provided.

Description

Glaucoma screening method, ophthalmic device, program, and recording medium

[0001] The present disclosure relates to a glaucoma screening method, an ophthalmic apparatus, a program, and a recording medium.

[0002] Glaucoma is one of the leading causes of acquired blindness. The number of glaucoma patients worldwide is estimated to be more than 60 million, and this number is expected to continue to increase. Several epidemiological studies have revealed that many glaucoma patients have not been diagnosed and are unaware that they have the disease.

[0003] To prevent limitations on daily life due to visual field loss and blindness caused by glaucoma, it is important to detect the disease early and start treatment. Widespread provision of screening using ophthalmologic examinations is thought to be an effective way to promote early detection. However, mass screening for glaucoma is not currently common in many countries.

[0004] JP 5-184542 A U.S. Patent Application Publication No. 2018 / 0025112

[0005] To realize glaucoma mass screening, several problems must be solved. Two representative problems are described below.

[0006] The first problem is the reliability of glaucoma identification. For example, conventional glaucoma identification methods using image analysis often incorrectly identify normal eyes as having retinal thinning due to artifacts and vascular images. In other words, conventional methods suffer from a high rate of false positives. A specialist can easily distinguish false positives caused by artifacts. However, because mass screening is typically performed in an environment without a specialist, normal eyes may be classified as suspected glaucoma. The subject undergoes detailed examinations at a specialized hospital, but the results are often negative. Such situations not only reduce the reliability of the mass screening facility and the screening method, but also reduce the quality of the statistical data generated from the data collected through mass screening. Furthermore, individual subjects incur unnecessary psychological, labor, financial, and time burdens. Therefore, improving the quality (precision, accuracy, etc.) of disease identification is a prerequisite for achieving glaucoma mass screening.

[0007] The second problem is the difficulty and complexity of identifying normal-tension glaucoma. Normal-tension glaucoma is a type of glaucoma that differs from the type of glaucoma that develops when elevated intraocular pressure damages the optic nerve. Normal-tension glaucoma is a type of glaucoma in which the optic nerve is damaged despite normal intraocular pressure. Conventional glaucoma screening methods often begin with screening using intraocular pressure as an indicator (biomarker), making it impossible to identify normal-tension glaucoma, which does not exhibit elevated intraocular pressure. The only way to identify normal-tension glaucoma is for a doctor to make a judgment after performing fundus examinations, visual field examinations, and other tests in addition to intraocular pressure tests. It is clear that such a method is unsuitable for mass screening.

[0008] One object of the present disclosure is to provide a glaucoma identification technology suitable for mass screening.

[0009] One exemplary aspect of the present disclosure is a method for glaucoma screening, comprising: preparing a computational model including a linear combination of one or more disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area; applying an optical coherence tomography scan to the fundus of a test eye to generate OCT data; calculating a plurality of measurement values ​​based on the OCT data, the measurement values ​​including one or more disc measurement values ​​corresponding to the one or more disc parameters, and one or more macular measurement values ​​corresponding to the one or more macular parameters; inputting the plurality of measurement values ​​into the computational model; and providing a calculation result output from the computational model in response to the input of the plurality of measurement values.

[0010] Another exemplary aspect of the present disclosure is an ophthalmic device that can be used for glaucoma screening, comprising: a memory unit that stores a pre-created computational model including a linear combination of one or more disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area; an OCT data acquisition unit that acquires optical coherence tomography (OCT) data of the fundus of the test eye; a measurement calculation unit that calculates a plurality of measurement values ​​based on the OCT data, including one or more disc measurement values ​​corresponding to the one or more disc parameters and one or more macular measurement values ​​corresponding to the one or more macular parameters; and a risk information generation unit that generates glaucoma risk information for the test eye based on the computational model and the plurality of measurement values.

[0011] Yet another exemplary aspect of the present disclosure is a program capable of causing a computer including a processor and a memory to perform glaucoma screening, the program causing the processor to perform the following operations: control to store in the memory a computational model including a linear combination of one or more disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area; control to accept optical coherence tomography (OCT) data of the fundus of the test eye; process to calculate, based on the OCT data, a plurality of measurement values ​​including one or more disc measurement values ​​corresponding to the one or more disc parameters and one or more macular measurement values ​​corresponding to the one or more macular parameters; process to input the plurality of measurement values ​​into the computational model; and process to provide a calculation result output from the computational model in response to the input of the plurality of measurement values.

[0012] Yet another exemplary aspect of the present disclosure is a computer-readable non-transitory recording medium having a program recorded thereon capable of causing a computer including a processor and a memory to perform glaucoma screening, the program causing the processor to perform the following operations: control to store in the memory a computational model including a linear combination of one or more disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area; control to accept optical coherence tomography (OCT) data of the fundus of the test eye; process to calculate, based on the OCT data, a plurality of measurement values ​​including one or more disc measurement values ​​corresponding to the one or more disc parameters, respectively, and one or more macular measurement values ​​corresponding to the one or more macular parameters, process to input the plurality of measurement values ​​into the computational model; and process to provide a calculation result output from the computational model in response to the input of the plurality of measurement values.

[0013] According to an exemplary aspect of the present disclosure, a glaucoma identification technique suitable for mass screening can be provided.

[0014] 1 is a flowchart showing the procedure of a glaucoma screening method according to an exemplary aspect of an embodiment. 2 is a flowchart showing the procedure of a glaucoma screening method according to an exemplary aspect of an embodiment. 3 is a flowchart for explaining the creation of a computational model in a glaucoma screening method according to an exemplary aspect of an embodiment and verification of its quality. 4 is a diagram for explaining OCT measurement predictors in a glaucoma screening method according to an exemplary aspect of an embodiment. 5 is a diagram for explaining OCT measurement predictors in a glaucoma screening method according to an exemplary aspect of an embodiment. 6 is a diagram for explaining OCT measurement predictors in a glaucoma screening method according to an exemplary aspect of an embodiment. 7 is a table showing characteristics of cases and controls in development data used in the development of a computational model in a glaucoma screening method according to an exemplary aspect of an embodiment. 8 is a histogram showing the distribution of cases and controls for OCT measurement predictors in a glaucoma screening method according to an exemplary aspect of an embodiment. 9 is a table showing three computational models in a glaucoma screening method according to an exemplary aspect of an embodiment. FIG. 1 is a receiver operating characteristic curve of three calculation models in a glaucoma screening method according to an exemplary aspect of an embodiment; FIG. 2 is a table for explaining verification results of three calculation models in a glaucoma screening method according to an exemplary aspect of an embodiment; FIG. 3 is a table for explaining verification results of three calculation models in a glaucoma screening method according to an exemplary aspect of an embodiment; FIG. 4 is a block diagram showing the configuration of an ophthalmologic apparatus according to an exemplary aspect of an embodiment.

[0015] Some non-limiting embodiments of the present disclosure will now be described in detail with reference to the drawings.

[0016] Any known technology can be combined with the embodiments of the present disclosure. For example, any matter freely selected from the matters described in any document cited in this specification can be combined with the embodiments of the present disclosure. Furthermore, any technology known in any technical field related to the present disclosure can be combined with the embodiments of the present disclosure. The matters that can be combined with the embodiments of the present disclosure are not limited thereto. For example, any technical matter disclosed by the applicant of the present application (e.g., any technical matter disclosed by any means such as a patent application, a paper, or a website) can be incorporated by reference into the present disclosure.

[0017] Two or more exemplary aspects of the present disclosure may be at least partially combined.

[0018] Any of the various functions of the various elements described in the embodiments according to the present disclosure is implemented using circuitry or processing circuitry. Circuitry or processing circuitry may be a general-purpose processor, a special-purpose processor, an integrated circuit, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), a Field Programmable Gate Array (FPGA)), or a combination of these devices configured and / or programmed to perform a certain function. Array), conventional circuit configurations, and any combination thereof. A processor is considered to be processing circuitry or circuit configurations including transistors and / or other circuitry. In this disclosure, a circuit configuration, unit, means, or similar terms refers to hardware that performs a certain function or hardware that is programmed to perform a certain function. The hardware may be any of the various hardware described in the embodiments of the present disclosure, or may be hardware that is programmed and / or configured to perform a certain function. When a processor that can be considered to be a certain type of circuit configuration is employed as hardware, the terms circuit configuration, unit, means, or similar terms refer to an element that includes a combination of hardware and software, and the software is used to configure the hardware and / or processor.

[0019] <Overview of Embodiments> One objective of some exemplary aspects of the present disclosure is to propose a glaucoma identification technology suitable for mass screening. To this end, the glaucoma screening technology according to these exemplary aspects focuses on and aims to solve or alleviate the two problems described above, namely, the problem of reliability in identifying glaucoma (problem 1) and the problem of difficulty and complexity in identifying normal-tension glaucoma (problem 2). Those skilled in the art will be able to understand from the embodiments described below that some exemplary aspects of the present disclosure solve or alleviate the first and second problems.

[0020] The embodiments of the present disclosure are not limited to those that focus on both of these two problems, and may, for example, focus on only one of these two problems, or may focus on one or both of these two problems and another problem, or may focus on a problem other than these two problems. In other words, the embodiments of the present disclosure may solve or mitigate only one of these two problems, or may solve or mitigate one or both of these two problems and another problem, or may solve or mitigate a problem other than these two problems. Hereinafter, unless otherwise specified, "solving a problem" and "mitigating a problem" are synonymous.

[0021] Optical coherence tomography (OCT) is a non-contact, high-speed, and non-invasive imaging modality. Although OCT is a relatively new technology, its adoption in the field of ophthalmology has progressed remarkably and is now used in many medical institutions, including small clinics. Therefore, OCT is an imaging modality suitable for use in mass screening.

[0022] In an embodiment according to the present disclosure, OCT is used to generate a cross-sectional image of the fundus. More specifically, the glaucoma identification in the embodiment according to the present disclosure is performed essentially based solely on an image obtained using OCT (referred to as an OCT image). In contrast, various known glaucoma identification techniques essentially use not only an image obtained using OCT (referred to as an OCT image), but also subject background information, intraocular pressure data, visual field test data, and images obtained using ophthalmic modalities other than OCT (such as a slit lamp microscope, a fundus camera, or a scanning laser ophthalmoscope). Even if a glaucoma identification technique using only OCT images exists, it would differ from the glaucoma identification in the embodiment according to the present disclosure in that it is unable to solve at least the first and / or second problems and does not have a configuration for solving these problems.

[0023] The type of OCT used in the embodiments of the present disclosure may be any type, but typically may be a spectral domain OCT (SD-OCT) or a swept source OCT (SS-OCT). Although the embodiments described below use a spectral domain OCT, those skilled in the art will understand that embodiments with similar configurations and similar effects can be obtained using other types of OCT.

[0024] In an embodiment according to the present disclosure, an algorithm suitable for glaucoma mass screening using OCT is created and verified. To this end, in the embodiment described below, an algorithm (computational model) is created that functions to generate a glaucoma risk score from OCT images using important indices selected from various indices obtained from OCT data of the fundus.

[0025] Several aspects of glaucoma screening methods that can be implemented using this algorithm are described below, which are exemplary and not intended to limit the scope of the disclosed embodiments.

[0026] A glaucoma screening method (glaucoma identification method) according to one embodiment is shown in Figure 1. Details of each step will be described later, with non-limiting examples.

[0027] In the method according to this aspect, first, a calculation model for determining a glaucoma risk score from an OCT image is created (step S11).

[0028] Step S11 is executed by a computer operating according to a computational model creation program. This program may include a program utilizing machine learning technology. For example, development data (learning data, teacher data, training data) including a large number of OCT data (or a large number of measurement values ​​calculated from a large number of OCT data) labeled with labels (glaucoma risk score values) created by a specialist or another computational model, and a mathematical model such as a neural network are prepared, and machine learning using this development data is applied to this mathematical model to construct the computational model. Some examples of methods for creating a computational model will be described later.

[0029] The computational model created in step S11 includes parameters related to the optic disc (optic disc parameters) and parameters related to the macula (macular parameters). That is, this computational model is a mathematical model configured to receive inputs of measured values ​​corresponding to the optic disc parameters and measured values ​​corresponding to the macular parameters, and to output calculation results based on these measured values.

[0030] The optic disc parameter is any parameter measured or calculated with respect to the optic disc, and is, for example, a parameter representing the retinal tissue thickness in an area defined based on the optic disc (optic disc area). The optic disc area may be, for example, at least a portion of the optic disc, an area surrounding the optic disc, or a combination thereof. The retinal tissue thickness represents the dimensions (thickness) of one or more tissues constituting the retina. The retina is histologically divided into the following 10 layers: retinal pigment epithelium layer, photoreceptor layer, outer limiting membrane, outer nuclear layer, outer plexiform layer, inner nuclear layer, inner plexiform layer, ganglion cell layer, nerve fiber layer, and internal limiting membrane. The retinal tissue thickness may be the thickness of one or more of these layers. The optic disc parameter is not limited to these examples, and may be any parameter that can be derived based on data collected by OCT scanning from an area of ​​the fundus defined based on the optic disc, such as cup dimensions (diameter, area, etc.), disk dimensions, rim dimensions, the ratio of the two dimensions, or optic disc tilt.

[0031] The macular parameter is any parameter measured or calculated with respect to the macula, such as a parameter representing the thickness of retinal tissue in an area defined for the macula (macular area). The macular area may be, for example, at least a portion of the macula, an area surrounding the macula, or a combination thereof. The macular parameter is not limited to these examples and may be any parameter that can be derived based on data collected by OCT scanning from an area of ​​the fundus defined based on the macula, such as macular dimensions (diameter, area, depth, etc.). The retinal tissue corresponding to the macular parameter may be the same as or different from the retinal tissue corresponding to the disc parameter.

[0032] The computational model of this embodiment is expressed as a mathematical formula in a predetermined form that includes one or more disc parameters and one or more macular parameters as variables. The mathematical formula may be in any form.

[0033] Some exemplary computational models include a linear combination of one or more disc parameters and / or a linear combination of one or more macular parameters, in other words, some exemplary computational models include linear equations with one or more disc parameters and / or one or more macular parameters as variables.

[0034] As is well known, N parameters P １ , P ２ , ..., P Ｎ The linear combination of the coefficients c １ , c ２ , ..., c Ｎ Using the formula, it is expressed as follows: １ P １ +c ２ P ２ +...+c Ｎ P Ｎ Also, N variables x １ , x ２ , ..., x Ｎ The linear equation y = f(x) includes coefficients a0, a １ , a ２ , ..., a Ｎ Using this, it is expressed as follows: y = a0 + a １ x １ +a ２ x ２ +...+a Ｎ x Ｎ .

[0035] The mathematical expressions representing some exemplary computational models may include terms of a type other than zero-order terms and first-order terms. For example, the mathematical expressions of some exemplary computational models may include polynomials of second or higher order. Furthermore, the mathematical expressions of some exemplary computational models may include mathematical symbols other than the four basic arithmetic operations (e.g., radicals, derivatives, integrals, etc.).

[0036] Three examples of calculation models defined as multivariable linear equations with both disc parameters and macular parameters as variables (Y(Model 1), Y(Model 2), and Y(Model 3)) are shown below. Each variable (e.g., TSNITlower) will be described later.

[0037] [First calculation model] Y (Model 1) = 14.4305 - 0.0404 × TSNITlower - 0.0303 × cpRNFLqS - 0.0304 × cpRNFLqIt - 0.0271 × cpRNFLqIn - 0.1424 × mGCIPL_IT + 1.1427 × log_mGCIPL_STvsIT + 0.6971 × log_mGCIPL_ITvsIN + 0.0602 × mGCC_IN

[0038] [Second calculation model] Y (Model 2) = 12.6694 - 0.0329 × TSNITlower - 0.0344 × cpRNFLqS - 0.0318 × cpRNFLqIt - 0.0398 × cpRNFLqIn + 1.1646 × log_mGCIPL_STxvsITx + 0.712 × log_mGCC_ITxvsINx

[0039] [Third calculation model] Y (Model 3) = 12.5935 - 0.066 × TSNITlower - 0.0296 × cpRNFLqIn - 0.0289 × mGCC + 0.9845 × log_mGCIPL_SNyvsINy + 0.5671 × log_mGCIPL_STyvsITy + 0.6613 × log_mGCC_ITyvsINy

[0040] In this embodiment, step S11 (creation of a glaucoma risk score calculation model) may be performed at any time before the evaluation of the subject's eye using this calculation model (steps S12 to S15). The created calculation model is stored in a computer that performs the evaluation of the subject's eye and / or in a storage device accessible by this computer. This storage device may be, for example, a node in a computer network to which this computer belongs, such as a local area network (LAN), the Internet, or a wide area network (WAN). Alternatively, this storage device may be a peripheral device directly or indirectly connected to this computer.

[0041] As will be described later in this embodiment with reference to FIG. 2, the calculation model can also be updated using data used to evaluate the subject's eye (corresponding to the OCT data generated in step S12 in this embodiment).

[0042] The exemplary computational model described above is defined as a mathematical formula with only parameters derived based on OCT data as variables, but some other aspects of the computational model may include other parameters as variables, for example, to improve the quality of the calculation (accuracy, precision, reproducibility, etc.). Examples of such additional or auxiliary parameters include: (1) parameters obtained from OCT data other than papilla parameters and macular parameters; (2) parameters related to specific sub-tissues of the retina; (3) parameters related to tissues other than the retina (e.g., parameters related to the choroid, sclera, vitreous, angle, etc.); (4) parameters related to the blood vessels in the fundus (e.g., blood vessel density, etc.); (5) parameters related to the size and / or shape of the area where the lesion or disorder occurs; (6) parameters related to the blood flow in the fundus; (7) parameters related to the subject's background information (e.g., age, sex, race, medical history, treatment history, medication history, family history, etc.); (8) parameters derived from data obtained by ophthalmic modalities other than OCT; (9) parameters related to data obtained in ophthalmic examinations (e.g., intraocular pressure, visual field test parameters, etc.); (10) parameters related to data obtained in examinations related to medical fields other than ophthalmology (e.g., blood pressure, blood test parameters, genetic test parameters, etc.).

[0043] In step S12, an OCT scan is performed on the fundus of the subject's eye to generate OCT data. The OCT scan is performed, for example, by a spectral domain OCT device (OCT scanner).

[0044] The type (e.g., type, format) of the OCT data generated in step S12 is typically determined by the processing content of step S13. For example, this OCT data may include any of data collected by OCT scanning (collected data), image data (OCT image data) obtained by applying imaging processing such as Fourier transform to the collected data, data obtained during the imaging processing (pre-imaging data), and data obtained by applying predetermined processing to OCT image data (processed OCT image data).

[0045] When the collected data is processed to generate OCT data, this processing is executed by a computer operating according to an OCT data generation program. The OCT data generation program may be part of a program that causes a computer to execute the series of processes in steps S12 to S15.

[0046] The number of OCT scans applied to the fundus of the test eye in step S12 may be any number of times greater than or equal to one. In some examples, multiple OCT scans can be applied to the same region of the fundus of the test eye. In this case, based on multiple sets of collected data acquired from the same region, it is possible to generate an average image (such as an arithmetic average image) with reduced random noise, an OCT angiography image (OCTA image) depicting fundus blood vessels, OCT blood flow measurement data showing fundus blood flow dynamics, and the like.

[0047] In some other examples, OCT scans may be performed on two or more different regions of the fundus of the test eye, respectively, to generate separate OCT data for the two or more regions. For example, an OCT scan of a region including the optic disc area (optic disc area scan) and an OCT scan of a region including the macular area (macular area scan) may be performed separately. In this case, OCT data for the optic disc area may be generated based on the optic disc area scan, and OCT data for the macular area may be generated based on the macular area scan.

[0048] Alternatively, a single OCT scan (wide-area scan) may be applied to an area including both the optic disc area and the macular area, in which case OCT data for both the optic disc area and the macular area can be generated based on data collected by the wide-area scan.

[0049] In the next step S13, data to be input into the calculation model created in step S11 is determined based on the OCT data generated in step S12. As described above, the calculation model of this embodiment includes one or more disc parameters and one or more macular parameters as variables. Therefore, in step S13, disc measurement values ​​corresponding to each disc parameter and macular measurement values ​​corresponding to each macular parameter are calculated. Note that measurement values ​​other than these may also be calculated additionally or supplementarily.

[0050] Step S13 is executed by a computer operating according to a measurement value calculation program, which may be part of a program that causes a computer to execute the series of processes from steps S12 to S15.

[0051] In an aspect in which information other than the measurement values ​​calculated from the OCT data (e.g., the above-mentioned additional or auxiliary parameters) is also used, the computer executes a process to acquire the information. For example, the computer may execute a process to access a medical information database (e.g., an electronic medical record system, a medical image archiving system, a medical information sharing system, etc.) to acquire information about the subject.

[0052] In the next step S14, the disc measurements and macular measurements calculated in step S13 are input into the calculation model created in step S11. Each disc measurement is substituted for a corresponding disc parameter (variable), and each macular measurement is substituted for a corresponding macular parameter (variable).

[0053] In some exemplary embodiments, other information is input to the calculation model in addition to the measurement values ​​calculated in step S13. Also, in some exemplary embodiments, the results output from the calculation model based on the input of the measurement values ​​calculated in step S13 are adjusted (corrected) based on the other information.

[0054] Step S14 is executed by a computer operating according to a measurement value input program, which may be part of a program that causes a computer to execute the series of processes from steps S12 to S15.

[0055] The computer obtains a calculation result (glaucoma risk score) based on the calculation model created in step S11 and the measurements (and other information) input into the calculation model in step S14. In step S15, the obtained glaucoma risk score is provided.

[0056] The destination of the glaucoma risk score is predetermined or determined for each process. Examples of the destination include a user, another computer, a medical system, an ophthalmic device, a storage device, etc. In some exemplary embodiments, the glaucoma risk score can be displayed on a display device, transmitted to a doctor's terminal, transmitted to a subject's terminal, transmitted to a server, transmitted to a database, transmitted to an examination device, transmitted to a treatment device, or stored in a storage device.

[0057] Step S15 is executed by a computer operating according to an information provision program, which may be part of a program that causes a computer to execute a series of processes from steps S12 to S15.

[0058] The embodiment shown in FIG. 1 is configured to calculate a glaucoma risk score from papilla measurement values ​​and macular measurement values ​​calculated from OCT data of the fundus of the test eye, and does not require the intraocular pressure of the test eye to evaluate glaucoma. Therefore, this embodiment can identify normal-tension glaucoma. Furthermore, as will be described in detail later, this embodiment can identify glaucoma with high reliability. Therefore, this embodiment provides a glaucoma identification technology suitable for mass screening.

[0059] As previously mentioned, some exemplary embodiments may be configured to update a computational model using data used to evaluate the subject's eye. One such embodiment of a glaucoma screening method (glaucoma identification method) is described with reference to FIG.

[0060] Steps S21 (creating a computational model), S22 (generating OCT data), S23 (calculating disc and macular measurements), and step S24 (inputting measurements into the computational model) may be performed in the same manner as steps S11, S12, S13, and S14, respectively, of Figure 1, although other procedures may also be applied.

[0061] In the next step S25, the computer obtains a glaucoma risk score based on the calculation model created in step S21 and the measurement values ​​input into the calculation model in step S24, and adjusts the coefficients of the calculation model based on this glaucoma risk score.

[0062] Typically, the computational model of this embodiment is defined as a multivariate linear equation with disc parameters and macular parameters as variables. Each term of this multivariate linear equation is the product of a coefficient and a variable. In this embodiment, for example, the development data used in creating the computational model in step S21 is updated by adding the disc measurement values ​​and macular measurement values ​​calculated in step S23. The coefficients of the computational model can be adjusted by applying machine learning using the updated development data to the computational model.

[0063] Step S25 is executed by a computer operating in accordance with a calculation model update program. The calculation model update program may be part of a program that causes a computer to execute a series of processes in steps S22 to S25 (or steps S21 to S25). The calculation model update program may also be the same as the calculation model creation program used in step S21.

[0064] In the next step S26, each of the disc measurements and each of the macular measurements calculated in step S23 are input into the calculation model updated in step S25.

[0065] The computer derives a glaucoma risk score based on the new calculation model obtained in step S25 and the measurements (and other information) input into the new calculation model in step S26. In step S27, the glaucoma risk score thus obtained is provided. The provision of the glaucoma risk score is performed, for example, in the same manner as step S15 in FIG. 1 , although other procedures may also be applied.

[0066] 2 can identify normal-tension glaucoma and can identify glaucoma with high reliability, similar to the embodiment in Fig. 1, and therefore can provide a glaucoma identification technology suitable for mass screening. In addition, this embodiment can use OCT data of the fundus of the test eye that is the target of glaucoma evaluation to update the calculation model, making it possible to improve the quality of the calculation model as needed.

[0067] Since this aspect uses subject data to update the computational model, it is desirable to perform processing to protect the subject's personal information. In some exemplary aspects, items corresponding to personal information (personal information items) are predefined in advance. Examples of personal information items include a subject identifier (subject ID), name, address, official personal identification number, and background information. In the step of processing information for a specific subject, the computer removes information corresponding to each personal information item from all information related to the subject, and uses only the remaining information (including OCT data and / or measurement values) to update the computational model. This improves the level of personal information protection. Such processing is performed by a computer operating in accordance with a personal information protection program. The personal information protection program may be part of the computational model update program described above.

[0068] Exemplary Aspects Having described an overview of the embodiments above, some exemplary aspects of such embodiments will now be described.

[0069] First, the creation of a computational model used in a glaucoma screening method according to one exemplary embodiment will be described with reference to an example performed by the inventors. The flowchart in Figure 3 shows an overview of the procedure for creating a computational model used in the glaucoma screening method according to this embodiment.

[0070] The first step, step S31, is the creation of development data. The development data used in this example included data from 11,487 eyes of 7,572 participants. The average age of the participants was 51.3 years (±10.0 years), with an age range of 35 to 74 years. Each participant underwent a medical history review, fundus photography using a digital fundus camera, visual field testing using an FDT (Frequency Doubling Technology) perimeter, automated OCT measurement using spectral domain OCT, and axial length measurement using an optical axial length measuring device. In other words, each participant's data included medical history data, fundus photography, visual field data, fundus OCT data, and axial length data. Note that, considering the detection of normal-tension glaucoma, the development data in this example did not include intraocular pressure values.

[0071] The next step, S32, is the selection of eligible cases and eligible controls. In the development of the glaucoma risk score performed in this example, a matched case-control design was adopted. Each case was a participant who was confirmed to have glaucoma as a result of a diagnosis by an ophthalmologist based on visual field testing and fundus examination. Each control was a participant who had no history of glaucoma and for whom no signs of glaucoma were found by fundus photography or FDT visual field testing.

[0072] The next step, S33, is matching background conditions between the cases and controls. In this example, one control was randomly selected for each case by matching gender (male or female) and age (five-year range). The final development dataset used for analysis included 284 eyes of 191 cases and 287 eyes of 277 controls.

[0073] The next step S34 is to develop a calculation model using the development data obtained in step S33. Details of the calculation model development method (statistical analysis method) used in this example will be described later with some examples.

[0074] The next step S35 is to create a validation data set. The validation data is data different from the development data. In this example, the development data set was obtained from participants who underwent a health check in 2016, and the validation data set was obtained from participants who underwent a health check in 2018. The validation data used in this example was OCT data from 9,720 eyes. Note that the validation data in this example does not include intraocular pressure values.

[0075] The next step S36 is to verify the computational model for positive predictive value. Note that the positive predictive value is a value defined as the probability that a positive result is a true positive, and in some embodiments, it is defined as the probability that a positive result is actually obtained when the glaucoma risk score is a "high" score, as described below.

[0076] In this example, the computational model obtained in step S34 was applied to each validation data (OCT data for each of the 9,720 eyes). Furthermore, for validation, the score range was divided into three levels: "low," "medium," and "high," and 723 eyes were randomly selected from each score level. The inventors then independently evaluated the OCT reports of each selected eye to determine whether each eye fell into one of four categories: "normal," "glaucoma," "ocular disease other than glaucoma," or "undetermined (further detailed ophthalmic examination required)."

[0077] This step was verified using the following five OCT parameters to determine glaucomatous changes: (1) the mean thickness of the retinal nerve fiber layer (RNFL) on a predetermined circle (a circular area set around the optic disc); (2) the presence or absence of local thinning of the RNFL and the height difference between the double humps in the TSNIT plot; (3) a quadrant RNFL thickness chart and a clock hour RNFL thickness chart; (4) a retinal thickness map, a ganglion cell layer++ thickness map (GCL++ thickness map), and a GCL+ thickness map in the macular area, and an RNFL thickness map in the optic disc area; (5) a GCL++ thickness deviation map (GCL++ thickness deviation map). GCL+ thickness deviation map), GCL+ thickness deviation map, and RNFL thickness deviation map. Here, GCL+ refers to the combined layer of the ganglion cell layer and inner plexiform layer, also called GCIPL. GCL++ refers to the combined layer of the retinal nerve fiber layer, ganglion cell layer, and inner plexiform layer, also called ganglion cell complex (GCC).

[0078] The TSNIT plot is data representing the distribution of RNFL thickness in a predetermined circular region around the optic disc. More specifically, the TSNIT plot is RNFL thickness distribution data expressed using a two-dimensional coordinate system spanned by a first coordinate axis representing a position (angle) in this circular region and a second coordinate axis representing RNFL thickness. The reference position (angle zero) of the first coordinate axis of the TSNIT plot is the temporal side (T), and the positive direction of the first coordinate axis is the rotation direction that passes through the temporal side (T), superior side (S), nasal side (N), and inferior side (I) in that order and returns to the temporal side (T).

[0079] Glaucomatous RNFL defects often present as arcuate defects with temporal borders. For OCT reports where the inventors disagreed on their assessment, the assessment was repeated until all the assessors agreed. All ophthalmologists who participated in the evaluation of the validation data were blinded to the glaucoma scores and performed the evaluation in an environment where they had no access to information other than the OCT reports.

[0080] The next step S37 is to verify the computational model regarding sensitivity and specificity. Here, "sensitivity" is a value defined as the probability that a case that should be determined as positive is correctly determined as positive, and "specificity" is a value defined as the probability that a case that should be determined as negative is correctly determined as negative.

[0081] To verify this step, a follow-up survey was conducted on participants who were identified as suspected of having glaucoma based on fundus photographs (fundus photographs) among those who participated in the development data creation in step S31. In this follow-up survey, 129 eyes of 66 participants were additionally examined using a Humphrey Perimeter (HFA®) using a standard program based on the 24-2 Swedish Interactive Threshold Algorithm. The inventors evaluated the results of the Humphrey Perimeter test and verified the consistency of the results by applying a computational model to the dataset of the 129 eyes of the follow-up survey subjects.

[0082] The verified calculation model according to this example can be obtained by the procedure (steps S31 to S37) of the flowchart in FIG. 3, which has been outlined above.

[0083] Next, the OCT measurement predictors used to create the computational model of this example will be described. The OCT measurement predictors are predictors generated from data obtained by OCT scanning.

[0084] In this example, inner retinal layer thickness was determined using OCT scanning based on a three-dimensional wide scan protocol. The three-dimensional wide scan protocol is a protocol for applying OCT scanning to a wide three-dimensional region including both the retina area and the optic disc area. The inner retina refers to the area from the inner limiting membrane to the outer limiting membrane. Layer thickness distribution data in the retina area and layer thickness distribution data in the optic disc area are derived from the three-dimensional data set obtained by OCT scanning using the three-dimensional wide scan protocol.

[0085] In another example, an OCT scan of a region including the retina area and an OCT scan of a region including the optic disc area may be separately performed, and layer thickness distribution data for the retina area may be generated from the former, and layer thickness distribution data for the optic disc area may be generated from the latter. Note that the OCT scan of the region including the retina area may be an OCT scan of the entire region, or may be a combination of separate OCT scans of two or more partial regions of the region.

[0086] The inventors re-established the segmentation of the optic disc area and the grid of the macular area, and created variables as potential predictors, which were selected from 312 variables defined as shown in Figures 4A, 4B, and 4C.

[0087] In the first step, layer thickness data for the optic disc area (peri-optic disc RNFL (cpRNFL) thickness data) and several types of layer thickness data for the macular area are determined.

[0088] Of the two charts shown in Figure 4A, the left chart 41A shows 12 segments set in the optic disc area. The optic disc area is a circular area (circumferential area) centered on the optic disc center and having a predetermined radius. The optic disc area is divided into 12 segments (arcuate regions) with equal central angles of 30 degrees. The dimensions (diameters) of the optic disc area may be, for example, a predetermined value, or may be a value set or selected depending on the dimensions (diameters) of the optic disc.

[0089] In this example, the mean thickness of the RNFL was calculated for each of 12 segments of the optic disc area based on a three-dimensional data set collected by OCT scanning with a three-dimensional wide-scan protocol.

[0090] Those skilled in the art will understand that the names of the 12 segments of the optic disc area shown in Chart 41A are defined as superior (S), inferior (I), temporal (T), and nasal (N), as well as combinations of these (same below).

[0091] Of the three charts shown in Figure 4B, the left chart 42A shows a grid dividing the macular area into 10 x 10 = 100 segments. The macular area is a rectangular area centered on the fovea and having predetermined dimensions, and is divided into 100 segments (rectangular regions) of equal size. The dimensions (diameters) of the macular area may be, for example, a default value or a value set or selected depending on the subject's eye.

[0092] In this example, for each of 100 segments of the macular area, the mean thickness of the macular RNFL (mRNFL), the mean thickness of the macular ganglion cell layer-inner plexiform layer (mGCIPL), and the mean thickness of the macular ganglion cell complex (mGCC) were calculated based on the three-dimensional data set obtained from the OCT scan using a three-dimensional wide-scan protocol.

[0093] In the second step, variables related to the optic disc area are determined. In this example, five regions (segments) qT, qS, qN, qIn, and qIt shown in the chart 41B on the right side of FIG. 4A were set based on the 12 segments of the optic disc area (cpRNFL). Furthermore, five variables cpRNFLqT, cpRNFLqS, cpRNFLqN, cpRNFLqIn, and cpRNFLqIt corresponding to these five regions qT, qS, qN, qIn, and qIt, respectively, were created.

[0094] Here, cpRNFLqT is the temporal quadrant, cpRNFLqS is the superior quadrant, and cpRNFLqN is the nasal quadrant. In contrast, the inferior quadrant is divided into two regions, qIn and qIt, based on the results of the analysis described below. This division of the inferior quadrant was adopted because the inferotemporal RNFL thickness is considered to be highly useful in detecting glaucoma.

[0095] In the third step, variables related to the macular area are determined. In this example, the above-mentioned three layer thickness distributions (mRNFL thickness distribution, mGCIPL thickness distribution, and mGCC thickness distribution) are first segmented (divided into 10x10 segments) into four regions ST, SN, IT, and IN shown in the chart 42A on the left side of Figure 4B, and variables related to the macular area are defined based on these four regions. The variables in this example are used, for example, in the first example Y (Model 1) of the above-mentioned calculation model.

[0096] Furthermore, variables are defined by excluding the peripheral portion of the macular area from the 100 segments of the macular area (10 × 10 segments). In this example, 64 segments within a range of two segments from the outer edge (periphery) of the macular area are excluded from the 100 segments of the macular area, and the remaining 36 segments are focused on. These 36 segments are then used to define variables corresponding to the four regions STx, SNx, ITx, and INx shown in the central chart 42B of FIG. 4B. The variables in this example are used, for example, in the second example Y (Model 2) of the calculation model described above.

[0097] Furthermore, variables are defined by excluding both the peripheral and central portions of the macular area from the 100 segments of the macular area. In this example, 64 segments within a range of two segments from the outer edge of the macular area and four segments within a range of one segment from the center of the macular area are excluded, and the remaining 32 segments are focused on. These 32 segments are then used to define variables corresponding to the four regions STy, SNy, ITy, and INy shown in chart 42C on the right side of Figure 4B. The variables in this example are used, for example, in the third example Y (Model 3) of the calculation model described above.

[0098] Each variable defined for the macular area is the average thickness (in micrometers) of the corresponding segment. For example, the RNFL thickness (cpRNFLqS) in segment qS shown in chart 41B of FIG. 4A is defined as the average RNFL thickness in the three segments S, ST, and SN in chart 41A corresponding to this segment qS: cpRNFLqS = (cpRNFL_S + cpRNFL_ST + cpRNFL_SN) / 3 (micrometers). Similarly, the GCC thickness (mGCC_SN) in segment SN shown in chart 42A of FIG. 4B is defined as the average GCC thickness in the 25 segments belonging to that segment SN out of 100 segments divided by the grid: mGCC_SN = (mGCC 01_01 + 01_02 + ... + 05_04 + 05_05) / 25 (micrometers).

[0099] In the fourth step, variables representing the difference in layer thickness between two regions are created. In this example, a variable representing the difference between the superior and inferior regions, and a variable representing the difference between the temporal and nasal regions are created. These variables do not follow a normal distribution. In addition, to avoid a calculated value of zero when the layer thicknesses of two regions are equal, these variables are defined as the antilogarithm obtained by adding 1 to the absolute value of the difference in layer thickness values ​​between the two regions and applying a natural logarithm transformation to the antilogarithm. For example, the variable representing the difference in mGCIPL thickness between the two regions ST and IT in FIG. 4B is defined as follows: log_mGCIPL_ST vs IT = log ｅ (|mGCIPL_ST−mGCIPL_IT|+1).

[0100] In the fifth and final step, we defined a variable TSNITlower, which corresponds to the layer thickness corresponding to the lower of the two peaks of a double hump in the TSNIT plot, to account for cases where the difference in cpRNFL thickness between the upper and lower regions is large but the thickness value on the thinner side is not significant. An example of this variable TSNITlower is shown in Figure 4C. Generally, a double hump is a spike near the upper (S) and a spike near the lower (I), and the variable TSNITlower is defined as the smaller of the two layer thickness values ​​at the apex positions of these two spikes.

[0101] Next, the statistical analysis used to create the computational model of this example will be described.

[0102] In this example, all variables obtained from the OCT measurements were treated as continuous variables. In the statistical analysis of this example, the optimal predictive model for the development dataset was selected by backward elimination using a multivariate logistic regression model with a P value of 0.1. This resulted in the three calculation models Y (Model 1), Y (Model 2), and Y (Model 3) described above.

[0103] The first computational model Y (Model 1) includes cpRNFL-related values, mRNFL-related variables, mGCIPL-related variables, and mGCC-related variables as potential predictors, where each macula-related variable is defined for the entire 10x10 macular grid (the entire macular area) (see Chart 42A in Figure 4B). For macular-related variables, the second computational model Y (Model 2) uses variables defined for the macular area excluding the peripheral portion (see Chart 42B), and the third computational model Y (Model 3) uses variables defined for the macular area excluding both the peripheral and central portions (see Chart 42C).

[0104] All three computational models include the average layer thickness over the entire defined region (cpRNFL, mRNFL, mGCIPL, and mGCC), a variable related to the difference in layer thickness between the two regions (e.g., log_mGCIPL_STvsIT), and a variable related to the spikes in the TSNIT plot graph representing the layer thickness distribution around the optic nerve head (cpRNFL) (TSNITlower).

[0105] The validity of the statistical analysis performed in this example was estimated and evaluated by the area under the receiver operating characteristic curve (AUC-ROC).

[0106] In the validation step of this example, the intercepts and betas (slope, regression coefficients) obtained from the three regression models in the development step were applied, and the risk score was calculated using the following two formulas (Formula 1 and Formula 2).

[0107] [Formula 1] log ｅ (p / (1-p))=Y⇔p=exp(Y) / (1+exp(Y)) [Formula 2] Glaucoma screening score=round(p×100) Here, round(X) is a function that performs fractional rounding of X and represents the integer closest to X.

[0108] According to the glaucoma risk score defined in this way, the scores of all participants who underwent fundus OCT measurement fell within the range of 0 to 100. In other words, the range of the glaucoma risk score in this example is 0 to 100. Furthermore, a higher value of the glaucoma risk score in this example indicates a higher possibility of suffering from glaucoma.

[0109] Based on the distribution of scores, the population was classified into three groups: a low (low risk) group with a score ranging from 0 to 49; a middle (moderate risk) group with a score ranging from 50 to 89; and a high (high risk) group with a score ranging from 90 to 100. Furthermore, for each group, the positive predictive value (PPV) for screening for the need for further detailed ophthalmologic examination was calculated. In addition, the diagnostic results obtained from a subset additionally supplemented by Humphrey visual field testing were used to evaluate sensitivity and specificity.

[0110] To clarify the impact of axial length on glaucoma screening in this case, we investigated the effect of information on axial length on improving the risk score. Furthermore, to confirm the accuracy of distinguishing between retinal thinning due to glaucoma and retinal thinning due to high myopia, we estimated axial length based on the angle (orientation) of the double hump of the RNFL around the optic nerve head.

[0111] For this purpose, first, the raw values ​​(measured values) of the axial length were independently added to the third calculation model Y (Model 3), and then the axial length was added to all three calculation models Y (Model 1), Y (Model 2), and Y (Model 3), and the variable selection in the logistic regression described above was performed.

[0112] The results of such statistical analysis are explained below. The characteristics of cases and controls in the development data used in this example are shown in Figure 5.

[0113] In Figure 5, cpRNFL indicates the peripapillary retinal nerve fiber layer, mRNFL indicates the macular retinal nerve fiber layer, mGCIPL indicates the macular ganglion cell layer-inner plexiform layer, and mGCC indicates the macular ganglion cell layer complex (retinal nerve fiber layer-macular ganglion cell layer-inner plexiform layer in the macular area).

[0114] Furthermore, cpRNFLqIn indicates the cpRNFL thickness in the inferior nasal quadrant, cpRNFLqIt indicates the cpRNFL thickness in the inferior temporal quadrant, and cpRNFLqS indicates the cpRNFL thickness in the superior quadrant.

[0115] Furthermore, log_mGCIPL_ITvsIN indicates a logarithmic transformation (first calculation model) in which the difference in mGCIPL thickness between the inferotemporal quadrant and the inferonasal quadrant is taken as the antilogarithm, log_mGCIPL_SNyvsINy indicates a logarithmic transformation (third calculation model) in which the difference in mGCIPL thickness between the superiorotemporal quadrant and the inferiorotemporal quadrant is taken as the antilogarithm, log_mGCIPL_STxvsITx indicates a logarithmic transformation (second calculation model) in which the difference in mGCIPL thickness between the superiorotemporal quadrant and the inferiorotemporal quadrant is taken as the antilogarithm, and log_mGCIPL_STvsIT indicates a logarithmic transformation (second calculation model) in which the difference in mGCIPL thickness between the superiorotemporal quadrant and the inferiorotemporal quadrant is taken as the antilogarithm. log_mGCIPL_STyvsITy indicates a logarithmic transformation (first calculation model) in which the difference in mGCIPL thickness between the superior temporal quadrant and the inferior temporal quadrant is taken as the antilogarithmic number, log_mGCC_ITxvsINx indicates a logarithmic transformation (second calculation model) in which the difference in mGCC thickness between the inferior temporal quadrant and the inferior nasal quadrant is taken as the antilogarithmic number, and log_mGCC_ITyvsINy indicates a logarithmic transformation (third calculation model) in which the difference in mGCC thickness between the inferior temporal quadrant and the inferior nasal quadrant is taken as the antilogarithmic number.

[0116] Furthermore, mGCC_IN indicates the mGCC thickness (first calculation model) in the inferior nasal quadrant, and mGCIPL_IT indicates the mGCIPL thickness in the inferior temporal quadrant.

[0117] Also, TSNITlower indicates the cpRNFL thickness at the spike (peak) on the lower side of the double hump of the TSNIT plot.

[0118] Each value in Figure 5 represents a numerical value (percentage in parentheses) or the mean ± standard deviation. Cases are participants who were confirmed by an ophthalmologist to have glaucoma based on a detailed ophthalmic examination, and controls are participants who were confirmed by an ophthalmologist to have no findings of glaucoma in either fundus photography or FDT visual field testing. P values ​​for age are calculated using a chi-square test, and for continuous variables, they are calculated using a t-test.

[0119] In the case group, the cpRNFL was significantly thinner than in the normal control group, and the overall mean values ​​of the 10 × 10 grid for the mRNFL, mGCIPL, and mGCC were significantly thinner than in the normal control group. The Spearman's rank correlation coefficient for cpRNFL thickness between the left and right eyes was 0.85 (P < 0.01) in the control group and 0.24 (P < 0.01) in the case group.

[0120] Figure 6 shows the distribution of cases and the distribution of controls for each OCT measurement predictor in Figure 5. For each OCT measurement predictor, the histogram with the higher peak corresponds to the distribution of the normal control group, and the histogram with the lower peak corresponds to the distribution of the case group.

[0121] The diagnostic model for glaucoma screening developed in this example is shown in Figure 7. Figure 7 shows information on the intercept and information on retinal thickness-related predictors for each of three calculation models Y (Model 1), Y (Model 2), and Y (Model 3). In Figure 7, B represents the slope (beta, regression coefficient), SE represents the standard error, OR represents the odds ratio, CI represents the confidence interval, and AUC-ROC represents the area under the receiver operating characteristic curve.

[0122] After selecting variables for the logistic regression model, negative regression coefficients were generally obtained for predictors of each region's layer thickness (cpRNFLqS, cpRNFLqIt, cpRNFLqIn, mGCIPL_IT, mGCC, and TSNITlower). This result indicates that if the retinal layer in the selected region is thicker, the likelihood of having glaucoma is low.

[0123] In contrast, positive regression coefficients were obtained for the predictors of the difference between the superior and inferior regions (log_mGCIPL_STvsIT, log_mGCIPL_STxvsITx, log_mGCIPL_SNyvsINy, log_mGCIPL_STyvsITy) and the predictors of the difference in retinal layer thickness between the temporal and nasal regions (log_mGCIPL_ITvsIN, log_mGCC_ITxvsINx, log_mGCC_ITyvsINy). This result indicates that when there is a large difference in retinal layer thickness between the upper and lower and / or left and right, there is a high possibility that the patient is suffering from glaucoma.

[0124] Furthermore, the predicted values ​​of the areas under the receiver operating characteristic curves of the three calculation models Y (Model 1), Y (Model 2), and Y (Model 3) were all approximately 0.97 (see also the three ROC curves shown in Figure 8). This result indicates that there is no significant difference between these three calculation models.

[0125] The inventors applied the glaucoma risk score constructed in this way to a validation dataset (9,720 eyes of 6,006 people). The glaucoma risk score is calculated using the above-mentioned [Equation 1] and [Equation 2]. Specifically, the glaucoma risk score can be calculated using three calculation models Y (Model 1), Y (Model 2), and Y (Model 3).

[0126] The results of the validation of the three computational models with respect to positive predictive value are described with reference to Figure 9. Figure 9 shows the distribution of glaucoma risk scores obtained by the three computational models and the positive predictive value (PPV) for the need for further detailed ophthalmic examination.

[0127] 9 shows the results of classifying the 9,720 risk scores obtained by applying each calculation model (Model 1, Model 2, Model 3) to 9,720 eyes in the validation dataset into the three levels "Low," "Middle," and "High" described above. According to the results, the high-risk group (risk score of 90 or higher) corresponds to 6.1 percent of the population for the first calculation model Y (Model 1), 6.2 percent for the second calculation model Y (Model 2), and 6.2 percent for the third calculation model Y (Model 3).

[0128] To verify the positive predictive value (and negative predictive value), two-thirds of the high-risk group, one-fifth of the medium-risk group, and 2 percent of the low-risk group were randomly selected from all subjects, and four ophthalmologists evaluated each selected subject's OCT report. The results showed that the positive predictive values ​​of the first, second, and third computational models for the high-risk group were 80.7%, 83.3%, and 90.8%, respectively. The negative predictive values ​​for the low-risk group were 87.9%, 88.4%, and 88.2%, respectively. These results indicate that the third computational model Y (Model 3) had the best glaucoma screening accuracy among the three computational models.

[0129] Next, the results of the verification of the third computational model Y (Model 3) regarding sensitivity and specificity will be described with reference to Figures 10 and 11. Figure 10 shows the results of SD-OCT examinations and Humphrey visual field tests performed on participants (129 eyes) who were identified as suspected of having glaucoma by reading fundus photographs among the participants who created the development data, as well as a comparison of these test results with the glaucoma risk score of the third computational model. Figure 11 also shows the sensitivity and specificity of the risk score in glaucoma screening using the third computational model.

[0130] 10 , according to the third calculation model, 67 eyes were found to have a high-risk score (90 or higher). Of these 67 eyes, 53 were diagnosed with glaucoma, 10 were judged to be indiagnosable by OCT and Humphrey visual field testing, and 4 were judged to have no findings (normal).

[0131] Furthermore, as shown in FIG. 11, when subjects with diseases other than glaucoma were excluded, the sensitivity of the risk score based on the third computational model was 85 percent (85.4 percent) and the specificity was 91 percent (91.3 percent).

[0132] In the false-positive group, two cases were suggestive of preperitoneal glaucoma (PPG), and two cases had optic nerve tracking failure during OCT scans. In the false-negative group, two cases had mild retinal nerve fiber layer defect (NFLD), two cases had narrow NFLD, and three cases had diffuse myopic retinal thinning. Two cases were diagnosed with glaucoma, but their scores were 89 and 88, respectively, both of which were very close to the threshold of 90 between high-risk and medium-risk scores.

[0133] The Spearman's rank correlation coefficients between the risk score of the third computational model and the mean deviation and pattern standard deviation of the Humphrey visual field test were −0.503 (P<0.001) and 0.618 (P<0.001). For cases with high-risk scores and confirmed glaucoma, the mean deviation and pattern standard deviation (mean ± standard deviation) were −6.80 ± 5.91 and 6.42 ± 4.64, respectively. For the nine cases with low- or intermediate-risk scores (<90) and confirmed glaucoma (i.e., false-negative cases), the mean deviation and pattern standard deviation (mean ± standard deviation) were −1.96 ± 1.53 and 3.02 ± 0.76, respectively. A t-test confirmed a significant difference between these two groups.

[0134] The inventors also investigated the effect of improving axial length on the risk score. First, the raw value (measured value) of axial length was independently added to the third calculation model. As a result, the odds ratio (95 percent confidence interval) of axial length was 1.30 (95 percent confidence interval: 1.02 to 1.65). However, no improvement in precision (accuracy) was observed.

[0135] Furthermore, the value of axial length was added to the OCT-related variables, and variable selection in logistic regression was performed for the first, second, and third calculation models. As a result, the odds ratios for axial length for the first, second, and third calculation models were 1.52 (95 percent confidence interval: 1.21 to 1.91), 1.56 (95 percent confidence interval: 1.32 to 2.10), and 1.52 (95 percent confidence interval: 1.20 to 1.93), respectively. Even in this case, no improvement in precision (accuracy) was observed.

[0136] In addition, in the variable selection of any of the three computational models, the estimated axial length angle of the double hump exhibited by the peripapillary RNFL was not selected as a significant variable.

[0137] The glaucoma risk score of this embodiment described above is obtained using SD-OCT and can be used for population-based glaucoma mass screening.

[0138] The development of the glaucoma risk score of this embodiment references the diagnostic logic of a glaucoma specialist, which includes the following seven items: (1) the absolute value of each layer thickness in the entire region or the absolute value of each layer thickness in separate regions; (2) the difference in the thickness of the peripapillary RNFL in the vertical direction; (3) the difference in the thickness of each layer in the macular area in the vertical direction and the difference in the thickness of each layer in the macular area in the horizontal direction; (4) the localization of a wedge-shaped optic nerve fiber layer defect (NFLD) around the optic disc; (5) the difference in the double-hump pattern of the peripapillary RNFL thickness distribution (TSNIT plot); (6) the layer thickness value with the lower peak in the double-hump pattern of the peripapillary RNFL thickness distribution (TSNIT plot); and (7) the value of the axial length estimated from the angle (orientation) of the double-hump pattern of the peripapillary RNFL.

[0139] In this embodiment, the following three patterns were considered for the macular area: (A) a first pattern (Chart 42A in Figure 4B, first calculation model) that uses data from the entire macular area; (B) a second pattern (Chart 42B, second calculation model) that uses data from an area excluding the peripheral part of the macular area to avoid the influence of relatively large blood vessels in the fundus; and (C) a third pattern (Chart 42C, third calculation model) that uses data from a donut-shaped area that excludes the peripheral part of the macular area as in the second pattern, but also excludes the central part where abnormalities do not (or rarely occur) in early glaucoma.

[0140] In all of the algorithms of the three calculation models according to this embodiment, the odds ratio of the variables related to the vertical and horizontal differences in each layer is the most influential index of glaucoma. Furthermore, of these three calculation models, the model that exhibits the highest sensitivity and specificity in risk assessment using only SD-OCT data is the third calculation model.

[0141] The validation step of this embodiment involves two steps to consider the following two objectives. The first objective is to develop a score that enables automatic detection of glaucoma findings using OCT, and the second objective is to screen for glaucoma. Therefore, in the first validation, OCT data was interpreted by a glaucoma specialist, and in the second validation, glaucoma diagnosis was evaluated using a Humphrey perimeter. Cases with evenly distributed risk scores were selected, and cases in which it was difficult to diagnose glaucoma were added, taking into account the importance of accurately assessing subtle changes in early glaucoma in mass screening. The validation process of this embodiment, designed in this way, simulates a situation in an actual mass screening (population-based mass screening), in which an ophthalmologist determines the need for further detailed glaucoma examinations at a hospital.

[0142] Additionally, for the purpose of glaucoma screening, in this embodiment, a panel of experts blinded to the risk scores was formed, and in practice, as previously mentioned, all four ophthalmologists made the necessary decisions in consensus.

[0143] To verify the accuracy of the risk score using real data, the dataset used for verification of this embodiment intentionally included cases with low-quality data. Poor image quality can be caused by poor fixation, blinking, eye movement, small pupils, etc. In some of these cases, it was difficult to identify glaucoma findings due to artifacts, segmentation errors, optic nerve tracking errors, etc., but four ophthalmologists reviewed the data until a consensus was reached and made the final decision. The risk score obtained in this manner was highly relevant to the assessment required for detailed glaucoma examinations.

[0144] Furthermore, in this embodiment, the validity (effectiveness) of the risk score was examined using ophthalmological diagnosis based on the Humphrey visual field test and OCT report, and it was confirmed that the risk score exhibited high performance and high accuracy (precision, accuracy) with a sensitivity of 85.4% and a specificity of 91.3%.

[0145] In this embodiment, four cases with high risk scores but no visual field abnormalities were considered to be false positive cases. Some of the false positive cases were determined to be likely to be pre-field glaucoma accompanied by retinal nerve fiber layer defects detected by OCT. Although this embodiment is primarily intended to detect early and intermediate stage glaucoma, it is believed that it can also be used to identify pre-field glaucoma.

[0146] Nine glaucoma cases showed low risk scores and were false negatives. Examination of these false negative cases revealed that risk scores tended to be lower in cases with mild RNFL thinning, narrow retinal nerve fiber layer defects, and diffuse myopic thinning of the retina.

[0147] In recent years, research into algorithm-based artificial intelligence and visualization of deep learning models has been actively conducted, but glaucoma identification technology suitable for glaucoma mass screening has not yet been put into practical use due to problems such as the black-box nature of the algorithm and the complexity of implementation in mass screening. As described above, the algorithm according to this embodiment is capable of detecting glaucoma with high reliability and is also capable of detecting normal-tension glaucoma. In addition, it is possible to provide a glaucoma risk score by installing simple calculation software in an existing OCT device or computer, and therefore it can be said to be suitable for practical use in glaucoma mass screening.

[0148] Furthermore, in this embodiment, the cutoff value (threshold value for dividing the risk score range according to the degree of risk) can be adjusted or set depending on conditions such as the prevalence of the population subject to mass screening. In the above example, the lower limit (cutoff value) of the risk score for determining that further detailed examination is necessary is set to 90 points. The risk score in this example has been shown to provide screening with a high positive predictive value; in fact, 6% of a dataset of approximately 9,000 subjects were subjects with a glaucoma prevalence of 5%.

[0149] Two glaucoma cases in the validation data set used to validate the sensitivity and specificity were determined to be false negatives, but considering that their risk scores were 88 and 89 points, it is possible to improve the sensitivity by adjusting (slightly lowering) the cutoff value. Since the most important characteristic in mass screening is considered to be a high positive predictive value, it can be said that the computational model according to this embodiment has both high feasibility and high reliability.

[0150] As described above, this embodiment provides an OCT screening algorithm suitable for glaucoma screening. Given the nature of glaucoma as an irreversible, progressive disease, it is extremely important to observe changes over time. Therefore, rather than expressing the new findings of this embodiment in terms of setting a clear cutoff value, it is considered more appropriate to express them in terms of a risk score.

[0151] Furthermore, in highly myopic eyes, the axial length elongates, which can cause deformation of the eyeball. OCT images of such eyes may reduce the reliability of data for glaucoma screening. Therefore, as the degree of myopia increases, it becomes difficult to differentiate between glaucoma and myopia. Meanwhile, the distribution of RNFL thickness varies significantly depending on the axial length, regardless of glaucoma. As mentioned above, using the estimated axial length calculated from the angle (orientation) of the double hump in the TSNIT plot did not improve the accuracy of the risk score. Similarly, taking raw axial length data into account did not improve the quality of the risk score.

[0152] This embodiment has several advantages, some examples of which are described below.

[0153] The first strength is that a large number of images to be analyzed are actually read by ophthalmologists, and the resulting data set is used to build a computational model, which is larger than conventional data sets. In other words, a data set that is favorable in terms of both quality and quantity (a high-quality, large-scale data set) is used to build the glaucoma risk score algorithm according to this embodiment.

[0154] The second strength is that all data used to construct the algorithm according to this embodiment is population-based data and is free from hospital bias (e.g., differences in the average severity of disease depending on the size, history, location, etc. of the hospital).

[0155] The third advantage is that the algorithm according to this embodiment does not provide a glaucoma diagnosis result (risk score) as an output from a black box, but can explain the reason for the result from an ophthalmological perspective, which makes it possible to improve the quality of informed consent, for example.

[0156] The fourth strength is that the algorithm according to this aspect was constructed using data from participants who underwent appropriate follow-up at the medical institution to which the inventors belong. For example, the inventors performed final verification of the algorithm according to this aspect based on the results of an examination conducted at an appropriate time after the first consultation (actually, four years after the first consultation), taking into account the progression of glaucoma.

[0157] The main purpose of the glaucoma risk score according to this embodiment is to provide information (decision support information) for a doctor to determine whether a detailed examination at a hospital is necessary for a subject, rather than to provide a final diagnosis of glaucoma. As described above, the glaucoma risk score according to this embodiment can provide a high-quality glaucoma identification technology suitable for mass screening.

[0158] The use of the glaucoma risk score according to this embodiment is not limited to providing the above-mentioned decision support information (typically, glaucoma mass screening). For example, the glaucoma risk score according to this embodiment can be used for the final diagnosis and treatment of glaucoma.

[0159] <Ophthalmic Device> Ophthalmic devices will be described as non-limiting examples of the glaucoma screening method according to the above-described embodiments. Any of the ophthalmic devices according to any of the examples can be at least partially combined with any of the features of the glaucoma screening method according to the embodiments. Furthermore, any of the ophthalmic devices according to any of the examples can be combined with any known technology.

[0160] The ophthalmic device of this embodiment is used, for example, in mass screening for ophthalmic diseases including glaucoma, and generates information indicating the glaucoma risk of the subject eye (glaucoma risk information) by processing fundus OCT data using a computational model.

[0161] The calculation model in this embodiment may be any one of the first calculation model Y (Model 1), the second calculation model Y (Model 2), and the third calculation model Y (Model 3) described above, but is not limited to these.

[0162] In some examples, two or more computational models can be selectively used. The method for selecting a computational model may be arbitrary. For example, by building a system that has a function for recording the type of computational model applied to each subject (each subject's eye, each examinee, each population, etc.) for each examination and a function for notifying the facility where a new examination is performed (or the ophthalmic device used for the new examination) of the type of computational model used in the previous examination when the new examination is performed, the consistency of the examination can be ensured. Furthermore, by building a system that has a function for notifying when a new computational model is introduced and / or when an existing computational model is updated, the quality of the examination can be improved.

[0163] The ophthalmic apparatus of this embodiment may acquire fundus OCT data using any method. In some examples, the ophthalmic apparatus may include an OCT scanner that applies an OCT scan to the fundus of the subject's eye. Furthermore, in some examples, the ophthalmic apparatus may include an element that acquires, from another device, OCT data generated by applying an OCT scan to the fundus of the subject's eye using another ophthalmic apparatus (ophthalmic OCT apparatus) that includes an OCT scanner. This other device may be, for example, an ophthalmic OCT apparatus that generated the OCT data, a medical image archiving system (e.g., a Picture Archiving and Communication System (PACS)) that stores the OCT data, or a recording medium on which the OCT data is recorded. The element that acquires the OCT data from another device may be, for example, a communication device for communicating data with the ophthalmic OCT apparatus or the medical image archiving system, or a drive device for reading the OCT data from a recording medium.

[0164] The glaucoma risk information of this embodiment may be information that expresses any type of glaucoma risk in any form or format. In some examples, the glaucoma risk information may include the glaucoma risk score described above, or may include information indicating the degree of glaucoma risk (e.g., high risk, medium risk, low risk, etc.).

[0165] An example of the configuration of an ophthalmic apparatus according to this embodiment is shown in Fig. 12. The ophthalmic apparatus 120 includes a processor 121, a memory 125, an OCT data acquisition unit 127, and a user interface (UI) 128.

[0166] The processor 121 is configured to execute various processes based on programs installed in the ophthalmic device 120 and / or programs stored in devices accessible by the ophthalmic device 120. The processor 121 can, for example, control each element of the ophthalmic device 120, control peripheral devices of the ophthalmic device 120, and execute various data processes.

[0167] The memory 125 is a storage unit of the ophthalmic apparatus 120. The memory 125 may include any type of storage device, such as a main storage device (primary storage device such as a main memory, cache memory, or working memory), an auxiliary storage device (secondary storage device such as a hard disk drive, flash memory, or RAM disk), a tertiary storage device, offline storage, or a register.

[0168] The memory 125 holds various types of data processed by the ophthalmic device 120. The memory 125 stores a calculation model 126. The calculation model 126 may be any calculation model according to the above-described embodiments. The calculation model 126 may be a linear equation including a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area.

[0169] The computational model 126 is created in advance and stored in the memory 125. For example, the computational model 126 may be any type of computational model created at least in part using the glaucoma screening method according to the above embodiment, and may include at least one of a first computational model Y (Model 1), a second computational model Y (Model 2), and a third computational model Y (Model 3).

[0170] The OCT data acquisition unit 127 acquires OCT data of the fundus of a living eye (examined eye) that is the target of glaucoma screening by the ophthalmologic apparatus 120. The OCT data acquisition unit 127 may include, for example, the OCT scanner described above and / or an element that acquires the OCT data described above from another device.

[0171] The OCT scanner includes, for example, a light source, a splitter that splits the light emitted from the light source into measurement light and reference light, a sample arm that projects the measurement light onto the fundus of the subject's eye and guides return light (reflected light, backscattered light, etc.) of the measurement light from the subject's eye, a reference arm that guides the reference light, a multiplexer that superimposes the return light of the measurement light and the reference light to generate interference light, and a photodetector that detects the interference light and generates an electrical signal (detection signal), as in the conventional OCT scanner. The sample arm is provided with an optical deflector (optical scanner) such as a galvanometer scanner to project the measurement light onto multiple different positions on the fundus of the subject's eye.

[0172] Furthermore, the OCT scanner includes an image construction unit (image construction circuit, image construction circuit configuration) for constructing OCT data (typically, OCT image data) by applying signal processing to the detection signal generated by the photodetector. The image construction unit, for example, generates a reflection intensity profile (A-line profile) for each A-line from a spectral distribution based on data for each measurement light projection position (scan point, A-line), visualizes each A-line profile, and generates multiple A-scan image data sets according to a scan pattern (arrangement of multiple scan points). The signal processing for generating the A-line profile includes noise reduction (denoising), filtering, fast Fourier transform (FFT), and the like.

[0173] The OCT modality used in the OCT scanner may be any, such as spectral domain OCT or swept-source OCT. Spectral domain OCT is an OCT modality that splits light emitted from a low-coherence light source (broadband light source) into measurement light and reference light, projects the measurement light onto an object to be tested, superimposes return light of the measurement light from the object on the reference light to generate interference light, detects the spectral distribution of the generated interference light with a spectroscope, and performs data processing such as Fourier transform on the detected spectral distribution to construct image data.

[0174] On the other hand, swept-source OCT is a method in which light from a wavelength-tunable light source is split into measurement light and reference light, the measurement light is projected onto the object to be tested, the return light of the measurement light from the object to be tested is superimposed on the reference light to generate interference light, the generated interference light is detected by a photodetector, and image data is constructed by performing data processing such as Fourier transform on the detection data collected in response to the wavelength sweep and the scanning of the measurement light.

[0175] In brief, spectral-domain OCT is an OCT modality that acquires the spectral distribution of interference light in a spatially-multiplexed manner, while swept-source OCT is an OCT technique that acquires the spectral distribution of interference light in a time-multiplexed manner. For non-limiting examples of spectral-domain OCT, see, for example, U.S. Pat. No. 9,295,386. For non-limiting examples of swept-source OCT, see, for example, U.S. Pat. No. 1,013,400.

[0176] The user interface 128 is an interface for exchanging information between the ophthalmic apparatus 120 and a user, and includes, for example, an input interface and an output interface. The input interface is an interface for the user to input information to the ophthalmic apparatus 120, and examples of such an interface include a pointing device, a keyboard, a switch, an audio input device, and an image (video) input device. The output interface is an interface for outputting information from the ophthalmic apparatus 120, and examples of such an interface include a display, an audio output device, and a printer.

[0177] The ophthalmic apparatus 120 of this example may include various elements not shown, such as any of the elements described in the above-mentioned documents (U.S. Pat. No. 9,295,386 and U.S. Pat. No. 1,101,3400) or any of the elements described in other documents related to OCT.

[0178] The processor 121 includes a control unit (control circuit, control circuit configuration) 122, a measurement value calculation unit (measurement value calculation circuit, measurement value calculation circuit configuration) 123, and a risk information generation unit (risk information generation circuit, risk information generation circuit configuration) 124.

[0179] The control unit 122 is configured to execute control relating to the measurement value calculation unit 123 , control relating to the risk information generation unit 124 , control relating to the memory 125 , control relating to the OCT data acquisition unit 127 , and control relating to the user interface 128 .

[0180] The control related to the measurement value calculation unit 123 may include, for example, turning on / off the operation of the measurement value calculation unit 123, control related to parameters of the measurement value calculation unit 123, input of data to the measurement value calculation unit 123, and acceptance of output data from the measurement value calculation unit 123. The control related to the risk information generation unit 124 may be similar.

[0181] Control related to memory 125 may include, for example, processing to store (save) data in memory 125, processing to retrieve data from memory 125, processing to process or edit data stored in memory 125, and the like.

[0182] Control of the OCT data acquisition unit 127 may include, for example, control of the light source, control of the optical scanner, control of the optical path length of the sample arm and / or reference arm, control of the focus of the measurement light, control of the polarization state of the measurement light and / or reference light, control of the amount (intensity) of the measurement light and / or reference light, control of the photodetector, and control of a data acquisition system (DAS, DAQ) not shown.

[0183] Control over the user interface 128 may include, for example, control of a display, control of a graphic user interface (GUI), and the like.

[0184] The control unit 122 can execute control of other elements of the ophthalmic apparatus 120. The control may be executed in the same manner as in the past (for example, alignment control, focus control, tracking control, etc.).

[0185] The control unit 122 inputs the OCT data of the fundus of the subject's eye generated by the OCT data acquisition unit 127 to the measurement value calculation unit 123. The measurement value calculation unit 123 is configured to calculate, based on the OCT data, a plurality of measurement values ​​to be input to a calculation model 126 used for glaucoma screening of the subject's eye.

[0186] When the calculation model 126 includes two or more calculation models, for example, the control unit 122 sends information indicating the type of measurement value corresponding to the calculation model used for glaucoma screening of the subject's eye (measurement value type information) to the measurement value calculation unit 123. The measurement value calculation unit 123 recognizes the types of multiple measurement values ​​to be calculated based on this measurement value type information, and calculates each of these measurement values ​​from OCT data of the fundus of the subject's eye.

[0187] As described above, in some examples, the computational model 126 may be a linear equation including a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area. When such a computational model 126 is used for glaucoma screening of the test eye, the multiple measurement values ​​calculated by the measurement value calculation unit 123 include one or more optic disc measurement values ​​corresponding to the one or more optic disc parameters, respectively, and one or more macular measurement values ​​corresponding to the one or more macular parameters, respectively.

[0188] When the calculation model 126 used for glaucoma screening of the test eye is the first calculation model Y (Model 1), the measurement value calculation unit 123 calculates a measurement value corresponding to TSNITlower, a measurement value corresponding to cpRNFLqS, a measurement value corresponding to cpRNFLqIt, a measurement value corresponding to cpRNFLqIn, a measurement value corresponding to mGCIPL_IT, a measurement value corresponding to log_mGCIPL_STvsIT, a measurement value corresponding to log_mGCIPL_ITvsIN, and a measurement value corresponding to mGCC_IN based on OCT data of the fundus of the test eye.

[0189] When the calculation model 126 used for glaucoma screening of the test eye is the second calculation model Y (Model 2), the measurement value calculation unit 123 calculates a measurement value corresponding to TSNITlower, a measurement value corresponding to cpRNFLqS, a measurement value corresponding to cpRNFLqIt, a measurement value corresponding to cpRNFLqIn, a measurement value corresponding to log_mGCIPL_STxvsITx, and a measurement value corresponding to log_mGCC_ITxvsINx based on OCT data of the fundus of the test eye.

[0190] When the calculation model 126 used for glaucoma screening of the test eye is the third calculation model Y (Model 3), the measurement value calculation unit 123 calculates a measurement value corresponding to TSNITlower, a measurement value corresponding to cpRNFLqIn, a measurement value corresponding to mGCC, a measurement value corresponding to log_mGCIPL_SNyvsINy, a measurement value corresponding to log_mGCIPL_STyvsITy, and a measurement value corresponding to log_mGCC_ITyvsINy based on OCT data of the fundus of the test eye.

[0191] Similarly, when a different type of calculation model 126 is used, the measurement value calculation unit 123 obtains measurement values ​​corresponding to each of the multiple parameters (multiple variables) included in the calculation model 126 from the OCT data of the fundus of the test eye.

[0192] The risk information generation unit 124 generates glaucoma risk information for the test eye based on multiple measurement values ​​calculated from OCT data of the fundus of the test eye by the measurement calculation unit 123 and a calculation model 126 used for glaucoma screening of the test eye.

[0193] In some examples, the risk information generation unit 124 substitutes each measurement value calculated by the measurement value calculation unit 123 into a corresponding parameter (variable) of the calculation model 126, and calculates a value (e.g., a glaucoma risk score) based on the calculation model 126. The glaucoma risk information generated and output by the risk information generation unit 124 in this example includes, for example, either or both of the result of a calculation using the calculation model 126 (e.g., a glaucoma risk score) and information obtained by applying a predetermined algorithm (procedure, calculation method) to the result of the calculation (e.g., information indicating the degree of glaucoma risk).

[0194] When the calculation model 126 used for glaucoma screening of the subject's eye is the first calculation model Y (Model 1), the risk information generation unit 124 obtains a glaucoma risk score by substituting a plurality of measurement values ​​into a plurality of parameters (a plurality of variables) of the first calculation model Y (Model 1) as follows: (1) Substitute a measurement value corresponding to TSNITlower into the variable TSNITlower; (2) Substitute a measurement value corresponding to cpRNFLqS into the variable cpRNFLqS; (3) Substitute a measurement value corresponding to cpRNFLqIt into the variable cpRNFLqIt; (4) Substitute a measurement value corresponding to cpRNFLqIn into the variable cpRNFLq (5) Substitute the measurement value corresponding to mGCIPL_IT into the variable mGCIPL_IT; (6) Substitute the measurement value corresponding to log_mGCIPL_STvsIT into the variable log_mGCIPL_STvsIT; (7) Substitute the measurement value corresponding to log_mGCIPL_ITvsIN into the variable log_mGCIPL_ITvsIN; (8) Substitute the measurement value corresponding to mGCC_IN into the variable mGCC_IN. Furthermore, the risk information generator 124 of this example can generate glaucoma risk information including the glaucoma risk score calculated in this manner and / or information indicating the level of glaucoma risk determined from the glaucoma risk score (e.g., high risk, medium risk, or low risk).

[0195] When the calculation model 126 used for glaucoma screening of the subject's eye is a second calculation model Y (Model 2), the risk information generation unit 124 converts a plurality of measurement values ​​into the second calculation model Y (Model 2) as follows: 2) A glaucoma risk score is calculated by substituting the following into multiple parameters (multiple variables): (1) Substitute the measurement value corresponding to TSNITlower into the variable TSNITlower; (2) Substitute the measurement value corresponding to cpRNFLqS into the variable cpRNFLqS; (3) Substitute the measurement value corresponding to cpRNFLqIt into the variable cpRNFLqIt; (4) Substitute the measurement value corresponding to cpRNFLqIn into the variable cpRNFLqIn; (5) Substitute the measurement value corresponding to log_mGCIPL_STxvsITx into the variable log_mGCIPL_STxvsITx; (6) Substitute the measurement value corresponding to log_mGCC_ITxvsINx into the variable log_mGCC_ITxvsINx. Furthermore, the risk information generation unit 124 in this example can generate glaucoma risk information that includes the glaucoma risk score calculated in this manner and / or information indicating the degree of glaucoma risk determined from the glaucoma risk score (e.g., high risk, medium risk, or low risk).

[0196] When the calculation model 126 used for glaucoma screening of the subject's eye is a third calculation model Y (Model 3), the risk information generation unit 124 converts a plurality of measurement values ​​into the third calculation model Y (Model 3) as follows: 3) A glaucoma risk score is calculated by substituting the following into multiple parameters (multiple variables): (1) Substitute the measurement value corresponding to TSNITlower into the variable TSNITlower; (2) Substitute the measurement value corresponding to cpRNFLqIn into the variable cpRNFLqIn; (3) Substitute the measurement value corresponding to mGCC into the variable mGCC; (4) Substitute the measurement value corresponding to log_mGCIPL_SNyvsINy into the variable log_mGCIPL_SNyvsINy; (5) Substitute the measurement value corresponding to log_mGCIPL_STyvsITy into the variable log_mGCIPL_STyvsITy; (6) Substitute the measurement value corresponding to log_mGCC_ITyvsINy into the variable log_mGCC_ITyvsINy. Furthermore, the risk information generation unit 124 in this example can generate glaucoma risk information that includes the glaucoma risk score calculated in this manner and / or information indicating the degree of glaucoma risk determined from the glaucoma risk score (e.g., high risk, medium risk, or low risk).

[0197] Similarly, even when a different type of computational model 126 is used, the risk information generation unit 124 can substitute multiple measurement values ​​calculated by the measurement value calculation unit 123 into multiple variables of the computational model 126 used for glaucoma screening of the test eye in order to generate glaucoma risk information.

[0198] The control unit 122 executes control for providing the glaucoma risk information generated by the risk information generation unit 124. In some examples, the control unit 122 can cause the glaucoma risk information to be displayed on a display of the user interface 128. In some examples, the control unit 122 can control a communication device (not shown) to transmit the glaucoma risk information to another device (e.g., a device, a system, a recording medium, etc.).

[0199] When the computational model 126 includes a plurality of computational models and glaucoma screening of the subject's eye is performed using two or more of these computational models, the measurement value calculation unit 123 calculates two or more measurement value groups corresponding to the two or more computational models from the OCT data of the fundus of the subject's eye, and the risk information generation unit 124 generates glaucoma risk information based on each of the two or more measurement value groups. This allows for the acquisition of two or more pieces of glaucoma risk information corresponding to the two or more computational models used in glaucoma screening of the subject's eye.

[0200] In some examples, the risk information generator 124 may provide two or more pieces of glaucoma risk information.

[0201] In some examples, the risk information generator 124 may select one or more pieces of glaucoma risk information from the two or more pieces of glaucoma risk information and provide the selected one or more pieces of glaucoma risk information.

[0202] In some examples, the risk information generation unit 124 can generate final glaucoma risk information based on the two or more pieces of glaucoma risk information and provide the final glaucoma risk information (and the two or more pieces of glaucoma risk information).

[0203] In some examples, the risk information generator 124 can generate a final glaucoma risk score and / or final information indicating the degree of glaucoma risk based on two or more glaucoma risk scores. The process of generating the final glaucoma risk information from two or more pieces of glaucoma risk information may include, for example, a process of calculating statistical values ​​(e.g., mean, weighted mean, median, standard deviation, variance, etc.) of the two or more glaucoma risk scores, a process of removing outliers from the two or more glaucoma risk scores, etc. The final information indicating the degree of glaucoma risk may be generated from the final glaucoma risk score, for example.

[0204] In some examples, the risk information generator 124 can generate final information indicating the level of glaucoma risk based on a plurality of pieces of information indicating the level of glaucoma risk. This final information generation process may include, for example, a process of selecting one piece of information from the plurality of pieces of information by majority vote, a process of comparing information acquired from the subject's eye (subject) in the current glaucoma screening with the plurality of pieces of information and selecting one piece of information from the plurality of pieces of information, a process of comparing information acquired in a previous examination of the subject's eye (subject) with the plurality of pieces of information and selecting one piece of information from the plurality of pieces of information, etc.

[0205] As described above, the ophthalmic device 120 according to this embodiment uses the glaucoma screening method according to the above embodiment to evaluate the risk of glaucoma in the subject's eye, and is therefore capable of providing high-quality glaucoma identification technology suitable for mass screening.

[0206] <Program and Recording Medium> A program and a recording medium, which are non-limiting examples of the glaucoma screening method according to the above-described embodiment, will be described.

[0207] The program according to this embodiment is a novel program to be executed by a computer. The computer includes a processor and a memory. The processor may have a configuration similar to that of the processor 121 of the exemplary ophthalmic device 120 shown in FIG. 12 . That is, the processor according to this embodiment may include a processor that functions as the control unit 122, a processor that functions as the measurement value calculation unit 123, and a processor that functions as the risk information generation unit 124. Furthermore, the memory according to this embodiment may store a calculation model (126), similar to the memory 125 of the ophthalmic device 120.

[0208] The program according to this embodiment is configured to cause a processor to execute a procedure including a series of steps described below.

[0209] In a first step, the processor executes control to store a computational model in the memory in accordance with the program according to the present aspect, which may be any of the computational models according to the above embodiments, for example, a linear equation including a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area.

[0210] In the second step, the processor executes control for accepting OCT data of the fundus of the subject's eye (OCT data acceptance control) in accordance with the program according to this aspect.

[0211] In some examples, OCT data acceptance control may include a process of acquiring OCT data acquired from the fundus of the test eye by another ophthalmic device (ophthalmic OCT device) including an OCT scanner from the ophthalmic OCT device, a medical image archiving system, or a recording medium.

[0212] In some examples, the computer is a component of an ophthalmic device (e.g., ophthalmic device 120). The ophthalmic device includes an OCT data acquisition unit (e.g., OCT data acquisition unit 127). In this case, the OCT data reception control may include control for causing the OCT data acquisition unit to perform an OCT scan of the fundus of the subject's eye, and control for causing the OCT data acquisition unit to generate OCT data based on data acquired by the OCT scan.

[0213] That is, the OCT data acceptance control performed in the second step may include either control for the processor to accept OCT data generated by another ophthalmic device from outside the computer, or control for the processor to accept OCT data generated by the ophthalmic device itself including the computer.

[0214] In the third step, the processor executes a process (measurement value calculation process) of calculating a plurality of measurement values ​​from the OCT data acquired in the second step in accordance with the program according to this aspect.

[0215] In some examples, the measurement value calculation process calculates, based on the OCT data acquired in the second step, a plurality of measurement values ​​corresponding to the computational model stored in the memory in the first step. When the computational model is a linear equation including a linear combination of one or more disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area, the measurement value calculation process includes a process of calculating, based on the OCT data acquired in the second step, a plurality of measurement values ​​including one or more disc measurement values ​​corresponding to the one or more disc parameters respectively and one or more macular measurement values ​​corresponding to the one or more macular parameters respectively.

[0216] In the fourth step, the processor executes a process of inputting the multiple measurement values ​​calculated in the third step into the calculation model stored in the memory in the first step in accordance with the program according to this aspect.

[0217] If the computational model stored in the memory in the first step is a linear equation including a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and one or more macular parameters representing retinal tissue thickness in the macular area, and if a plurality of measurements including one or more optic disc measurements corresponding to the one or more optic disc parameters and one or more macular measurements corresponding to the one or more macular parameters are calculated in the third step, then in the fourth step, the corresponding optic disc measurements and the corresponding macular measurements are input into each optic disc parameter (variable) and each macular parameter (variable) of the computational model, respectively. This derives a glaucoma risk score for the test eye. The processor can generate glaucoma risk information from the glaucoma risk score.

[0218] In the fifth step, the processor executes a process for providing the glaucoma risk information generated in the fourth step in accordance with the program according to this aspect. That is, in the fifth step, the processor executes a process for providing the calculation results output from the calculation model in response to the input of the multiple measurement values ​​calculated in the third step.

[0219] As such, the program according to this aspect assesses the risk of glaucoma in the subject's eye using the glaucoma screening method according to the above embodiment, and therefore can provide a high-quality glaucoma identification technology suitable for mass screening.

[0220] The program according to this aspect can be recorded on a recording medium. That is, the recording medium according to this aspect is a recording medium on which the program according to this aspect is recorded. The recording medium according to this aspect is a computer-readable non-transitory recording medium. The computer-readable non-transitory recording medium according to this aspect may be a recording medium in any form, for example, a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0221] Any feature related to the glaucoma screening method according to the embodiment and / or any feature related to the ophthalmologic apparatus according to the embodiment may be at least partially combined with the program according to any aspect. Also, any known technology may be combined with the program according to any aspect.

[0222] Similarly, any feature related to the glaucoma screening method according to the embodiment and / or any feature related to the ophthalmologic apparatus according to the embodiment may be at least partially combined with the recording medium according to any aspect. Also, any known technology may be combined with the recording medium according to any aspect.

[0223] <Some Non-Limiting Features> Some non-limiting features of the various embodiments detailed in this disclosure will be described below. Note that the features of the embodiments are not limited to these.

[0224] A first example of a glaucoma screening method according to an embodiment includes a calculation model preparation step, an OCT data generation step, a measurement value calculation step, a measurement value input step, and a calculation result provision step.

[0225] In the computational model preparation step, a computational model is prepared that includes a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area.

[0226] In the OCT data generating step, an optical coherence tomography (OCT) scan is applied to the fundus of the subject's eye to generate OCT data.

[0227] In the measurement value calculation step, a plurality of measurement values ​​are calculated based on the OCT data generated in the OCT data generation step, including one or more papilla measurement values ​​each corresponding to one or more papilla parameters in the computational model prepared in the computational model preparation step, and one or more macular measurement values ​​each corresponding to one or more macular parameters.

[0228] In the measurement value input step, the plurality of measurement values ​​calculated in the measurement value calculation step are input to the calculation model prepared in the calculation model preparation step.

[0229] In the calculation result providing step, calculation results output from the calculation model in response to input of a plurality of measurement values ​​to the calculation model executed in the measurement value input step are provided.

[0230] A second example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the first example aspect: The one or more disc parameters in the computational model prepared in the computational model preparation step include a disc parameter representing a retinal nerve fiber layer thickness in the optic disc area.

[0231] A third example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the second example aspect: the computational model prepared in the computational model preparation step includes a linear combination of two or more head parameters, the two or more head parameters including a first head parameter (TSNITlower) indicating a layer thickness value corresponding to the lower of two peaks of a double hump in a TSNIT plot representing the distribution of retinal nerve fiber layer thickness in a predetermined circular region around the optic nerve head, and a second head parameter (cpRNFLqIn) indicating a mean value of retinal nerve fiber layer thickness in an arc-shaped region in the circular region around the optic nerve head, the arc-shaped region being defined as zero in the temporal direction and a positive direction in the direction rotated from the temporal direction toward the superior direction, and corresponding to an angle ranging from 225 degrees to 285 degrees.

[0232] A fourth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the third example aspect: The computational model prepared in the computational model preparation step includes a linear combination of four optic disc parameters. The four optic disc parameters include a first optic disc parameter (TSNITlower), a second optic disc parameter (cpRNFLqIn), a third optic disc parameter (cpRNFLqS) and a fourth optic disc parameter (cpRNFLqIt). The third optic disc parameter (cpRNFLqS) is a optic disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped region corresponding to the defined angle range of 45 degrees to 135 degrees in the circular region around the optic disc. The fourth disc parameter (cpRNFLqIt) is a disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped area corresponding to the defined angle range of 285 degrees to 315 degrees in the circular area around the optic disc.

[0233] A fifth example aspect of the glaucoma screening method according to the embodiments includes the following non-limiting feature in addition to the non-limiting features of any of the first to fourth example aspects: the computational model prepared in the computational model preparation step includes a linear combination of two or more macular parameters, the two or more macular parameters including a first macular parameter based on a thickness of a composite tissue of ganglion cells and inner plexiform layer (mGCIPL) in the macular area, and a second macular parameter based on a thickness of retinal ganglion cell complexes (mGCCs) in the macular area.

[0234] A sixth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include a layer thickness value parameter (mGCIPL_IT) representing a value of the thickness of the composite tissue (mGCIPL).

[0235] A seventh example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include comparison value parameters (log_mGCIPL_STvsIT, log_mGCIPL_ITvsIN; log_mGCIPL_STxvsITx; log_mGCIPL_SNyvsINy, log_mGCIPL_STyvsITy) representing a comparison value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0236] An eighth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the seventh example aspect: The comparison value represented by the comparison value parameter included in the first macular parameter (parameter related to mGCIPL) is calculated by applying a logarithmic operation to an antilogarithm including a difference value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0237] A ninth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of any of the fifth to eighth example aspects: The second macular parameter (parameter related to mGCC) includes a layer thickness value parameter (mGCC_IN; mGCC) representing a value of the retinal ganglion cell complex (mGCC) thickness.

[0238] A tenth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the fifth to eighth example aspects: The second macular parameter (parameter related to mGCC) includes a comparison value parameter (log_mGCC_ITx vs INx; log_mGCC_ITy vs INy) representing a comparison value between the first retinal ganglion cell complex thickness in the third subarea and the second retinal ganglion cell complex thickness in the fourth subarea of ​​the macular area.

[0239] An eleventh example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the tenth example aspect: The comparison value represented by the comparison value parameter included in the second macular parameter (parameter related to mGCC) is calculated by applying a logarithmic operation to an antilogarithm including a difference value between the first retinal ganglion cell complex thickness in the third sub-area and the second retinal ganglion cell complex thickness in the fourth sub-area of ​​the macular area.

[0240] A twelfth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the first to eleventh examples: In the computational model preparation step, a computational model is created based on a development dataset collected from a first population, a first validation dataset collected from a second population different from the first population, and a second validation dataset collected from a subpopulation of the first population determined to be glaucoma suspects.

[0241] A thirteenth exemplary aspect of the glaucoma screening method according to the embodiments includes the following non-limiting features in addition to the non-limiting features of the twelfth exemplary aspect: The developed dataset includes fundus photographs acquired by digital fundus photography, fundus OCT data acquired by optical coherence tomography, and visual field data acquired by visual field testing.

[0242] A fourteenth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the twelfth or thirteenth example aspects: In the computational model preparation step, the first validation dataset includes fundus OCT data acquired by optical coherence tomography and is used to validate the positive predictive value.

[0243] A fifteenth example aspect of the glaucoma screening method according to the embodiment includes the following non-limiting features in addition to the non-limiting features of any of the twelfth to fourteenth examples: In the computational model preparation step, the second validation dataset includes visual field data acquired in a visual field test, and is used to validate the sensitivity and specificity.

[0244] Any of the first to fifteenth exemplary aspects of the glaucoma screening method according to the embodiment can be at least partially combined with any of the matters described or suggested in the present disclosure.

[0245] A first example of an ophthalmologic apparatus according to an embodiment includes a storage unit, an OCT data acquisition unit, a measurement value calculation unit, and a risk information generation unit.

[0246] The storage unit stores a pre-created computational model. The computational model includes a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area. The memory 125 of the ophthalmic apparatus 120 shown in FIG. 12 is an example of a memory unit according to this embodiment.

[0247] The OCT data acquisition unit is configured to acquire optical coherence tomography data (OCT data) of the fundus of the subject's eye. The OCT data acquisition unit 127 of the ophthalmologic apparatus 120 shown in Fig. 12 is an example of the OCT data acquisition unit according to this embodiment.

[0248] The measurement value calculation unit is configured to calculate, based on the OCT data acquired by the OCT data acquisition unit, a plurality of measurement values ​​including one or more papilla measurement values ​​corresponding to one or more papilla parameters in the calculation model stored in the storage unit and one or more macular measurement values ​​corresponding to one or more macular parameters in the calculation model stored in the storage unit. The measurement value calculation unit 123 of the ophthalmologic apparatus 120 shown in FIG. 12 is an example of a measurement value calculation unit according to this embodiment.

[0249] The risk information generation unit is configured to generate glaucoma risk information for the subject's eye based on the calculation model stored in the storage unit and the multiple measurement values ​​calculated by the measurement value calculation unit. The risk information generation unit 124 of the ophthalmologic apparatus 120 shown in Figure 12 is an example of a risk information generation unit according to this embodiment.

[0250] A second example aspect of the ophthalmic device according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the first example aspect: The one or more disc parameters in the computational model stored in the memory unit include a disc parameter representing a retinal nerve fiber layer thickness in the optic disc area.

[0251] A third example aspect of the ophthalmic device according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the second example aspect: the computational model stored in the memory unit includes a linear combination of two or more head parameters, the two or more head parameters including a first head parameter (TSNITlower) indicating a layer thickness value corresponding to the lower of two peaks of a double hump in a TSNIT plot representing the distribution of retinal nerve fiber layer thickness in a predetermined circular region around the optic nerve head, and a second head parameter (cpRNFLqIn) indicating an average value of retinal nerve fiber layer thickness in an arc-shaped region in the circular region around the optic nerve head, the arc-shaped region corresponding to an angle range of 225 degrees to 285 degrees, the angle being defined as zero in the temporal direction and a positive direction in the direction rotated from the temporal direction toward the superior direction.

[0252] A fourth example aspect of the ophthalmologic apparatus according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the third example aspect: the computational model stored in the memory unit includes a linear combination of four optic disc parameters. The four optic disc parameters include a first optic disc parameter (TSNITlower), a second optic disc parameter (cpRNFLqIn), a third optic disc parameter (cpRNFLqS) and a fourth optic disc parameter (cpRNFLqIt). The third optic disc parameter (cpRNFLqS) is a optic disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped region corresponding to the defined angle range of 45 degrees to 135 degrees in the circular region around the optic disc. The fourth disc parameter (cpRNFLqIt) is a disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped area corresponding to the defined angle range of 285 degrees to 315 degrees in the circular area around the optic disc.

[0253] A fifth example aspect of the ophthalmic device according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the first to fourth examples: the computational model stored in the memory unit includes a linear combination of two or more macular parameters, the two or more macular parameters including a first macular parameter based on a thickness of a composite tissue of ganglion cells and inner plexiform layer (mGCIPL) in the macular area, and a second macular parameter based on a thickness of retinal ganglion cell complexes (mGCCs) in the macular area.

[0254] A sixth example aspect of the ophthalmic device according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include a layer thickness value parameter (mGCIPL_IT) representing a value of the thickness of the composite tissue (mGCIPL).

[0255] A seventh example aspect of the ophthalmic device according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include comparison value parameters (log_mGCIPL_STvsIT, log_mGCIPL_ITvsIN; log_mGCIPL_STxvsITx; log_mGCIPL_SNyvsINy, log_mGCIPL_STyvsITy) representing a comparison value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0256] An eighth example aspect of the ophthalmic device according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the seventh example aspect: The comparison value represented by the comparison value parameter included in the first macular parameter (parameter related to mGCIPL) is calculated by applying a logarithmic operation to an antilogarithm including a value of the difference between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0257] A ninth example aspect of the ophthalmic device according to the embodiment includes the following non-limiting features in addition to the non-limiting features of any of the fifth to eighth examples: The second macular parameters (parameters related to mGCC) include a layer thickness value parameter (mGCC_IN; mGCC) representing a value of the retinal ganglion cell complex (mGCC) thickness.

[0258] A tenth example aspect of the ophthalmic device according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the fifth to eighth example aspects: The second macular parameter (parameter related to mGCC) includes comparison value parameters (log_mGCC_ITx vs INx; log_mGCC_ITy vs INy) representing a comparison value between the first retinal ganglion cell complex thickness in the third subarea and the second retinal ganglion cell complex thickness in the fourth subarea of ​​the macular area.

[0259] An eleventh example aspect of the ophthalmic apparatus according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the tenth example aspect: The comparison value represented by the comparison value parameter included in the second macular parameter (parameter related to mGCC) is calculated by applying a logarithmic operation to an antilogarithm including a value of the difference between the first retinal ganglion cell complex thickness in the third sub-area and the second retinal ganglion cell complex thickness in the fourth sub-area of ​​the macular area.

[0260] A twelfth example aspect of the ophthalmic apparatus according to the embodiment includes the following non-limiting features in addition to the non-limiting features of any of the first to eleventh examples. The computational model according to this example aspect is created by a computational model creation device. The computational model creation device may be a component of the ophthalmic apparatus according to this example aspect, or may be a device separate from the ophthalmic apparatus according to this example aspect.

[0261] To create a computational model according to this example embodiment, a development data set, a first validation data set, and a second validation data set are prepared.

[0262] The development dataset is a dataset collected from a first population, i.e., the development dataset is collected by administering one or more predetermined medical tests (including ophthalmic tests) to each subject in the first population.

[0263] The first validation dataset is a dataset collected from a second population different from the first population corresponding to the development dataset. That is, the first validation dataset is a dataset collected by administering one or more predetermined medical tests (including ophthalmic tests) to each subject included in the second population. The one or more medical tests administered for the first validation dataset and the one or more medical tests administered for the development dataset may have at least one test in common, or may not have any common test.

[0264] The second validation dataset is a dataset collected from a subpopulation determined to be glaucoma suspects in the first population corresponding to the development dataset. The subpopulation is a group consisting of subjects who have been determined (diagnosed) to be glaucoma suspects among multiple subjects belonging to the first population that was the subject of the development dataset. It is clear from the definition that the subpopulation is a part (subset) of the first population.

[0265] The development data set, the first validation data set, and the second validation data set are input to a computational model creation device, which creates a computational model based on these data sets and a predetermined algorithm.

[0266] The algorithm used to create the computational model may include a machine learning algorithm. This machine learning technique is typically, but is not limited to, supervised learning. In some examples, in the machine learning for constructing the computational model, any known technique such as unsupervised learning, reinforcement learning, semi-supervised learning, transduction, multi-task learning, etc. may be used alone or in combination in addition to or instead of supervised learning.

[0267] The techniques used to construct the computational model are not limited to the examples shown here, and may include any known techniques, such as support vector machines, Bayesian classifiers, boosting, k-means, kernel density estimation, principal component analysis, independent component analysis, self-organizing maps, random forests, and generative adversarial networks (GANs).

[0268] A thirteenth example aspect of the ophthalmologic apparatus according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the twelfth example aspect: The development dataset input to the computational model creation device includes fundus photographs acquired by digital fundus photography, fundus OCT data acquired by optical coherence tomography, and visual field data acquired by visual field testing.

[0269] A fourteenth example aspect of the ophthalmologic apparatus according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the twelfth or thirteenth example aspects: the first validation data set input to the computational model creation device includes fundus OCT data acquired by optical coherence tomography, and the computational model creation device performs verification of the positive predictive value based on the input first validation data set.

[0270] A fifteenth example aspect of the ophthalmologic apparatus according to the embodiment includes the following non-limiting features in addition to the non-limiting features of any of the twelfth to fourteenth examples: The second validation data set input to the computational model creation device includes visual field data acquired in a visual field test. The computational model creation device performs verification of sensitivity and specificity based on the input second validation data set.

[0271] Any of the first to fifteenth aspects of the ophthalmic apparatus according to the embodiment may be at least partially combined with any of the features described or suggested in this disclosure.

[0272] A first example of a program according to an embodiment is a program to be executed by a computer including a processor and a memory, and is configured to cause the processor to execute calculation model storage control, OCT data acceptance control, measurement value calculation processing, measurement value input processing, and calculation result provision processing.

[0273] In the computational model storage control, the program causes the processor to execute a process of storing in memory a computational model including a linear combination of one or more optic disc parameters representing retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representing retinal tissue thickness in the macular area.

[0274] In the OCT data reception control, the program causes the processor to execute a process of receiving optical coherence tomography data (OCT data) of the fundus of the subject's eye.

[0275] In the measurement value calculation process, the program causes the processor to perform a process of calculating multiple measurement values, including one or more papilla measurement values ​​each corresponding to one or more papilla parameters in the calculation model stored in memory by the calculation model storage control, and one or more macular measurement values ​​each corresponding to one or more macular parameters, based on the OCT data received in the OCT data reception process.

[0276] In the measurement value input process, the program causes the processor to execute a process of inputting the plurality of measurement values ​​calculated in the measurement value calculation process into the calculation model stored in the memory in the calculation model storage control.

[0277] In the calculation result providing process, the program causes the processor to execute a process of providing calculation results output from the calculation model in response to input of a plurality of measurement values ​​to the calculation model executed in the measurement value input process.

[0278] A first example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the first example of the embodiment recorded thereon.

[0279] A second example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the first example aspect: The one or more disc parameters in the computational model stored in the memory in the computational model storage control include a disc parameter representing a retinal nerve fiber layer thickness in the optic nerve head area.

[0280] A second example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium on which a program according to the second example of the embodiment is recorded.

[0281] A third example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the second example aspect: the computational model stored in the memory by the computational model storage control includes a linear combination of two or more optic disc parameters, the two or more head parameters including a first optic disc parameter (TSNITlower) indicating a layer thickness value corresponding to the lower of two peaks of a double hump in a TSNIT plot representing the distribution of retinal nerve fiber layer thickness in a predetermined circular region around the optic disc, and a second optic disc parameter (cpRNFLqIn) indicating an average value of retinal nerve fiber layer thickness in an arc-shaped region in the circular region around the optic disc, the arc-shaped region corresponding to an angle range of 225 degrees to 285 degrees, where the angle is defined as zero in the temporal direction and the positive direction is the direction rotated from the temporal direction toward the superior direction.

[0282] A third example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the third example of the embodiment recorded thereon.

[0283] A fourth example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the third example aspect: the computational model stored in the memory by the computational model storage control includes a linear combination of four optic disc parameters. The four optic disc parameters include a first optic disc parameter (TSNITlower), a second optic disc parameter (cpRNFLqIn), as well as a third optic disc parameter (cpRNFLqS) and a fourth optic disc parameter (cpRNFLqIt). The third optic disc parameter (cpRNFLqS) is a optic disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped region corresponding to the defined angle range of 45 degrees to 135 degrees in the circular region around the optic disc. The fourth disc parameter (cpRNFLqIt) is a disc parameter that represents the average value of the retinal nerve fiber layer thickness in an arc-shaped area corresponding to the defined angle range of 285 degrees to 315 degrees in the circular area around the optic disc.

[0284] A fourth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the fourth example of the embodiment recorded thereon.

[0285] A fifth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the first to fourth example aspects: the computational model stored in the memory in the computational model storage control includes a linear combination of two or more macular parameters, the two or more macular parameters including a first macular parameter based on a thickness of a composite tissue of ganglion cells and inner plexiform layer (mGCIPL) in the macular area, and a second macular parameter based on a thickness of retinal ganglion cell complexes (mGCCs) in the macular area.

[0286] A fifth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium on which a program according to the fifth example of the embodiment is recorded.

[0287] A sixth example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include a layer thickness value parameter (mGCIPL_IT) representing a value of the thickness of the composite tissue (mGCIPL).

[0288] A sixth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the sixth example of the embodiment recorded thereon.

[0289] A seventh example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the fifth example aspect: The first macular parameters (parameters related to mGCIPL) include comparison value parameters (log_mGCIPL_STvsIT, log_mGCIPL_ITvsIN; log_mGCIPL_STxvsITx; log_mGCIPL_SNyvsINy, log_mGCIPL_STyvsITy) representing a comparison value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0290] A seventh example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the seventh example of the embodiment recorded thereon.

[0291] An eighth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the seventh example aspect: The comparison value represented by the comparison value parameter included in the first macular parameter (parameter related to mGCIPL) is calculated by applying a logarithmic operation to an antilogarithm including a difference value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area.

[0292] An eighth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the eighth example of the embodiment recorded thereon.

[0293] A ninth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the fifth to eighth example aspects: The second macular parameter (parameter related to mGCC) includes a layer thickness value parameter (mGCC_IN; mGCC) representing a value of a retinal ganglion cell complex (mGCC) thickness.

[0294] A ninth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the ninth example of the embodiment recorded thereon.

[0295] A tenth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the fifth to eighth example aspects: The second macular parameter (parameter related to mGCC) includes comparison value parameters (log_mGCC_ITx vs INx; log_mGCC_ITy vs INy) representing a comparison value between the first retinal ganglion cell complex thickness in the third subarea and the second retinal ganglion cell complex thickness in the fourth subarea of ​​the macular area.

[0296] A tenth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the tenth example of the embodiment recorded thereon.

[0297] An eleventh example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the tenth example aspect: The comparison value represented by the comparison value parameter included in the second macular parameter (parameter related to mGCC) is calculated by applying a logarithmic operation to an antilogarithm including a value of the difference between the first retinal ganglion cell complex thickness in the third sub-area and the second retinal ganglion cell complex thickness in the fourth sub-area of ​​the macular area.

[0298] An eleventh example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the eleventh example of the embodiment recorded thereon.

[0299] A twelfth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of any of the first to eleventh examples: The program according to this example aspect is configured to cause the processor to further execute a calculation model creation process that creates a calculation model to be stored in the memory by calculation model storage control.

[0300] In the computational model creation process, the program causes the processor to execute a process of creating a computational model based on a development dataset collected from a first population, a first validation dataset collected from a second population different from the first population, and a second validation dataset collected from a subpopulation of the first population that has been determined to be suspected of having glaucoma.

[0301] A twelfth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the twelfth example of the embodiment recorded thereon.

[0302] A thirteenth example aspect of the program according to the embodiment includes the following non-limiting features in addition to the non-limiting features of the twelfth example aspect: The development dataset input to the processor includes fundus photographs acquired by digital fundus photography, fundus OCT data acquired by optical coherence tomography, and visual field data acquired by visual field testing.

[0303] A thirteenth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the thirteenth example of the embodiment recorded thereon.

[0304] A fourteenth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting features of the twelfth or thirteenth example aspects: the first validation dataset input to the processor includes fundus OCT data acquired by optical coherence tomography, and the processor executes verification of the positive predictive value based on the input first validation dataset in accordance with the program.

[0305] A fourteenth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the fourteenth example of the embodiment recorded thereon.

[0306] A fifteenth example aspect of the program according to the embodiment includes the following non-limiting feature in addition to the non-limiting feature of any of the twelfth to fourteenth examples: the second validation dataset input to the processor includes visual field data acquired in a visual field test, and the processor executes validation of sensitivity and specificity based on the input second validation dataset in accordance with the program.

[0307] A fifteenth example of the recording medium according to the embodiment is a computer-readable non-transitory recording medium having a program according to the fifteenth example of the embodiment recorded thereon.

[0308] Any of the first to fifteenth example aspects of the program according to the embodiment can be at least partially combined with any of the features described or suggested in this disclosure.

[0309] Any of the first to fifteenth exemplary aspects of the recording medium according to the embodiment can be at least partially combined with any of the features described or suggested in this disclosure.

[0310] This disclosure presents several embodiments and several aspects thereof. These embodiments and aspects are merely non-limiting examples of the present invention. Therefore, any modifications (omissions, substitutions, additions, etc.) within the scope of the spirit of the present invention can be made to the embodiments and aspects presented in this disclosure.

[0311] 120 Ophthalmic apparatus 121 Processor 122 Control unit 123 Measurement value calculation unit 124 Risk information generation unit 125 Memory 126 Calculation model 127 OCT data acquisition unit 128 User interface

Claims

1. A memory unit that stores a pre-created computational model including a linear combination of one or more optic disc parameters representing a retinal tissue thickness in an optic disc area and a linear combination of one or more macular parameters representing a retinal tissue thickness in a macular area; an OCT data acquisition unit that acquires optical coherence tomography (OCT) data of the fundus of the subject's eye; a measurement calculation unit that calculates a plurality of measurements based on the OCT data, the measurement values ​​including one or more optic disc measurements corresponding to the one or more optic disc parameters and one or more macular measurements corresponding to the one or more macular parameters; a risk information generating unit that generates glaucoma risk information of the subject's eye based on the calculation model and the plurality of measurement values; Including, Ophthalmology equipment.

2. The one or more optic disc parameters in the computational model stored in the memory unit include an optic disc parameter representing a retinal nerve fiber layer thickness in the optic disc area. The ophthalmic device of claim 1.

3. the computational model includes a linear combination of two or more nipple parameters; The two or more nipple parameters are A first optic disc parameter indicating a layer thickness value corresponding to the lower peak of two peaks of a double hump exhibited by a TSNIT plot representing a distribution of retinal nerve fiber layer thickness in a predetermined circular area around the optic disc; a second papilla parameter representing an average value of the retinal nerve fiber layer thickness in an arc-shaped region corresponding to an angle range of 225 degrees to 285 degrees, the angle being defined as zero in the temporal direction and a positive direction in a direction rotating upward from the temporal direction in the circular region; Including, The ophthalmic device of claim 2.

4. The computational model includes a linear combination of four nipple parameters, The four nipple parameters are: The first nipple parameter; The second nipple parameter; a third papilla parameter representing an average value of the retinal nerve fiber layer thickness in an arcuate region corresponding to a range of the angle from 45 degrees to 135 degrees in the circular region; a fourth papilla parameter representing an average value of the retinal nerve fiber layer thickness in an arcuate region corresponding to a range of the angle from 285 degrees to 315 degrees in the circular region; Including, The ophthalmic apparatus of claim 3.

5. the computational model includes a linear combination of two or more macular parameters; The two or more macular parameters are a first macular parameter based on a combined tissue thickness of the ganglion cells and the inner plexiform layer in the macular area; a second macular parameter based on a retinal ganglion cell complex thickness in the macular area; Including, The ophthalmic apparatus according to any one of claims 1 to 4.

6. The first macular parameter includes a layer thickness value parameter representing a value of the composite tissue thickness.

6. An ophthalmic device according to claim 5.

7. the first macular parameter includes a comparison value parameter representing a comparison value between a first composite tissue thickness in a first sub-area and a second composite tissue thickness in a second sub-area of ​​the macular area; 6. An ophthalmic device according to claim 5.

8. The comparison value is calculated by applying a logarithmic operation to an antilogarithm comprising a difference value between the first composite tissue thickness and the second composite tissue thickness.

8. The ophthalmic device of claim 7.

9. The second macular parameter includes a layer thickness value parameter representing a value of the retinal ganglion cell complex thickness.

6. An ophthalmic device according to claim 5.

10. the second macular parameter includes a comparison value parameter representing a comparison value between a first retinal ganglion cell complex thickness in a third sub-area and a second retinal ganglion cell complex thickness in a fourth sub-area of ​​the macular area; 6. An ophthalmic device according to claim 5.

11. the comparison value is calculated by applying a logarithmic operation to an antilog containing a difference value between the first retinal ganglion cell complex thickness and the second retinal ganglion cell complex thickness. The ophthalmic device of claim 10.

12. The computational model is created based on a development dataset collected from a first population, a first validation dataset collected from a second population different from the first population, and a second validation dataset collected from a subpopulation of the first population determined to be suspected of having glaucoma. The ophthalmic apparatus according to any one of claims 1 to 4.

13. Further comprising a computational model creation device that creates the computational model based on the development dataset, the first validation dataset, and the second validation dataset. The ophthalmic device of claim 12.

14. The development data set includes fundus photographs acquired by digital fundus photography, fundus OCT data acquired by optical coherence tomography, and visual field data acquired by visual field testing; The ophthalmic device of claim 12.

15. The first validation data set includes fundus OCT data acquired by optical coherence tomography and is used to validate the positive predictive value; The ophthalmic device of claim 12.

16. The second validation data set includes visual field data acquired in a visual field test and is used to validate sensitivity and specificity. The ophthalmic device of claim 12.

17. A program for causing a computer including a processor and a memory to execute the program, The processor, storing in the memory a computational model including a linear combination of one or more optic disc parameters representative of retinal tissue thickness in the optic disc area and a linear combination of one or more macular parameters representative of retinal tissue thickness in the macular area; A control that receives optical coherence tomography (OCT) data of a fundus of the test eye; calculating a plurality of measurements based on the OCT data, the measurements including one or more papilla measurements each corresponding to the one or more papilla parameters and one or more macular measurements each corresponding to the one or more macular parameters; inputting the measurements into the computational model; providing a calculation result output from the calculation model in response to the input of the plurality of measured values; Execute the program.

18. A non-transitory computer-readable recording medium having the program of claim 17 recorded thereon.