WO2023195554A1

WO2023195554A1 - Apparatus and method for diagnosing risk level of glaucoma

Info

Publication number: WO2023195554A1
Application number: PCT/KR2022/004919
Authority: WO
Inventors: 김성재; 오세종; 조경진
Original assignee: 경상국립대학교병원
Priority date: 2022-04-06
Filing date: 2022-04-06
Publication date: 2023-10-12

Abstract

Disclosed are an apparatus and method for diagnosing a risk level of glaucoma. The apparatus for diagnosing a risk level of glaucoma, according to an aspect, comprises: a data input unit which receives clinical data of a subject; a pre-processing unit which pre-processes the clinical data; and an index calculation unit which extracts a certain number of cases most similar to the clinical data or the pre-processed clinical data from a reference database, calculates a ratio of the number of cases of glaucoma to the extracted certain number of cases, and calculates an integrated glaucoma risk index on the basis of the pre-processed clinical data and the calculated ratio, wherein the clinical data includes an upper retinal nerve fiber layer thickness, a lower retinal nerve fiber layer thickness, a temporal retinal nerve fiber layer thickness, an intraocular pressure, and a pattern standard deviation and an average deviation obtained through visual field tests.

Description

Glaucoma risk diagnosis device and method

It is related to technology for calculating glaucoma risk index from ophthalmic examination data and technology for using it for glaucoma diagnosis.

Glaucoma is a leading cause of irreversible blindness worldwide and progressively affects the optic nerve. The number of glaucoma patients is rapidly increasing every year, and this rapid increase is a global trend. Therefore, accurate diagnosis of glaucoma is an important issue. Recently, machine learning techniques have been widely used to diagnose glaucoma.

Meanwhile, monitoring the possibility or progression of glaucoma in subjects is as important as diagnosing glaucoma. However, although machine learning-based classification models can determine whether a subject has glaucoma, they cannot determine the risk or progression of glaucoma for the subject.

Therefore, there is a need to develop technology that can more accurately determine the risk or progression of glaucoma.

The purpose is to provide a glaucoma risk diagnosis device and method that calculates an integrated glaucoma risk index indicating the risk or progression of glaucoma.

A glaucoma risk diagnosis device according to one aspect includes a data input unit that receives clinical data of a subject; a preprocessing unit that preprocesses the clinical data; and extracting a predetermined number of cases most similar to the clinical data or the preprocessed clinical data from a reference database, calculating a ratio of the number of cases with glaucoma among the extracted cases to the preprocessed clinical data, and calculating the ratio of the number of cases with glaucoma to the preprocessed clinical data. an index calculation unit that calculates an integrated glaucoma risk index based on the data and the calculated ratio; Including, the clinical data may include the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, intraocular pressure, and pattern standard deviation and average deviation obtained through visual field testing.

The preprocessor may normalize the clinical data to a value between 0 and 1, and convert the normalized value of some of the clinical data into an inverted value by subtracting the normalized value from 1.

The preprocessor may subtract each of the normalized values of the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, and the average deviation from 1 to convert them into inverted values.

The predetermined number may be 5.

The index calculation unit may calculate the integrated glaucoma risk index using the following equation.

[Equation]

(I-GRI represents the integrated glaucoma risk index, NNI represents the ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, F _i represents the preprocessed value of each clinical data, a _i represents the weight applied to each clinical data, m and n are NNI and

represents the weight applied to, m+n=1)

The a _i may be the importance of each clinical data derived from a machine learning-based glaucoma diagnosis model created by learning clinical data for learning.

The glaucoma risk diagnosis device includes a diagnosis unit that diagnoses the risk of glaucoma for the subject based on an integrated glaucoma risk index; It may further include.

The diagnostic unit may diagnose that the greater the integrated glaucoma risk index, the higher the risk of glaucoma.

The diagnosis unit may generate a graphic chart representing the integrated glaucoma risk index or glaucoma risk diagnosis results.

The integrated glaucoma risk index has a value between 0 and 1 and may reflect the risk or progression of glaucoma.

A glaucoma risk diagnosis method according to another aspect includes receiving clinical data of a subject; Preprocessing the clinical data; extracting a predetermined number of cases most similar to the clinical data or the preprocessed clinical data from a reference database; and calculating a ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, and calculating an integrated glaucoma risk index based on the preprocessed clinical data and the calculated ratio. Including, the clinical data may include the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, intraocular pressure, and pattern standard deviation and average deviation obtained through visual field testing.

Preprocessing the clinical data includes normalizing the clinical data to a value between 0 and 1; and subtracting normalized values of some of the clinical data from 1 and converting them into inverted values; may include.

In the step of converting the clinical data, each of the normalized values of the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, and the average deviation can be subtracted from 1 and converted into inverted values. .

The predetermined number may be 5.

In the step of calculating the integrated glaucoma risk index, the integrated glaucoma risk index can be calculated using the following equation.

[Equation]

represents the weight applied to, m+n=1)

The glaucoma risk diagnosis method includes: diagnosing the glaucoma risk of the subject based on an integrated glaucoma risk index; It may further include.

In the step of diagnosing glaucoma, the greater the integrated glaucoma risk index, the higher the risk of glaucoma.

The glaucoma risk diagnosis method includes generating a graphic chart representing the integrated glaucoma risk index or glaucoma risk diagnosis results; It may further include.

By calculating and using an integrated glaucoma risk index that reflects the risk or progression of glaucoma, the risk or progression of glaucoma can be more accurately determined.

1 is a diagram illustrating a glaucoma risk diagnosis device according to an exemplary embodiment.

Figure 2 is an example diagram showing the NNI distribution of normal and glaucoma cases.

Figure 3 is an example diagram to explain the effect of applying NNI in calculating the integrated glaucoma risk index.

Fig. 4 is a diagram illustrating a graphic chart according to an exemplary embodiment.

Fig. 5 is a diagram illustrating a graphic chart according to an exemplary embodiment.

Figure 6 is a diagram illustrating a glaucoma risk diagnosis device according to an exemplary embodiment.

Figure 7 is a diagram illustrating a glaucoma risk diagnosis method according to an exemplary embodiment.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the attached drawings. When adding reference numerals to components in each drawing, it should be noted that identical components are given the same reference numerals as much as possible even if they are shown in different drawings. Additionally, in describing the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

Meanwhile, in each step, unless a specific order is clearly stated in the context, each step may occur in a different order from the specified order. That is, each step may be performed in the same order as specified, may be performed substantially simultaneously, or may be performed in the opposite order.

The terms described below are terms defined in consideration of functions in the present invention, and may vary depending on the intention or custom of the user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Terms such as first, second, etc. may be used to describe various components, but the components should not be limited by the terms. Terms are used only to distinguish one component from another. Singular expressions include plural expressions unless the context clearly indicates otherwise, and terms such as 'include' or 'have' refer to the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. It is intended to specify that something exists, but it should be understood as not precluding the possibility of the existence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

In addition, the division of components in this specification is merely a division according to the main function each component is responsible for. That is, two or more components may be combined into one component, or one component may be divided into two or more components for more detailed functions. In addition to the main functions that each component is responsible for, each component may additionally perform some or all of the functions that other components are responsible for, and some of the main functions that each component is responsible for may be performed by other components. It may also be carried out. Each component may be implemented as hardware (eg, memory, processor, etc.) or software, or may be implemented as a combination of hardware and software.

The glaucoma risk diagnosis device 100 according to an exemplary embodiment is capable of calculating an integrated glaucoma risk index based on the subject's clinical data and diagnosing the subject's risk or progression of glaucoma based on the calculated integrated glaucoma risk index. As a device, it may be mounted on an electronic device or implemented as a separate device. Here, electronic devices may include cart-type devices and portable devices, and portable devices include mobile phones, smartphones, tablets, laptops, PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), navigation devices, MP3 players, and digital devices. It may include cameras, wearable devices, etc. Wearable devices include glasses type, wrist watch type, wrist band type, ring type, earring type, belt type, necklace type, ankle band type, thigh band type, forearm band type, etc. can do. However, portable devices are not limited to the above-described examples, and wearable devices are also not limited to the above-described examples.

Referring to FIG. 1, the glaucoma risk diagnosis device 100 according to an exemplary embodiment includes a data input unit 110 and a processor 120, and the processor 120 includes a preprocessor 121 and an index calculation unit 122. ) and a diagnostic unit 123.

The data input unit 110 may receive the subject's clinical data used to calculate the integrated glaucoma risk index or determine the risk/progress of glaucoma from the user. Here, clinical data are closely related to the progression of glaucoma obtained through visual field tests, retinal nerve fiber layer (RNFL) optical coherence tomography (OCT) tests, and intraocular pressure tests. It may be a characteristic or diagnostic indicator. For example, clinical data includes pattern standard deviation (PSD), mean deviation (MD), and superior retinal nerve fiber layer (RNFL) thickness ( Hereinafter, RNFL_S), retinal nerve fiber layer inferior thickness (hereinafter, RNFL_I), retinal nerve fiber layer temporal thickness (hereinafter, RNFL_T), and intraocular pressure (hereinafter, IOP) may be included. Here, PSD and MD can be obtained through visual field testing, RNFL_S, RNFL_I, and RNFL_T can be obtained through retinal nerve fiber layer optical coherence tomography testing, and IOP can be obtained through intraocular pressure testing.

According to an exemplary embodiment, the data input unit 110 includes a key pad, a dome switch, a mouse, a touch pad, a jog wheel, a jog switch, It may include hardware or software buttons, etc. In particular, when the touch pad forms a layered structure with the display, it can be called a touch screen.

The preprocessing unit 121 may preprocess the subject's clinical data.

According to an exemplary embodiment, the preprocessor 121 may normalize the subject's clinical data and convert some clinical data values of the normalized clinical data into reverse values.

For example, the preprocessor 121 may normalize the subject's clinical data, such as PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP, to values between 0 and 1 through a min-max normalization technique. Additionally, the preprocessor 121 may subtract the normalized values of some of the normalized clinical data, such as MD, RNFL_S, RNFL_I, and RNFL_T, from 1 and invert them.

The smaller the value of PSD and IOP, the smaller the risk or progression of glaucoma, while the smaller the value of MD, RNFL_S, RNFL_I, and RNFL_T, the greater the risk or progression of glaucoma. That is, PSD and IOP are positively correlated with glaucoma, and MD, RNFL_S, RNFL_I, and RNFL_T are negatively correlated with glaucoma. Therefore, in order to calculate the integrated glaucoma risk index described later, the normalized values of some clinical data, such as PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP, are all positively correlated with glaucoma. There is a need to turn it around. For example, the preprocessor 121 sets the normalized values of MD, RNFL_S, RNFL_I, and RNFL_T, which are negatively correlated with glaucoma, to their maximum values, respectively, so that the normalized values of all clinical data have a positive correlation with glaucoma. It can be inverted by subtracting from 1.

The index calculation unit 122 extracts a predetermined number of cases that are most similar to the subject's clinical data or preprocessed clinical data from a pre-built reference database, and creates an integrated glaucoma risk index based on the data of the extracted cases and the preprocessed clinical data. can be calculated. Here, the predetermined number may be 5, but this is only an example and is not limited thereto. The reference database contains data for multiple cases, including glaucoma and normal, and the data for each case may include whether glaucoma is present, clinical data, or preprocessed clinical data.

According to an exemplary embodiment, the index calculation unit 122 may calculate an integrated glaucoma risk index using Equation 1.

Here, I-GRI represents the integrated glaucoma risk index, NNI represents the ratio of the number of cases with glaucoma among the extracted cases to the number of cases extracted from the reference database (e.g., 5), and F _i represents the number of cases with glaucoma in each clinical trial. Represents the preprocessed values of the data, such as PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP, respectively, a _i represents the weight applied to each clinical data, and m and n represent NNI and NNI, respectively.

It can indicate the weight applied to . Additionally, m+n=1.

According to an exemplary embodiment, a _i may be the importance of each clinical data, and may be derived from a machine learning-based glaucoma diagnosis model generated by learning clinical data for learning, so that glaucoma can be diagnosed based on the clinical data. there is. At this time, the machine learning model may be XGboost, which supports a method of self-calculating the importance of each clinical data, but this is only an example and is not limited to this. For example, a _i for each of PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP may be 0.27, 0.14, 0.11, 0.31, 0.1, and 0.07, respectively, but this is only an example and is not limited thereto.

According to an exemplary embodiment, m and n can be derived experimentally and can be 0.2 and 0.8, respectively.

For example, if the input clinical data is (PSD, MD, RNFL_S, RNFL_I, RNFL_T, IOP) = (9.54, -7.84, 56, 54, 48, 11), the preprocessor 121 performs min-max normalization Through the technique, PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP are normalized respectively to obtain (0.5386, 0.5333, 0.3012, 0.2769, 0.3111, 0.25), and the normalized values of MD, RNFL_S, RNFL_I, and RNFL_T are reduced from 1, respectively. By subtracting, the inversion values 0.4667, 0.6889, 0.7231, and 0.6889 can be obtained. As a result, the preprocessor 121 can obtain preprocessed values (0.5386, 0.4667, 0.6889, 0.7231, 0.6889, 0.25) of PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP. The index calculation unit 122 may extract the five cases most similar to the subject's clinical data or preprocessed clinical data from a pre-built reference database. If all of the extracted cases are glaucoma cases, NNI is 1, and the index calculation unit 122 calculates I-GRI = 0.2*1+0.8*(0.27*0.5386+0.14*0.4667+ through Equation 1. 0.11*0.6889+0.31*0.7231+0.1*0.6889+0.07*0.25)≒0.678 can be calculated.

The diagnosis unit 123 may diagnose the subject's glaucoma risk based on the integrated glaucoma risk index.

The integrated glaucoma risk index according to an exemplary embodiment has a value between 0 and 1 and may reflect the risk or progression of glaucoma. For example, the diagnosis unit 123 determines that the greater the value of the integrated glaucoma risk index, the higher the risk or progression of glaucoma, and if the integrated glaucoma risk index is greater than a predetermined threshold, glaucoma may be diagnosed. At this time, the predetermined threshold may be experimentally derived and, for example, may be 0.36, but is not limited thereto.

The diagnosis unit 123 may output the integrated glaucoma risk index and/or glaucoma risk diagnosis result through an output means. For example, the diagnostic unit 123 may generate a graphic chart representing the integrated glaucoma risk index and/or the glaucoma risk diagnosis result and output it through an output means. At this time, the graphic chart can express the location of the integrated glaucoma risk index or the location of each clinical data value. For example, graphic charts may include bar charts, gauge charts, and radar charts.

As described above, when the NNI extracts a predetermined number (e.g., 5) of cases that are most similar to the subject's clinical data or preprocessed clinical data from the reference database, the number of cases (e.g., 5) extracted from the reference database is calculated. It can represent the ratio of the number of cases with glaucoma among the extracted cases. NNI has a value between 0 and 1, and the larger the value, the greater the risk or progression of glaucoma.

Referring to Figure 2, it can be seen that the NNI value of the glaucoma group is very different from the NNI value of the normal group. This may mean that NNI can be used to classify normal and glaucoma cases.

Figure 3 is an example diagram to explain the effect of applying NNI in calculating the integrated glaucoma risk index. In Figure 3 (a), the distribution of the integrated glaucoma risk index (I-GRI) is shown when NNI is not applied to calculate the integrated glaucoma risk index, that is, when the integrated glaucoma risk index is calculated using Equation 2 below: , (b) can represent the distribution of the integrated glaucoma risk index when NNI is applied to calculate the integrated glaucoma risk index, that is, when the integrated glaucoma risk index is calculated using Equation 1 above.

Comparing Figures 3 (a) and (b), it can be seen that NNI brings the I-GRI value of the normal group to the left and the I-GRI value of the glaucoma group to the right. As a result, it can be seen that the area of the overlap region between the glaucoma group and the normal group decreases. In other words, it can be seen that NNI contributes to improving the performance of I-GRI.

Fig. 4 is a diagram illustrating a graphic chart according to an exemplary embodiment. Figure 4 is an example of a bar chart representing the location of the integrated glaucoma risk index.

Referring to Figure 4, the bar chart is divided into a normal area (Healthy), a glaucoma area (Glaucoma), and a border area (Border), which is the overlapping range of normal and glaucoma, and the integrated glaucoma risk index (I-GRI) value ( 0.739) can be displayed on a bar chart. In the illustrated example, it can be seen that the integrated glaucoma risk index is located in the glaucoma area (Glaucoma).

A vertical line 411 may be displayed on the border of the bar chart indicating a threshold value (eg, 0.36) for distinguishing between normal and glaucoma.

Fig. 5 is a diagram illustrating a graphic chart according to an exemplary embodiment. Figure 5 is an example diagram of a radial chart representing the location of each clinical data value. In FIG. 5, (a) the chart may be for a glaucoma patient, (b) the chart may be for a person in a boundary area, which is the overlapping range between the glaucoma patient and a normal person, and (c) the chart may be for a normal person.

As shown in FIG. 5, a radar chart is a chart that visualizes multivariate data in the form of a two-dimensional chart of three or more quantitative variables expressed as axes starting from the same point.

The subject's clinical data values may be displayed on a radial chart. At this time, the displayed clinical data value may be a preprocessed clinical data value obtained by preprocessing the input clinical data.

A polygon 511 indicating the threshold value of each clinical data for distinguishing glaucoma from normal may be displayed on the radial chart. Based on the degree to which the I-

GRI polygons

512a, 512b, 512c, which display the subject's clinical data in the radial chart, cover the polygon 511, which displays the threshold value of each clinical data for distinguishing glaucoma from normal, the subject's The risk or progress of glaucoma and whether or not glaucoma exists can be determined.

Referring to FIG. 6, the glaucoma risk diagnosis device 600 according to an exemplary embodiment may include a data input unit 110, a processor 120, a storage unit 610, a communication unit 620, and an output unit 630. You can. Here, since the data input unit 110 and the processor 120 are the same as described above with reference to FIG. 3, their detailed description will be omitted.

The storage unit 610 may store programs or commands for operating the glaucoma risk diagnosis device 600, and may store data input/output to the glaucoma risk diagnosis device 600 and processed data. For example, the storage unit 610 includes the subject's clinical data input through the data input unit 110, the result data preprocessed by the preprocessor 121, the integrated glaucoma risk index calculated by the index calculation unit 122, and the diagnosis. The diagnosis results of the unit 123, a reference database, etc. can be stored.

The storage unit 610 is a flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM. (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), PROM (Programmable Read Only Memory), magnetic memory, magnetic disk, optical disk. It may include at least one type of storage medium, etc. Additionally, the glaucoma risk diagnosis device 600 may operate an external storage medium such as web storage that performs the storage function of the storage unit 610 on the Internet.

The communication unit 620 can communicate with an external device. For example, the communication unit 620 transmits data input to the glaucoma risk diagnosis device 600, stored data, and processed data to an external device, or calculates the subject's integrated glaucoma risk index from an external device and various types of data to be used for glaucoma diagnosis. Data can be received. At this time, external devices include medical servers, medical devices, print or display devices for outputting results, digital TVs, desktop computers, mobile phones, smart phones, tablets, laptops, PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), It may be, but is not limited to, a navigation device, MP3 player, digital camera, wearable device, etc.

The communication unit 620 can communicate with an external device using wired or wireless communication technology. At this time, wired communication can use twisted pair cable, coaxial cable, optical fiber cable, Ethernet cable, etc., and wireless communication can use Bluetooth communication, BLE communication, short-range wireless communication, WLAN communication, Zigbee communication, infrared communication, WFD communication, UWB communication, Ant+ communication, WIFI communication, RFID communication, 3G communication, 4G communication, and 5G communication can be used. However, these are only examples and are not limited thereto.

The output unit 630 may output data input to the glaucoma risk diagnosis device 600, stored data, processed data, etc. According to one embodiment, the output unit 630 includes the subject's clinical data input through the data input unit 110, the result data preprocessed by the preprocessor 121, and the integrated glaucoma risk index calculated by the index calculation unit 122. , the diagnosis results of the diagnosis unit 123, graphic charts, reference databases, etc. may be output in at least one of an auditory method, a visual method, and a tactile method. To this end, the output unit 630 may include a display, speaker, vibrator, etc.

The glaucoma risk diagnosis method of FIG. 7 may be performed by the glaucoma

risk diagnosis apparatus

100 or 600 of FIG. 1 or FIG. 6 .

Referring to FIG. 7 , the glaucoma risk diagnosis device may receive the subject's clinical data used to calculate the integrated glaucoma risk index or determine the risk/progression of glaucoma from the user (710). Here, the clinical data may include, for example, PSD and MD obtained through visual field testing, RNFL_S, RNFL_I, and RNFL_T obtained through retinal nerve fiber layer optical coherence tomography testing, and IOP obtained through intraocular pressure testing. .

The glaucoma risk diagnosis device may preprocess the clinical data by normalizing the subject's clinical data and converting some clinical data values of the normalized clinical data into reverse values (720).

For example, the glaucoma risk diagnosis device normalizes the subject's clinical data, such as PSD, MD, RNFL_S, RNFL_I, RNFL_T, and IOP, to values between 0 and 1 through a min-max normalization technique, and selects among the normalized clinical data Some clinical data, such as the normalized values of MD, RNFL_S, RNFL_I and RNFL_T, can be inverted by subtracting each from 1.

The glaucoma risk diagnosis device may extract a predetermined number of cases that are most similar to the subject's clinical data or preprocessed clinical data from a pre-built reference database (730). Here, the reference database includes data on multiple cases, including glaucoma and normal, and the data for each case may include whether glaucoma is present, clinical data, or preprocessed clinical data.

The glaucoma risk diagnosis device can calculate an integrated glaucoma risk index based on data from extracted cases and preprocessed clinical data (740). Here, the predetermined number may be 5, but this is only an example and is not limited thereto.

For example, a glaucoma risk diagnosis device can calculate an integrated glaucoma risk index using Equation 1 described above.

The glaucoma risk diagnosis device can diagnose the risk of glaucoma for a subject based on the integrated glaucoma risk index (750).

The integrated glaucoma risk index according to an exemplary embodiment has a value between 0 and 1 and may reflect the risk or progression of glaucoma. For example, the glaucoma risk diagnosis device determines that the greater the value of the integrated glaucoma risk index, the higher the risk or progression of glaucoma, and can diagnose glaucoma if the integrated glaucoma risk index is greater than a predetermined threshold. At this time, the predetermined threshold may be experimentally derived and, for example, may be 0.36, but is not limited thereto.

The glaucoma risk diagnosis device may output the integrated glaucoma risk index and/or glaucoma risk diagnosis result through an output means. For example, the glaucoma risk diagnosis device may generate a graphic chart representing the integrated glaucoma risk index and/or the glaucoma risk diagnosis result and output it through an output means. At this time, the graphic chart can express the location of the integrated glaucoma risk index or the location of each clinical data value. For example, graphic charts may include bar charts, gauge charts, and radar charts.

An aspect of the present invention may be implemented as computer-readable code on a computer-readable recording medium. Codes and code segments implementing the above program can be easily deduced by a computer programmer in the art. Computer-readable recording media may include all types of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media may include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, etc. Additionally, the computer-readable recording medium may be distributed over network-connected computer systems and written and executed as computer-readable code in a distributed manner.

So far, the present invention has been examined focusing on its preferred embodiments. A person skilled in the art to which the present invention pertains will understand that the present invention may be implemented in a modified form without departing from the essential characteristics of the present invention. Accordingly, the scope of the present invention is not limited to the above-described embodiments, but should be construed to include various embodiments within the scope equivalent to the content described in the patent claims.

Claims

a data input unit that receives the subject's clinical data;

a preprocessing unit that preprocesses the clinical data; and

Extract a predetermined number of cases most similar to the clinical data or the preprocessed clinical data from a reference database, calculate the ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, and calculate the preprocessed clinical data. and an index calculation unit that calculates an integrated glaucoma risk index based on the calculated ratio. Including,

The clinical data includes the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, intraocular pressure, and pattern standard deviation and average deviation obtained through visual field testing,

Glaucoma risk diagnosis device.
According to paragraph 1,

The preprocessor,

Normalizing the clinical data to a value between 0 and 1, subtracting the normalized value of some of the clinical data from 1 and converting it to an inverted value,

Glaucoma risk diagnosis device.
According to paragraph 2,

The preprocessor,

Subtracting each of the normalized values of the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, and the average deviation from 1 to convert them into inverted values,

Glaucoma risk diagnosis device.
According to paragraph 1,

The predetermined number is 5,

Glaucoma risk diagnosis device.
According to paragraph 1,

The index calculation unit,

Calculating the integrated glaucoma risk index using the following equation,

Glaucoma risk diagnosis device.

[Equation]

(I-GRI represents the integrated glaucoma risk index, NNI represents the ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, F i represents the preprocessed value of each clinical data, a i represents the weight applied to each clinical data, m and n are NNI and
represents the weight applied to, m+n=1)
According to clause 5,

The a i is the importance of each clinical data derived from a machine learning-based glaucoma diagnosis model created by learning clinical data for learning,

Glaucoma risk diagnosis device.
According to paragraph 1,

a diagnosis unit that diagnoses the risk of glaucoma for the subject based on the integrated glaucoma risk index; Containing more,

Glaucoma risk diagnosis device.
In clause 7,

The diagnostic department,

Diagnosing that the greater the integrated glaucoma risk index, the higher the risk of glaucoma.

Glaucoma risk diagnosis device.
In clause 7,

The diagnostic department,

Generating a graphic chart representing the integrated glaucoma risk index or glaucoma risk diagnosis results,

Glaucoma risk diagnosis device.
According to paragraph 1,

The integrated glaucoma risk index has a value between 0 and 1 and reflects the risk or progression of glaucoma.

Glaucoma risk diagnosis device.
Receiving the subject's clinical data;

Preprocessing the clinical data;

extracting a predetermined number of cases most similar to the clinical data or the preprocessed clinical data from a reference database; and

calculating a ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, and calculating an integrated glaucoma risk index based on the preprocessed clinical data and the calculated ratio; Including,

The clinical data includes the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, intraocular pressure, and the pattern standard deviation and average deviation obtained through visual field testing,

How to diagnose glaucoma risk.
According to clause 11,

The step of preprocessing the clinical data is,

normalizing the clinical data to a value between 0 and 1; and

subtracting normalized values of some of the clinical data from 1 and converting them into inverted values; Including,

How to diagnose glaucoma risk.
According to clause 12,

The step of converting the clinical data is,

Subtracting each of the normalized values of the upper retinal nerve fiber layer thickness, the lower retinal nerve fiber layer thickness, the posterior retinal nerve fiber layer thickness, and the average deviation from 1 to convert them into inverted values,

How to diagnose glaucoma risk.
According to clause 11,

The predetermined number is 5,

How to diagnose glaucoma risk.
According to clause 11,

The step of calculating the integrated glaucoma risk index is,

Calculating the integrated glaucoma risk index using the following equation,

How to diagnose glaucoma risk.

[Equation]

(I-GRI represents the integrated glaucoma risk index, NNI represents the ratio of the number of cases with glaucoma among the extracted cases to the predetermined number, F i represents the preprocessed value of each clinical data, a i represents the weight applied to each clinical data, m and n are NNI and
represents the weight applied to, m+n=1)
According to clause 15,

The a i is the importance of each clinical data derived from a machine learning-based glaucoma diagnosis model created by learning clinical data for learning,

How to diagnose glaucoma risk.
According to clause 11,

Diagnosing the risk of glaucoma for the subject based on the integrated glaucoma risk index; Containing more,

How to diagnose glaucoma risk.
According to clause 17,

The step of diagnosing glaucoma is,

Diagnosing that the greater the integrated glaucoma risk index, the higher the risk of glaucoma.

How to diagnose glaucoma risk.
According to clause 17,

generating a graphic chart representing the integrated glaucoma risk index or glaucoma risk diagnosis results; Containing more,

How to diagnose glaucoma risk.
According to clause 11,

The integrated glaucoma risk index has a value between 0 and 1 and reflects the risk or progression of glaucoma.

How to diagnose glaucoma risk.