Simulation system, simulation method, and simulation program

Abstract

[Problem] To provide a technology with which it becomes more likely to be able to select a posterior chamber phakic intraocular lens having a size suitable for a subject's eye. [Solution] Provided is a simulation system comprising: an acquisition unit that acquires a non-insertion image, which is an image, captured by a tomographic device, of an anterior ocular segment of a subject's eye into which a posterior chamber phakic intraocular lens has not been inserted; a generation unit that generates, on the basis of the non-insertion image and the size of a posterior chamber phakic intraocular lens to be inserted into the subject's eye, an insertion image, which is an image of the anterior ocular segment of the subject's eye into which the posterior chamber phakic intraocular lens has been inserted; an identification unit that identifies, on the basis of the insertion image, an optimal size, which is the optimal size of the posterior chamber phakic intraocular lens to be inserted into the subject's eye; and an output unit that outputs information relating to the optimal size.

Description

Simulation system, simulation method, and simulation program

[0001] The present invention relates to a simulation system, a simulation method, and a simulation program.

[0002] As one of the myopia correction surgeries, a method of inserting a posterior chamber phakic intraocular lens into the eye is known. When the surgery is performed, a posterior chamber phakic intraocular lens of a size suitable for the eye is selected from a plurality of sizes. The size of the posterior chamber phakic intraocular lens needs to be determined before the surgery is performed.

[0003] Conventionally, technologies for determining the size of a posterior chamber phakic intraocular lens before surgery have been developed. For example, in Patent Document 1, inspection data such as the ATA (Angle To Angle) and WTW (White To White) of the person scheduled for surgery is acquired, and based on the inspection data of those who have undergone lens insertion surgery in the past and the size of the posterior chamber phakic intraocular lens, a technology for acquiring the size of the posterior chamber phakic intraocular lens to be used by the person scheduled for surgery is disclosed. Also, in Patent Document 2, inspection data such as ATA and WTW is input into a learning model, and a technology for predicting the size of the anterior chamber angle and the posterior chamber phakic intraocular lens is disclosed.

[0004] Patent No. 7482286, Patent Publication No. 2022-533285

[0005] In the conventional technology, inspection data is acquired for predetermined items, and the size of the posterior chamber phakic intraocular lens is acquired based on the acquired inspection data. However, even for eyes with the same inspection data values, the structure of each part of the eye, for example, the shape of the iris and ciliary body, can be different. Therefore, it has been difficult to accurately determine the size of the posterior chamber phakic intraocular lens based on specific inspection data. The present invention has been made in view of the above problems, and an object thereof is to increase the possibility of selecting a posterior chamber phakic intraocular lens of an appropriate size for the eye of the subject.

[0006] To achieve the above objectives, the simulation system comprises: an acquisition unit that acquires a non-insertion image, which is an image of the anterior segment of the eye to be examined in which a posterior chamber type phakic intraocular lens has not been inserted, taken by a tomography device; a generation unit that generates an insertion image, which is an image of the anterior segment of the eye to be examined in which the posterior chamber type phakic intraocular lens has been inserted, based on the size of the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined and the non-insertion image; an identification unit that identifies the optimal size, which is the optimal size for the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined, based on the insertion image; and an output unit that outputs information regarding the optimal size.

[0007] In other words, the simulation system generates an inserted image of the posterior chamber intraocular lens (IOL) with a posterior chamber IOL of a suitable size, based on an image of the eye without an IOL (non-inserted image) and the size of the IOL. The non-inserted image is taken by a tomography device, and the inserted image generated from this non-inserted image allows for analysis of the state of the eye after the IOL has been inserted, making it possible to determine whether the size is appropriate. Therefore, by adjusting the size until an appropriate size is identified, the optimal size can be determined. With this configuration, the optimal size can be determined based on the structure of each part of the eye. Consequently, the possibility of selecting an IOL of an appropriate size for the eye of the patient (scheduled for surgery) before surgery can be increased.

[0008] This figure shows a simulation system according to one embodiment of the present invention. Figures 2A and 2B schematically show an eye into which a posterior chamber type phakic intraocular lens has been inserted, Figures 2C and 2D show the cross-sectional direction, and Figure 2E shows a posterior chamber type phakic intraocular lens. Figure 3A shows an example of a non-inserted image, and Figures 3B and 3C show examples of inserted images. This figure schematically shows a machine learning model. This is a flowchart of the simulation process. This figure shows an example of displaying the simulation results. This figure shows an example of displaying the simulation results. This is a flowchart of the insertion direction determination process. This figure shows an example of displaying the simulation results. This figure shows an example of displaying the simulation results.

[0009] Here, embodiments of the present invention will be described in the following order: (1) Configuration of the simulation system; (2) Simulation processing; (3) Insertion direction determination processing; (4) Other embodiments:

[0010] (1) Configuration of the simulation system: Figure 1 shows a simulation system 100 according to one embodiment of the present invention. The simulation system 100 according to this embodiment is connected to an optical coherence tomography (OCT) apparatus 200, a storage medium 300, an input unit 400, and a display unit 500. The optical coherence tomography apparatus 200 is a device that splits light from a light source into a reference light and a measurement light, and performs optical coherence tomography by measuring the interference between the reference light returned by a mirror or the like and the measurement light returned by the eye under examination.

[0011] In this embodiment, the optical coherence tomography (OCT) apparatus 200 measures the eye of the subject to be measured, generates measurement information 300a, and stores it in the storage medium 300. The OCT apparatus 200 according to this embodiment can acquire measurement information 300a by measuring at least the anterior segment of the eye of the subject. The configuration of the OCT apparatus 200 is not limited, and it may be capable of measuring a wide range of areas, including the retina.

[0012] The storage medium 300 is a non-volatile storage medium such as an HDD (Hard Disk Drive) or SSD (Solid State Drive). Various types of data can be stored in the storage medium 300, and the simulation system 100 can save data to the storage medium 300 at any time and read data stored in the storage medium 300. In this embodiment, measurement information 300a obtained by measuring the eye of a subject without a posterior chamber type phakic intraocular lens inserted is stored in the storage medium 300. In addition, a cross-sectional image of the anterior segment of the eye is generated based on the measurement information 300a and is stored in the storage medium 300 as a non-insertioned image 300b, which is an image of the anterior segment of the eye of a subject without a posterior chamber type phakic intraocular lens taken by an optical coherence tomography (OCT) device 200.

[0013] Furthermore, in this embodiment, an inserted image 300c, which is an image of the anterior segment of the eye in which a posterior chamber type phakic intraocular lens has been inserted, is generated based on the non-inserted image 300b and stored in the storage medium 300. Furthermore, in this embodiment, a machine learning model 300d, which has been trained to generate the inserted image 300c based on the non-inserted image 300b, is stored in the storage medium 300. Furthermore, in this embodiment, regression equation information 300e for calculating the Vault value, which will be described later, from the measured values ​​of the eye in the examination is stored in the storage medium 300. The various types of information stored in the storage medium 300 will be described in detail later.

[0014] The input unit 400 is a device operated by a user, such as an examiner, and is used to input various types of information to the simulation system 100. The form of the input unit 400 is not particularly limited, but examples include a keyboard or mouse. The display unit 500 is a display device that displays various types of information, and displays various images such as characters and images generated by the simulation system 100.

[0015] The simulation system 100 includes a control unit 110, a communication interface (I / F) 120, and a display I / F 130. The communication I / F 120 is an interface that connects the simulation system 100 to external devices. In this embodiment, an optical coherence tomography (OCT) scanner 200, a storage medium 300, and an input unit 400 are connected to the simulation system 100 via the communication I / F 120. The communication I / F 120 may be of various standards, such as the USB standard, PCI-Express standard, Thunderbolt standard, Ethernet standard, etc. (these are registered trademarks).

[0016] The display I / F 130 is an interface that connects the simulation system 100 and the display unit 500. The display I / F 130 may be of various standards, such as HDMI (High-Definition Multimedia Interface: registered trademark) or DVI (Digital Visual Interface). Of course, the communication I / F 120 and the display I / F 130 are not limited and may be wireless communication interfaces, etc.

[0017] The control unit 110 includes a CPU, RAM, and ROM (not shown), and can execute various programs stored in the storage medium 300, ROM, etc. The programs executed by the control unit 110 include various programs. In this embodiment, a simulation program 111 is included that simulates an image of the posterior chamber type phakic intraocular lens being inserted into the anterior segment of the eye under examination based on measurement information 300a, and identifies the optimal size and insertion direction of the posterior chamber type phakic intraocular lens from the image. When the simulation program 111 is executed, the control unit 110 functions as an acquisition unit 111a, a generation unit 111b, a specification unit 111c, and an output unit 111d.

[0018] The acquisition unit 111a has the function of acquiring a non-insertioned image, which is an image of the anterior segment of the eye under examination that does not have a posterior chamber type phakic intraocular lens inserted, taken by an optical coherence tomography (OCT) scanner. That is, the control unit 110 refers to the measurement information 300a, which shows the raw data measured by the OCT scanner 200, and performs predetermined processing to generate a non-insertioned image 300b. The non-insertioned image 300b is a cross-sectional image showing the state of the eye under examination after it has been cut. In this embodiment, the control unit 110 can generate non-insertioned images 300b cut in multiple cutting directions.

[0019] The cutting direction can be in various directions, but in this embodiment, the cutting direction is the same as one of the insertion directions of the posterior chamber type phakic intraocular lens. Figures 2A and 2B schematically show the eye E into which the posterior chamber type phakic intraocular lens has been inserted. Figure 2A shows the state in which the posterior chamber type phakic intraocular lens L has been inserted in a direction parallel to the vertical direction of the subject (scheduled for surgery). Figure 2B shows the state in which the posterior chamber type phakic intraocular lens L has been inserted in a direction parallel to the left-right direction of the subject.

[0020] Figures 2C and 2D show the cutting directions when generating images. In Figures 2C and 2D, the cutting directions are indicated by dashed lines. Figure 2C shows an example in which the anterior segment of the eye is cut with a cutting plane parallel to the vertical and anterior-posterior directions of the subject and passing through the corneal apex; in this specification, this direction is referred to as the vertical direction. That is, the cutting direction in Figure 2C is the vertical direction, and the insertion direction of the posterior chamber phakic intraocular lens in Figure 2A is the vertical direction. Figure 2D shows an example in which the anterior segment of the eye is cut with a cutting plane parallel to the left-right and anterior-posterior directions of the subject and passing through the corneal apex; in this specification, this direction is referred to as the horizontal direction. That is, the cutting direction in Figure 2D is the horizontal direction, and the insertion direction of the posterior chamber phakic intraocular lens in Figure 2B is the horizontal direction. In this embodiment, it is assumed that the posterior chamber phakic intraocular lens does not move or rotate after it has been inserted into the eye. Therefore, the insertion direction of a posterior chamber type phakic intraocular lens can also be said to be the direction of lens fixation.

[0021] Figure 3A shows an example of a non-inserted image 300b. In this embodiment, the control unit 110 generates a non-inserted image 300b with a horizontal cutting direction based on the measurement information 300a using the function of the acquisition unit 111a, and stores it in the storage medium 300. Figure 3A shows a non-inserted image 300b when the anterior segment of the eye under examination is cut horizontally. In Figure 3A, areas that are not clear in the optical coherence tomography apparatus 200 are indicated by dashed lines (the same applies hereafter).

[0022] The generation unit 111b has the function of generating an inserted image, which is an image of the anterior segment of the eye in which the posterior chamber type phakic intraocular lens is inserted, based on the size of the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined and the non-inserted image. In this embodiment, the control unit 110 generates an inserted image 300c based on the machine learning model 300d and stores it in the storage medium 300.

[0023] Figure 4 is a schematic diagram illustrating the machine learning model in this embodiment. In this embodiment, a generative model G and a discriminative model D are prepared as training models. The generative model G is a model that generates one image from another image. The discriminative model D is a model that compares the first image and the second image and determines whether the second image is a fake image generated from the first image or not.

[0024] In this embodiment, the non-inserted image 300b is input to the generation model G, and machine learning is performed to generate the inserted image 300c. Furthermore, during the machine learning process, pairs of images It without a posterior chamber phakic intraocular lens inserted and ground truth images Ip with a posterior chamber phakic intraocular lens inserted are used as training data.

[0025] Furthermore, in the machine learning process, there are two possibilities: one where the non-inserted image 300b is used as the first image and the inserted image 300c generated by the generative model G is input to the discrimination model D as the second image; and another where the non-inserted image 300b, used as training data It, is used as the first image and the ground truth image Ip associated with the non-inserted image 300b is input to the discrimination model D as the second image.

[0026] The discriminative model D is trained to output false if the second image is an inserted image 300c generated by the generative model G, and true if the second image is a ground truth image Ip associated with the training data It. The structure of the model and the setting of the loss function when performing machine learning can be done using various methods; for example, a well-known GAN (Generative Adversarial Network) can be used.

[0027] In this embodiment, the machine learning described above is performed for each combination of the size of the posterior chamber intraocular lens and the insertion direction of the posterior chamber intraocular lens, and the results are stored in the storage medium 300 as a machine learning model 300d, associated with the posterior chamber intraocular lens and the insertion direction. With this configuration, the control unit 110 can generate an inserted image 300c from the non-inserted image 300b, which represents the case when a posterior chamber intraocular lens of a specific size is inserted in a specific insertion direction.

[0028] While the method for defining the size of a posterior chamber phakic intraocular lens is not limited, in this embodiment, the size is defined by the diagonal distance Ls of the posterior chamber phakic intraocular lens L, as shown in Figure 2E. In this embodiment, several typical sizes of the posterior chamber phakic intraocular lens L are identified in advance, and a machine learning model 300d is generated by associating each of these sizes with any combination of two insertion directions (vertical and horizontal). For example, if four sizes are selectable, eight (= 4 × 2) machine learning models 300d are generated.

[0029] The control unit 110 sets the size and insertion direction of the posterior chamber phakic intraocular lens using the function of the generation unit 111b, and acquires a machine learning model 300d corresponding to the set size and insertion direction. Then, the control unit 110 inputs a non-inserted image 300b to the acquired machine learning model 300d to generate an inserted image 300c, which is an image of the anterior segment of the eye in which the posterior chamber phakic intraocular lens has been inserted.

[0030] Figures 3B and 3C show examples of inserted images 300c. Figure 3B shows an example of an inserted image 300c generated based on the non-inserted image 300b shown in Figure 3A, with the size of the posterior chamber intraocular lens set to a certain size and the insertion direction set to the horizontal. Figure 3C shows an example of an inserted image 300c generated based on the non-inserted image 300b shown in Figure 3A, with the size of the posterior chamber intraocular lens set to a certain size and the insertion direction set to the vertical.

[0031] In other words, in this embodiment, even if the insertion direction of the posterior chamber type phakic intraocular lens is vertical, the cutting direction of the cross-sectional view is horizontal. This is because collecting samples of cross-sectional views when cut horizontally is easier than collecting samples of cross-sectional views when cut vertically. Specifically, it is relatively difficult to perform measurements with the optical coherence tomography (OCT) device 200 when the anterior segment of the eye, for example, the cornea, is not obscured by the upper and lower eyelids, making it difficult to collect samples. On the other hand, in a cross-sectional view cut horizontally, the possibility of the anterior segment of the eye being obscured by the upper and lower eyelids is low. Furthermore, even if training data showing a vertical cross-sectional view with the posterior chamber type phakic intraocular lens inserted vertically could be collected without being affected by poor eyelid opening, it is often difficult to sufficiently open the upper and lower eyelids of the subject. For this reason, even if the insertion direction of the posterior chamber type phakic intraocular lens is vertical, by making the cutting direction of the cross-sectional view horizontal, the possibility of misanalysis due to such poor eyelid opening can be reduced. Based on the above, in this embodiment, when collecting training data for generating the machine learning model 300d, the cutting direction of the cross-sectional view is set to the horizontal direction. Therefore, regardless of whether the insertion direction of the posterior chamber type phakic intraocular lens is horizontal or vertical, the system is configured to generate an inserted image 300c by inputting a non-inserted image 300b showing a cross-section cut horizontally.

[0032] The identification unit 111c has the function of determining the optimal size, which is the size of the posterior chamber type phakic intraocular lens to be inserted into the eye under examination, based on the inserted image. In this embodiment, the control unit 110 determines the distance between the lens and the posterior chamber type phakic intraocular lens based on the inserted image 300c, using the function of the identification unit 111c. Specifically, in this embodiment, the control unit 110 sets the size of the posterior chamber type phakic intraocular lens to a specific size and insertion direction to a specific insertion direction, and then inputs the non-inserted image 300b into a machine learning model 300d corresponding to the specific size and specific insertion direction to generate the inserted image 300c.

[0033] Furthermore, the control unit 110 refers to the inserted image 300c and identifies the positions of the crystalline lens and the posterior chamber phakic intraocular lens within the image. Identifying these positions can be achieved in various ways, for example, by using pattern matching or machine learning models. Once the positions of the crystalline lens and the posterior chamber phakic intraocular lens within the image are identified, the control unit 110 identifies the distance V between the posterior surface of the posterior chamber phakic intraocular lens (the posterior end face of the subject) and the entire surface of the crystalline lens (the anterior end face of the subject) on a straight line passing through the corneal apex (see Figures 3B and 3C). In this specification, this distance is referred to as the Vault value. In this embodiment, the distance occupied by each pixel in the cross-sectional image in real space is predetermined, and the Vault value is identified as the distance in real space (μm) based on the number of pixels in the distance V. The control unit 110 can perform the above processing for each size and insertion direction of the posterior chamber phakic intraocular lens.

[0034] In this embodiment, the control unit 110 considers the size and insertion direction of the posterior chamber phakic intraocular lens to be optimal if the Vault value falls within a predetermined range. In this specification, the optimal size is referred to as the optimal size, and the optimal insertion direction as the optimal insertion direction. If the Vault value does not fall within a predetermined range, the control unit 110 changes the size and insertion direction of the posterior chamber phakic intraocular lens, generates an insertion image 300c based on a machine learning model 300d corresponding to the changed size and insertion direction, and identifies the Vault value. The control unit 110 repeats the process of changing the size and insertion direction of the posterior chamber phakic intraocular lens until the Vault value obtained in this way falls within a predetermined range.

[0035] The predetermined range set for the Vault value only needs to be set so that the state in which the posterior chamber type phakic intraocular lens is inserted is considered to be the optimal state. In this embodiment, the target value ± default value is the predetermined range. The target value may be determined by various methods, and in this embodiment, it is specified in advance by the examiner. The default value is fixed in advance. For example, a value of approximately 500 μm is assumed as the target value, and a value of approximately 200 μm is assumed as the default value.

[0036] In this embodiment, in addition to a configuration for determining the Vault value based on the analysis of the inserted image 300c, a configuration for determining the Vault value based on a regression equation is also provided. Specifically, regression equation information 300e is stored in the storage medium 300. The regression equation information 300e is information that shows a regression equation for calculating the Vault value based on the measurement results identified from the measurement information 300a of the eye being examined.

[0037] The regression equation can be defined in various ways, for example, as a function of the ATA, ACW (Anterior Chamber Depth), CLR (Crystalline Lens Rise), and the size of the posterior chamber intraocular lens. Examples of such regression equations include those called the NK equation and the KS equation, and in this embodiment, regression equations for the horizontal and vertical directions are defined in advance.

[0038] The control unit 110 refers to the regression equation information 300e and selects a regression equation corresponding to the insertion direction of the posterior chamber type phakic intraocular lens assumed when generating the inserted image 300c. Then, the control unit 110 identifies the ATA, ACW, and CLR by performing predetermined processing based on the measurement information 300a. Furthermore, the control unit 110 identifies the size of the posterior chamber type phakic intraocular lens assumed when generating the inserted image 300c and obtains a Vault value by substituting these values ​​into the regression equation.

[0039] The output unit 111d has the function of outputting information regarding the optimal size. The information regarding the optimal size can be any information that the examiner refers to in order to determine the optimal size and optimal insertion direction. In this embodiment, the information regarding the optimal size includes the optimal size, optimal insertion direction, Vault value, and inserted image 300c. The output unit 111d controls the display unit 500 via the display I / F 130 to display a screen for outputting information regarding the optimal size, and displays the optimal size, optimal insertion direction, Vault value, and inserted image 300c at a predetermined position on the screen. Various types of information may be displayed on the screen, for example, the examiner's identification information, the date of the examination, comments, and the measured value of the eye under examination (ATA, etc.) identified from the measurement information 300a may be displayed.

[0040] In the above configuration, an uninserted image 300b is generated, showing the image of the subject's eye without the posterior chamber phakic intraocular lens inserted. An inserted image 300c is then generated based on the size and insertion direction of the posterior chamber phakic intraocular lens. In the configuration that calculates the Vault value using the inserted image 300c, the optimal size and insertion direction can be determined based on the structure of each part of the subject's eye (e.g., shape, size, two-dimensional and three-dimensional position). The configuration that calculates the Vault value using the inserted image 300c has been statistically confirmed to be more accurate in calculating the Vault value compared to the configuration that calculates the Vault value based on regression equation information 300e. Therefore, it is possible to increase the likelihood of selecting a posterior chamber phakic intraocular lens of appropriate size for the subject's eye and selecting the optimal insertion direction before surgery.

[0041] (2) Simulation processing: Next, the simulation processing performed by the control unit 110 in the above-described configuration will be described in detail. FIG. 5 is a flowchart showing the simulation processing. The simulation processing may be executed when the measurement information 300a of the subject eye is measured by the optical coherence tomography device 200, or the simulation processing may be executed using the previously measured measurement information 300a.

[0042] When the simulation processing is started, the control unit 110 refers to the storage medium 300 and acquires the measurement information 300a (step S100). Next, the control unit 110 acquires the non-insertion image 300b by the function of the acquisition unit 111a (step S105). That is, the control unit 110 performs a predetermined process based on the measurement information 300a and generates a non-insertion image 300b showing a cross-sectional view in the horizontal direction of the subject eye. As a result, for example, a non-insertion image 300b as shown in FIG. 3A is generated.

[0043] Next, the control unit 110 sets the insertion direction of the posterior chamber phakic intraocular lens (step S110). The insertion direction may be set by various methods. For example, the insertion direction may be set based on the input of the examiner using the input unit 400, or the insertion direction may be set based on the state of the subject eye. The state of the subject eye may be specified by various elements. In the present embodiment, an example in which the insertion direction is determined based on the refractive error state of the subject eye is assumed. The details of the insertion direction determination process for determining the insertion direction will be described later.

[0044] Next, the control unit 110 sets the size of the posterior chamber phakic intraocular lens (step S115). In the present embodiment, a situation in which the size of the posterior chamber phakic intraocular lens can be selected from a plurality of sizes is assumed. Therefore, the examiner operates the input unit 400 to input a desired size, and the control unit 110 sets the input size as the size of the posterior chamber phakic intraocular lens. Although the selectable sizes are not limited, in the present embodiment, it is assumed that 12.1 mm, 12.6 mm, 13.2 mm, and 13.7 mm can be selected.

[0045] Next, the control unit 110 generates an inserted image 300c using the functions of the generation unit 111b (step S120). That is, the control unit 110 refers to the storage medium 300 and acquires a machine learning model 300d corresponding to the insertion direction set in step S110 and the size set in step S115. Then, the control unit 110 generates an inserted image 300c by inputting the non-inserted image 300b acquired in step S105 to the acquired machine learning model 300d. As a result, if the insertion direction is horizontal, for example, an inserted image 300c like the one shown in Figure 3B is generated, and if the insertion direction is vertical, for example, an inserted image 300c like the one shown in Figure 3C is generated.

[0046] Next, the control unit 110 acquires a Vault value based on the inserted image using the function of the identification unit 111c (step S125). That is, the control unit 110 refers to the inserted image 300c generated in step S120 and identifies the positions of the crystalline lens and the posterior chamber phakic intraocular lens within the image. The control unit 110 then identifies the distance V between the posterior surface of the posterior chamber phakic intraocular lens and the front surface of the crystalline lens as the Vault value, on a straight line passing through the corneal apex.

[0047] Next, the control unit 110 acquires measurement values ​​based on the measurement information 300a, the non-inserted image 300b, and the inserted image 300c (step S130). In this embodiment, the control unit 110 is configured to perform predetermined processing based on the measurement information 300a, the non-inserted image 300b, and the inserted image 300c, and to acquire measurement values ​​for various measurement items of the eye under examination. The measurement items are not limited, but in this embodiment, they include corneal shape analysis, WTW, CYL (Cylinder, astigmatism), AXIS (Axis, cylindrical axis), ACW, ATA, CLR, ACD (Anterior Chamber depth), and TIA. ACW, ATA, CLR, ACD, and TIA are identified based on the non-inserted image 300b and the inserted image 300c, respectively, and are defined as pre-operative and post-operative values.

[0048] Next, the control unit 110 acquires the Vault value based on the regression formula (step S135). That is, the control unit 110 refers to the regression formula information 300e and identifies the regression formula corresponding to the insertion direction set in step S110. Then, the control unit 110 uses the ATA, ACW, CLR, and the size of the posterior chamber phakic intraocular lens set in step S115 as input variables, and acquires the value of the regression formula using each of the NK formula and the KS formula. The acquired value is the Vault value specified based on the regression formula.

[0049] Next, the control unit 110 determines whether the Vault value is within a predetermined range by the function of the specifying unit 111c (step S140). In the present embodiment, the Vault value to be determined is the Vault value acquired in step S125. In the present embodiment, since the Vault value acquired based on the insertion image 300c is statistically more accurate than the Vault value acquired based on the regression formula information 300e, it is determined whether the former Vault value is within the predetermined range. Here, it is only necessary to refer to the Vault value acquired based on the insertion image 300c, and a determination combining the Vault value acquired based on the regression formula information 300e may also be performed.

[0050] If it is determined in step S140 that the Vault value is not within the predetermined range, the control unit 110 repeats the processing after step S110. On the other hand, if it is determined in step S140 that the Vault value is within the predetermined range, the control unit 110 regards the insertion direction set in step S110 as the optimal insertion direction and the size set in step S115 as the optimal size, and displays the simulation result by the function of the output unit 111d (step S145). That is, the control unit 110 controls the display unit 500 to display the measurement value acquired in step S130, the non-insertion image 300b acquired in step S105, the insertion image 300c generated in step S120, the Vault values acquired in steps S125 and S135, the optimal size, and the optimal insertion direction on the display unit 500.

[0051] Figures 6 and 7 show examples of simulation results. Figure 6 shows an example of the display when the posterior chamber intraocular lens is inserted vertically, and Figure 7 shows an example of the display when the posterior chamber intraocular lens is inserted horizontally. In the screens shown in these figures, an icon indicating either the left or right eye is displayed in the upper right corner of the screen, and in this example, it is indicated as the right eye.

[0052] In the examples shown in Figures 6 and 7, the subject's identification information and the date are displayed at the top of the screen. The non-inserted image 300b is displayed on the left side of the middle section of the screen, and the inserted image 300c is displayed on the right side of the middle section. An icon I1 is displayed in the upper left of the non-inserted image 300b to indicate that the cutting direction of the non-inserted image 300b is horizontal. An icon I2 is displayed in the upper left of the inserted image 300c to indicate that the cutting direction of the inserted image 300c is horizontal and the insertion direction of the posterior chamber phakic intraocular lens is vertical.

[0053] Furthermore, the lower part of the screen displays Table T1 showing corneal topography analysis, WTW, CYL, and AXIS; Table T2 showing ACW, ATA, CLR, ACD, and TIA; and Table T3 showing Vault values. In Table T2, the measured values ​​identified from the non-inserted image 300b are shown as Pre-Op (pre-surgery), and the measured values ​​identified from the inserted image 300c are shown as Post-Op (post-surgery). Between Table T3 and the inserted image 300c, it is indicated that the target value (Target Vault) for defining the predetermined range is 500 μm. Figure 6 shows an example where the optimal insertion direction is the vertical direction (V), and Figure 7 shows an example where the optimal insertion direction is the estimated direction (H).

[0054] Table T3 shows the Vault value identified by the regression equation (NK equation, KS equation) and the Vault value identified by the inserted image 300c for each size of the posterior chamber phakic intraocular lens set in step S115. In the example shown in Figures 6 and 7, when the size of the posterior chamber phakic intraocular lens is 13.2 mm, the Vault value identified by the inserted image 300c is 525 μm. Here, the target value of the Vault value is 500 μm, and the predetermined range is 500 ± 200 μm, so 13.2 is the optimal size. Also, here, it is assumed that the initially set size of 13.2 mm was the optimal size without setting other sizes, so the Vault values ​​for other sizes are not displayed (a minus sign is shown). Of course, if Vault values ​​for other sizes have been obtained, the obtained Vault values ​​may be displayed.

[0055] In the display described above, the optimal size and insertion direction are shown, allowing the examiner to determine the optimal size and insertion direction for the subject's eye. Furthermore, a Vault value is output, allowing the examiner to know the Vault value when a posterior chamber type phakic intraocular lens of the set size is inserted in the set direction. Therefore, the examiner can decide whether to use the size and insertion direction shown on the screen, or whether to consider other sizes and insertion directions.

[0056] Furthermore, since the inserted image 300c is output, the examiner can see the condition of the patient's eye after the insertion of the posterior chamber phakic intraocular lens. Therefore, it is easy to intuitively grasp the effects of inserting the posterior chamber phakic intraocular lens and the changes in the patient's eye. Moreover, by presenting the inserted image 300c to the patient, the effects of inserting the posterior chamber phakic intraocular lens and the changes in the patient's eye can be easily explained, and this can be used for informed consent, etc.

[0057] Furthermore, in the screens shown in Figures 6 and 7, various other types of information may be displayed in addition to the information illustrated. For example, information warning of the effects of inserting a posterior chamber phakic intraocular lens may be displayed. Various types of information can be envisioned for this purpose, but for example, a configuration could be adopted in which an acceptable range for the anterior chamber angle and anterior chamber depth is set after inserting a posterior chamber phakic intraocular lens, and a warning is issued if the range is outside the acceptable range. With this configuration, if there are concerns regarding the insertion of a posterior chamber phakic intraocular lens, the examiner can easily grasp the concerns.

[0058] (3) Insertion direction determination process: Next, the insertion direction determination process will be explained. Figure 8 is a flowchart of the insertion direction determination process. The control unit 110 executes the insertion direction determination process using the function of the specific unit 111c. For posterior chamber intraocular lenses, there are lenses for correcting astigmatism. When using lenses for correcting astigmatism, the distance between the ciliary body sulci is longer in the vertical direction than in the horizontal direction, so inserting in the vertical direction rather than the horizontal direction has the advantage that the posterior chamber intraocular lens is less likely to rotate and the postoperative astigmatism correction prognosis is more stable. In addition, when inserting in the vertical direction, the upper part of the cornea is incised in the left-right direction and the posterior chamber intraocular lens is inserted through this opening, which has the advantage that regular astigmatism is reduced by induced astigmatism. Even when lenses for correcting astigmatism are not used, it is preferable to insert in the vertical direction if the cylindrical power (astigmatism power) of the subject's corneal refractive power before insertion of the posterior chamber intraocular lens is greater than the threshold or if the patient has regular astigmatism. When inserted vertically, the posterior chamber phakic intraocular lens is less likely to rotate than when inserted horizontally, making it easier to avoid a reduction in astigmatism correction due to toric axis misalignment after insertion.

[0059] The insertion direction determination process is performed based on the above knowledge. However, there are multiple types of information that can be referenced regarding astigmatism. Specifically, the measured values ​​from subjective refractive power testing (measured with an ophthalmoscopic device), objective refractive power testing (measured with a refractometer), and corneal refractive power testing (measured with an optical coherence tomography device 200) can be referenced as information regarding astigmatism. However, there is a priority order for the measurement items referenced in the correction of the subject's visual function, and among the above three types of measurement items, subjective refractive power testing, objective refractive power testing, and corneal refractive power testing have the highest priority in that order.

[0060] In this embodiment, it is assumed that various measurement items may be measured in advance for the subject's eye. If measurements have been taken, the measured values ​​are stored in the storage medium 300, but if no measurements have been taken, no measured values ​​exist. Therefore, the control unit 110 determines whether or not there is subjective cylindrical power and astigmatism axis input information for the subject (step S200). That is, the control unit 110 determines that there is subjective cylindrical power and astigmatism axis input information for the subject if the measured values ​​of the eye from the subjective refractive power test are stored in the storage medium 300. If it is determined in step S200 that there is subjective cylindrical power and astigmatism axis input information for the subject, the control unit 110 skips steps S205 and S210.

[0061] In step S200, if it is not determined that there is subjective cylindrical power and astigmatism axis input information for the subject, the control unit 110 determines whether or not there is objective cylindrical power and astigmatism axis input information for the subject (step S205). That is, the control unit 110 determines that there is objective cylindrical power and astigmatism axis input information for the subject if the measured value of the subject's eye by the refractometer is stored in the storage medium 300. In step S205, if it is determined that there is objective cylindrical power and astigmatism axis input information for the subject, the control unit 110 skips step S210.

[0062] In step S205, if it is determined that there is no input information for the subject's objective cylindrical power and astigmatism axis, the control unit 110 acquires the subject's corneal cylindrical power and astigmatism axis based on the measurement information 300a (step S210). That is, the control unit 110 performs predetermined processing based on the measurement information 300a acquired by the optical coherence tomography apparatus 200, performs corneal shape analysis, and acquires the corneal cylindrical power and astigmatism axis.

[0063] Next, the control unit 110 determines whether the astigmatism axis is within the range of regular astigmatism (step S215). The range of regular astigmatism is any range predetermined as a typical range for the astigmatism axis in the case of regular astigmatism, for example, a range within ±20° of the astigmatism axis (axis direction of the weak principal meridian). Here, the astigmatism axis to be determined as to whether or not it is within the range of regular astigmatism differs depending on the determination results in steps S200 and S205.

[0064] Specifically, in step S200, if it is determined that subjective cylindrical power and astigmatism axis input information is available for the subject, the astigmatism axis measured by the subjective refractive power test will be used for the determination. In step S205, if it is determined that objective cylindrical power and astigmatism axis input information is available for the subject, the astigmatism axis measured by the refractometer will be used for the determination. In step S205, if it is not determined that objective cylindrical power and astigmatism axis input information is available for the subject, the astigmatism axis identified based on the measurement information 300a from the optical coherence tomography device 200 will be used for the determination.

[0065] In step S215, if it is determined that the astigmatism axis is within the range of regular astigmatism, the control unit 110 sets the insertion direction of the posterior chamber phakic intraocular lens to the vertical direction (step S230). Regular astigmatism is a form of astigmatism in which the vertical astigmatism is strong, and in cases of regular astigmatism, the astigmatism is reduced by making an incision in the left-right direction on the upper part of the cornea. For this reason, in cases of regular astigmatism, the insertion direction of the posterior chamber phakic intraocular lens is initially set to the vertical direction.

[0066] In this case, the process from step S115 onward determines whether the Vault value is within a predetermined range when the insertion direction of the posterior chamber phakic intraocular lens is set to the vertical direction. That is, according to the above process, the vertical direction is selected when the astigmatism axis is within the range of regular astigmatism, so the insertion direction is set so that the posterior chamber phakic intraocular lens is inserted in the direction corresponding to the astigmatism axis of the eye being examined. Then, based on the set insertion direction, the process from step S115 onward is executed, and the optimal size of the posterior chamber phakic intraocular lens is determined based on the insertion image 300c when the posterior chamber phakic intraocular lens is inserted in the set insertion direction.

[0067] Furthermore, if the Vault value is not within the predetermined range, steps S110 to S140 are repeated until processing is completed for all selectable sizes, and it is determined whether the Vault value is within the predetermined range with the insertion direction fixed vertically. If it is determined that the Vault value is not within the predetermined range with the insertion direction fixed vertically, then the optimal size cannot be determined when inserted vertically. In this case, the control unit 110 changes the direction in which the posterior chamber phakic intraocular lens is inserted and performs the processing from step S115 onwards to determine the optimal size. In the actual process of considering the size of a posterior chamber phakic intraocular lens, statistically, there are many cases where the Vault value is within an appropriate range without changing the insertion direction. Therefore, if the insertion direction is determined first based on the state of astigmatism, the optimal insertion direction and optimal size suitable for astigmatism correction can be efficiently determined.

[0068] On the other hand, if it is not determined in step S215 that the astigmatism axis is within the range of regular astigmatism, the control unit 110 determines whether the magnitude of the cylindrical power is greater than a threshold (step S220). The threshold that is compared with the magnitude of the cylindrical power can be any value predetermined as a value for determining whether or not a posterior chamber type phakic intraocular lens for astigmatism correction is used, for example, it may be 1.5D. Here, the cylindrical power for which it is determined whether or not the magnitude of the cylindrical power is greater than the threshold differs depending on the determination results in steps S200 and S205.

[0069] Specifically, in step S200, if it is determined that subjective cylindrical power and astigmatism axis input information is available for the subject, the cylindrical power measured by the subjective refractive power test will be used for the determination. In step S205, if it is determined that objective cylindrical power and astigmatism axis input information is available for the subject, the cylindrical power measured by the refractometer will be used for the determination. In step S205, if it is not determined that objective cylindrical power and astigmatism axis input information is available for the subject, the cylindrical power identified based on the measurement information 300a from the optical coherence tomography device 200 will be used for the determination.

[0070] In step S220, if it is determined that the cylindrical power is greater than the threshold, the control unit 110 sets the insertion direction of the posterior chamber phakic intraocular lens to the vertical direction (step S230). This is because if the cylindrical power is greater than the threshold, a posterior chamber phakic intraocular lens for astigmatism correction is used, and when inserted vertically, the posterior chamber phakic intraocular lens is less likely to rotate than when inserted horizontally, making it easier to avoid a reduction in the astigmatism correction effect due to toric axis misalignment after insertion. On the other hand, in step S220, if it is not determined that the cylindrical power is greater than the threshold, the control unit 110 sets the insertion direction of the posterior chamber phakic intraocular lens to the horizontal direction (step S225).

[0071] (4) Other Embodiments: The embodiments described above are just examples for carrying out the present invention, and various other embodiments can be adopted. For example, the configuration of the simulation system is not limited to the configuration shown in Figure 1. The devices shown in Figure 1 may be fewer or more devices that share functions. Specifically, the simulation system 100 may be an integrated device with at least one of the optical coherence tomography apparatus 200, storage medium 300, input unit 400, and display unit 500. Alternatively, the simulation system 100 may be distributed across more devices, such as by having at least a part of it configured as a server.

[0072] Furthermore, the cutting direction of the non-inserted image 300b and the inserted image 300c is not limited to the horizontal direction. For example, based on the non-inserted image 300b showing the eye of the subject cut vertically, an inserted image 300c showing a cross-sectional view of the eye of the subject with a posterior chamber phakic intraocular lens inserted vertically may be generated. In this case, based on the inserted image 300c, the Vault value of the vertically inserted posterior chamber phakic intraocular lens is determined, and processing is performed to determine the optimal size based on the Vault value.

[0073] Furthermore, the processes shown in Figures 5 and 8 are examples, and various omissions, additions, and substitutions may be made. For example, the processing order of steps S110 and S115 may be reversed. In steps S130 and S135, processes that do not require the use of the inserted image 300c may be executed before step S120. Moreover, if the Vault value based on the regression equation is not used, steps S130 and S135 may be omitted.

[0074] Furthermore, various procedures can be employed to determine the optimal size and insertion direction. For example, the control unit 110 may determine Vault values ​​for all (or more) possible combinations of size and insertion direction for a posterior chamber phakic intraocular lens, and then determine the size and insertion direction that gives the optimal (e.g., closest to the target value) Vault value from among the obtained Vault values.

[0075] Such a configuration can be achieved, for example, by repeating steps S110 to S135 of the simulation process in Figure 5 in the above-described embodiment until all possible combinations of size and insertion direction are covered. In this case, it is preferable to display all acquired Vault values ​​in step S145. Figures 9 and 10 show examples of displaying the simulation results. The screen configuration is the same as the example shown in Figures 6 and 7, but in Figures 9 and 10, the acquired Vault values ​​for all selectable sizes are displayed. The insertion direction may be displayed on separate screens for the vertical and horizontal directions, as shown in Figures 9 and 10, or the Vault values ​​for each size in both the vertical and horizontal directions may be displayed in a list. In addition, in the example shown in Figures 9 and 10, the inserted image 300c when a specific size is set is displayed, but of course, it may be possible to switch to an inserted image 300c of any size depending on the input to the input unit 400, or the inserted image 300c when set to each size may be displayed in a list.

[0076] With the above configuration, the examiner can compare the Vault value for all (or more) possible combinations of size and insertion direction. Therefore, for example, it is possible to compare the Vault value and the inserted image 300c for multiple conditions in which the Vault value is within a predetermined range, and to consider a more appropriate size and insertion direction.

[0077] Furthermore, while the above-described embodiment is configured to determine the optimal size and insertion direction for a posterior chamber phakic intraocular lens, it may also be configured so that either one is determined based on the insertion image 300c. For example, if the posterior chamber phakic intraocular lens is fixed in a specific insertion direction and the insertion direction is not selected, then the optimal size is determined based on the insertion image 300c.

[0078] The acquisition unit only needs to be able to acquire non-insertioned images, which are images of the anterior segment of the eye under examination that does not have a posterior chamber phakic intraocular lens inserted, taken with a tomography device. In other words, the acquisition unit only needs to be able to acquire non-insertioned images of the anterior segment of the patient's own eye under examination, taken with a tomography device before surgery. The target of imaging can be the anterior segment, that is, the area where the posterior chamber phakic intraocular lens will be inserted, and at a minimum, it should include the area that affects the size of the posterior chamber phakic intraocular lens.

[0079] Non-inserted images can be any images that include the structure of the anterior segment of the eye before surgery, and may be cross-sectional images of the anterior segment or three-dimensional images (information in which grayscale values ​​etc. indicating the structure of the eye at each three-dimensional position are specified). Of course, an image of the entire eye may also be taken. The type of tomography device used to take non-inserted images is not limited, and tomography devices with various imaging methods and configurations may be used. Specifically, examples of tomography devices include optical coherence tomography devices, as well as tomography devices using ultrasound and tomography devices using slit light. Furthermore, by changing the orientation of the cross-sections taken by the tomography device and taking multiple cross-sectional images, a three-dimensional image can be generated. For this reason, non-inserted and inserted images may be three-dimensional images generated based on cross-sectional images taken by any tomography device.

[0080] The generation unit should be able to generate an inserted image, which is an image of the anterior segment of the eye with the posterior chamber phakic intraocular lens (PLA) inserted, based on the size of the PLA to be inserted into the eye and the image without the lens. In other words, the generation unit should be able to generate an inserted image that shows the image when the PLA is inserted, and allows for analysis of the relationship between the PLA and the various parts of the anterior segment surrounding the PLA. The inserted image should be an image showing the anterior segment with the PLA inserted, and may be a cross-sectional image of the anterior segment including the PLA, or a three-dimensional image.

[0081] The identification unit only needs to be able to determine the optimal size of the posterior chamber intraocular lens to be inserted into the eye under examination, based on the inserted image. That is, the identification unit only needs to be able to determine whether the size of the posterior chamber intraocular lens is optimal or not, based on whether the state of the posterior chamber intraocular lens in the inserted image is in a predetermined state or not. If the size of the posterior chamber intraocular lens is determined to be not optimal based on a given inserted image, the identification unit can generate inserted images with different sizes of posterior chamber intraocular lenses and determine whether that size is optimal or not.

[0082] The indicators used to determine whether the size of a posterior chamber phakic intraocular lens is optimal are not limited to Vault values, but may include various other indicators. Furthermore, the optimal size may be determined based on multiple indicators. In addition, the optimal size may be determined based on a machine learning model.

[0083] The output unit only needs to be able to output information about the optimal size. That is, the information about the optimal size that the output unit can output may be information indicating the optimal size itself, an inserted image when a lens of the optimal size is inserted, or various parameters such as the Vault value when a lens of the optimal size is inserted. Furthermore, the output unit only needs to be able to output information about the optimal size identified from the inserted image, and may also output other information, such as information about the optimal size estimated based on a regression equation.

[0084] Furthermore, as described above, the application of the configuration to obtain a Vault value by setting the insertion direction of the posterior chamber type phakic intraocular lens in a direction corresponding to the astigmatism axis of the eye under examination is not limited to configurations that determine the Vault value and optimal size based on the insertion image 300c. For example, it may also be applied to configurations that determine the Vault value based on a regression equation and determine the optimal size based on that Vault value. That is, it may be a configuration in which the insertion direction of the posterior chamber type phakic intraocular lens is set based on the astigmatism axis of the eye under examination, and the optimal size of the posterior chamber type phakic intraocular lens is determined based on the Vault value obtained when the posterior chamber type phakic intraocular lens is inserted in the set insertion direction.

[0085] If the optimal size cannot be determined, the direction in which the posterior chamber intraocular lens is inserted is changed, and the optimal size of the posterior chamber intraocular lens is determined based on the Vault value when the posterior chamber intraocular lens is inserted in the changed insertion direction. With this configuration, it is possible to examine whether an optimal size exists when the posterior chamber intraocular lens is inserted in an insertion direction suitable for astigmatism correction, and thus the optimal size when using an astigmatism-correcting lens can be efficiently determined.

[0086] Furthermore, the method for determining the optimal size and insertion direction based on an inserted image generated from a non-inserted image can also be applied as a method or program invention. Moreover, such systems, methods, and programs can be implemented as standalone devices or as part of a device with multiple functions, encompassing various embodiments.

[0087] 100...Simulation system, 110...Control unit, 111...Simulation program, 111a...Acquisition unit, 111b...Generation unit, 111c...Specification unit, 111d...Output unit, 120...Communication I / F, 130...Display I / F, 200...Optical coherence tomography apparatus, 300...Storage medium, 300a...Measurement information, 300b...Non-inserted image, 300c...Inserted image, 300d...Machine learning model, 300e...Regression equation information, 400...Input unit, 500...Display unit

Claims

1. A simulation system comprising: an acquisition unit that acquires a non-insertion image, which is an image of the anterior segment of an eye to be examined in which a posterior chamber type phakic intraocular lens has not been inserted, taken by a tomography device; a generation unit that generates an insertion image, which is an image of the anterior segment of the eye to be examined in which the posterior chamber type phakic intraocular lens has been inserted, based on the size of the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined and the non-insertion image; an identification unit that identifies the optimal size, which is the optimal size for the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined, based on the insertion image; and an output unit that outputs information regarding the optimal size.

2. The simulation system according to claim 1, wherein the output unit outputs the optimal size.

3. The simulation system according to claim 1 or 2, wherein the output unit outputs the distance between the lens and the posterior chamber type phakic intraocular lens when the optimally sized posterior chamber type phakic intraocular lens is inserted into the eye under examination.

4. The simulation system according to claim 1 or 2, wherein the output unit outputs the insertion image obtained when the posterior chamber type phakic intraocular lens of the optimal size is inserted into the eye under examination.

5. The simulation system according to claim 1 or 2, wherein the identification unit identifies an optimal insertion direction, which is the optimal direction for inserting the posterior chamber type phakic intraocular lens of the optimal size into the anterior segment of the eye, based on the inserted image, and the output unit outputs information regarding the optimal insertion direction.

6. The simulation system according to claim 1 or 2, wherein the identifying unit determines the optimal size of the posterior chamber type phakic intraocular lens based on the insertion image when the posterior chamber type phakic intraocular lens is inserted in a direction corresponding to the astigmatism axis of the eye under examination, and if the optimal size cannot be determined, the direction in which the posterior chamber type phakic intraocular lens is inserted is changed to determine the optimal size.

7. The simulation system according to claim 1 or 2, wherein the identifying unit identifies the distance between the lens and the posterior chamber type phakic intraocular lens based on the inserted image for each size of the posterior chamber type phakic intraocular lens, and identifies the size of the posterior chamber type phakic intraocular lens as the optimal size when the distance is within a predetermined range.

8. A simulation method comprising: an acquisition step of acquiring a non-insertion image, which is an image of the anterior segment of an eye to be examined in which a posterior chamber type phakic intraocular lens has not been inserted, taken by a tomography device; a generation step of generating an insertion image, which is an image of the anterior segment of the eye to be examined in which the posterior chamber type phakic intraocular lens has been inserted, based on the size of the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined and the non-insertion image; a identification step of identifying an optimal size, which is the optimal size for the posterior chamber type phakic intraocular lens to be inserted into the eye to be examined, based on the insertion image; and an output step of outputting information regarding the optimal size.

9. A simulation program that causes a computer to function as: an acquisition unit that acquires an uninserted image, which is an image of the anterior segment of an eye without a posterior chamber phakic intraocular lens inserted, taken by a tomography device; a generation unit that generates an inserted image, which is an image of the anterior segment of the eye with the posterior chamber phakic intraocular lens inserted, based on the size of the posterior chamber phakic intraocular lens to be inserted into the eye and the uninserted image; an optimal size identification unit that identifies the optimal size for the posterior chamber phakic intraocular lens to be inserted into the eye, based on the inserted image; and an output unit that outputs information regarding the optimal size.

Images