Slit Lamp Microscope

Abstract

To widely provide high-quality slit lamp microscope inspection.SOLUTION: A slit lamp microscope according to an embodiment includes a scan part and a first evaluation part. The scan part is configured to collect an image group by scanning an anterior eye part of a subject eye with slit light. The first evaluation part is configured to evaluate an image reading property of the image group collected by the scan part by using an inference model constructed by machine learning.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a slit lamp microscope. [Background technology]

[0002] Diagnostic imaging plays an important role in the field of ophthalmology. Various ophthalmic imaging devices are used for diagnostic imaging. Types of ophthalmic imaging devices include slit lamp microscopes, fundus cameras, scanning laser ophthalmoscopes (SLO), and optical coherence tomography (OCT).

[0003] Among these various ophthalmic devices, the most widely and frequently used is the slit lamp microscope, which is used to illuminate the subject's eye with a slit light and observe or photograph the illuminated cross-section from an oblique or lateral direction using a microscope (see, for example, Patent Documents 1 and 2).

[0004] One of the main uses of slit lamp microscopes is the observation of the anterior segment of the eye. In anterior segment observation, doctors observe the entire anterior segment while moving the illumination field and focus position of the slit light to determine whether there is an abnormality. Slit lamp microscopes are also used in prescribing vision correction devices, such as checking the fitting of contact lenses. Furthermore, slit lamp microscopes are also used by people other than doctors, such as optometrists, paramedical staff, and eyeglass store clerks, for the purpose of screening for eye diseases.

[0005] In addition, with the recent progress of information and communication technology, research and development on telemedicine is progressing. Telemedicine is the act of providing medical treatment to patients in remote locations using a communication network such as the Internet. Patent Documents 3 and 4 disclose technologies for operating a slit lamp microscope from a remote location. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2016-159073 A [Patent Document 2] JP 2016-179004 A [Patent Document 3] JP 2000-116732 A [Patent Document 4] JP 2008-284273 A Summary of the Invention [Problem to be solved by the invention]

[0007] In general, to obtain good images using a slit lamp microscope, delicate and complicated operations such as adjusting the illumination angle and the imaging angle are required. In the techniques disclosed in Patent Documents 3 and 4, even when observing the eye of a subject who is in front of the examiner, an examiner in a remote location must perform difficult operations, which results in problems such as a long examination time and failure to obtain good images.

[0008] As described above, slit lamp microscopes are effective for screening and other examinations, but the current situation is that there is a shortage of people skilled in operating the device, making it difficult to provide high-quality slit lamp microscope examinations to many people.

[0009] Furthermore, in remote medical care and screening, the doctor who interprets the images (image diagnosis) is often not present at the imaging location. In such cases, images that are not suitable for interpretation may be provided to the doctor. Considering the difficulty of operation and the long imaging time, it is expected that this problem will be more likely to occur with slit lamp microscopes.

[0010] It is an object of the present invention to make high quality slit lamp microscopy widely available. [Means for solving the problem]

[0011] Some exemplary embodiments of the slit lamp microscope include a scanning unit that scans the anterior segment of the test eye with a slit light to collect a group of images, a first evaluation unit that evaluates the quality of the group of images collected by the scanning unit, and a control unit that selectively executes two or more predetermined controls depending on the evaluation result of the quality obtained by the first evaluation unit. Effect of the Invention

[0012] Exemplary aspects facilitate widespread availability of high quality slit lamp microscopy. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 2A] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Figure 2B] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Diagram 3] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Figure 4A] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 4B] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 4C] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 4D] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 5A] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 5B] FIG. 1 is a schematic diagram for explaining construction of an inference model of a slit lamp microscope according to an exemplary embodiment. [Figure 5C] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 5D] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 6] 1 is a flowchart illustrating the operation of a slit lamp microscope according to an exemplary embodiment. [Figure 7] 1 is a flowchart illustrating the operation of a slit lamp microscope according to an exemplary embodiment. [Figure 8] FIG. 1 is a schematic diagram showing the configuration of a slit lamp microscope according to an exemplary embodiment. [Figure 9A] 1 is a flowchart illustrating the operation of a slit lamp microscope according to an exemplary embodiment. [Figure 9B] 1 is a flowchart illustrating the operation of a slit lamp microscope according to an exemplary embodiment. [Figure 10A] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Figure 10B] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Figure 10C] 1 is a schematic diagram for explaining the operation of a slit lamp microscope according to an exemplary embodiment. FIG. [Figure 11] FIG. 1 is a schematic diagram illustrating a configuration of an ophthalmic system according to an exemplary embodiment. [Figure 12] FIG. 1 is a schematic diagram illustrating a configuration of an ophthalmic system according to an exemplary embodiment. [Figure 13] FIG. 1 is a schematic diagram illustrating a configuration of an ophthalmic system according to an exemplary embodiment. [Figure 14] FIG. 1 is a schematic diagram illustrating a configuration of an ophthalmic system according to an exemplary embodiment. [Figure 15] FIG. 1 is a schematic diagram illustrating a configuration of an ophthalmic system according to an exemplary embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Some exemplary aspects will be described in detail with reference to the drawings. Any known technology, such as those disclosed in the documents cited in this specification, may be combined with the exemplary aspects. The entire disclosure of other applications filed by the applicant may be incorporated by reference into the present disclosure. For example, the entire disclosure of applications filed by the applicant relating to a slit lamp microscope (or a system including a slit lamp microscope or a system connectable to a slit lamp microscope) may be incorporated by reference into the present disclosure.

[0015] The slit lamp microscope according to the exemplary embodiment may be a stationary type or a portable type. The slit lamp microscope according to the exemplary embodiment is typically used in a situation or environment where there is no expert (expert) in the device. The slit lamp microscope according to the exemplary embodiment may be used in a situation or environment where there is an expert nearby, or in a situation or environment where an expert can monitor, instruct, and operate the microscope from a remote location.

[0016] Examples of facilities where slit lamp microscopes are installed include eyeglass stores, optometrists, medical institutions, health checkup locations, medical examination locations, patients' homes, welfare facilities, public facilities, and medical examination vehicles.

[0017] The slit lamp microscope according to the exemplary embodiment is an ophthalmic imaging device having at least a function as a slit lamp microscope, and may further include other imaging functions (modalities). Examples of other modalities include an anterior eye camera, a fundus camera, SLO, and OCT. The slit lamp microscope according to the exemplary embodiment may further include a function for measuring characteristics of the subject's eye. Examples of the measurement functions include visual acuity measurement, refraction measurement, intraocular pressure measurement, corneal endothelial cell measurement, aberration measurement, and visual field measurement. The slit lamp microscope according to the exemplary embodiment may further include an application for analyzing captured images and measurement data. The slit lamp microscope according to the exemplary embodiment may further include a function for treatment or surgery. Examples of such a function include photocoagulation treatment and photodynamic therapy.

[0018] An ophthalmic system according to an exemplary embodiment (first ophthalmic system) may include one or more slit lamp microscopes, one or more information processing devices, and one or more image interpretation terminals, and may be usable for remote medical care, for example. The slit lamp microscope may be a slit lamp microscope according to any of the exemplary embodiments, or may be a slit lamp microscope including at least a part of the slit lamp microscope according to any of the exemplary embodiments.

[0019] The information processing device receives the images acquired by the slit lamp microscope and transmits them to the image interpretation terminal. The information processing device may also have a function of managing the images acquired by the slit lamp microscope.

[0020] The image reading terminal is a computer used by a doctor (typically a specialist such as an ophthalmologist or image reading physician) to read images acquired by a slit lamp microscope (observe the images to obtain clinical findings). Information input by the image reader into the image reading terminal may be converted, for example, by the image reading terminal or another computer into an image reading report or electronic medical record information and transmitted to an information processing device. In another example, information input by the image reader into the image reading terminal can be transmitted to an information processing device. In this case, the information processing device or another computer can convert the information input by the image reader into an image reading report or electronic medical record information. The information processing device may manage the image reading report or electronic medical record information by itself, or may transfer it to another medical system (for example, an electronic medical record system).

[0021] An ophthalmic system according to another exemplary embodiment (a second ophthalmic system) may include one or more slit lamp microscopes, one or more information processing devices, and one or more image reading devices. At least one of the slit lamp microscopes and the information processing devices may be the same as those of the first ophthalmic system.

[0022] The image reading device is, for example, a computer that uses an image processor that operates according to a predetermined program and / or an artificial intelligence engine (such as an inference engine or an inference model) to read images acquired by a slit lamp microscope. Information derived from an image by the image reading device may be converted, for example, by the image reading device or another computer into an image reading report or electronic medical record information and transmitted to an information processing device. In another example, the image reading device may transmit information derived from an image to an information processing device. In this case, the information processing device or another computer may convert the information derived from an image by the image reading device into an image reading report or electronic medical record information. The information processing device may manage the image reading report or electronic medical record information by itself, or may transfer it to another medical system.

[0023] An ophthalmic system according to yet another exemplary embodiment (third ophthalmic system) may include one or more slit lamp microscopes and one or more information processing devices. The slit lamp microscope may be similar to that of the first ophthalmic system. The information processing device may be similar to that of the first ophthalmic system and / or that of the second ophthalmic system. In other words, the information processing device may be capable of data communication with an image reading terminal of the first ophthalmic system and / or may be capable of data communication with an image reading device of the second ophthalmic system.

[0024] The slit lamp microscope and ophthalmic system according to the exemplary embodiment can be used in remote medical care. However, as described above, it is not easy to obtain a good image with a slit lamp microscope, and in order to effectively perform interpretation and diagnosis, it is necessary to obtain an image of a wide range of the anterior segment "in advance." Due to these circumstances, it can be said that effective remote medical care using a slit lamp microscope has not been realized. The exemplary embodiment can provide a technology that contributes to the realization of such remote medical care. However, the use of the exemplary embodiment is not limited to this, and the exemplary embodiment can also be applied to other uses.

[0025] The exemplary embodiment focuses on the following problem. That is, in the application form (remote medical care, etc.) assumed by the exemplary embodiment, a series of images (a group of images, a set of images) showing a sufficiently wide range of the anterior eye is directly or indirectly provided to the reader by the slit lamp microscope, so that even if the quality of the series of images provided is low, it is difficult to retake them, resulting in a problem that interpretation cannot be performed at all or only insufficient interpretation can be performed. Therefore, it is necessary to obtain images of "good" quality "in advance." In other words, it is necessary to obtain a series of images having a quality that allows effective diagnosis (interpretation, etc.) all at once before providing them to the reader. However, considering the difficulty of operating the slit lamp microscope and the occurrence of blinking and eye movement during imaging, it is extremely difficult to obtain a series of images that express the entire target area of ​​observation and interpretation with good image quality.

[0026] The slit lamp microscope according to the exemplary embodiment can be used to obtain a series of images (image group, image set) that represent a wide area of ​​the anterior segment of the eye with good image quality, and the ophthalmic system according to the exemplary embodiment can be used for remote medical treatment using such a slit lamp microscope.

[0027] Several exemplary aspects will be described below. Any two or more of these aspects can be at least partially combined. In addition, any known modification (addition, substitution, omission, etc.) can be applied to such combinations.

[0028] At least a portion of the functionality of the elements disclosed herein is implemented using circuitry or processing circuitry. The circuitry or processing circuitry may be a general purpose processor, a special purpose processor, an integrated circuit, a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), a field programmable gate array (FPGA), or a combination of a programmable logic device (PLC) and a programmable logic device (PGD)), configured and / or programmed to perform at least a portion of the functionality disclosed. The term "circuitry", "unit", "means" or the like includes any of the above-mentioned circuitry, including a transistor and / or other circuitry, a processor, a processor array, a conventional circuitry, and any combination thereof. A processor is considered to be a processing circuitry or circuitry, including transistors and / or other circuitry. In this disclosure, a circuitry, a unit, a means, or the like is hardware that performs at least some of the disclosed functions or is hardware that is programmed to perform at least some of the disclosed functions. The hardware may be hardware disclosed herein or may be known hardware that is programmed and / or configured to perform at least some of the described functions. In the case where the hardware is a processor that can be considered to be a type of circuitry, a circuitry, a unit, a means, or the like is a combination of hardware and software, and the software is used to configure the hardware and / or the processor.

[0029] <First aspect> An example of a slit lamp microscope according to the first embodiment is shown in FIG.

[0030] The slit lamp microscope 1 is used to photograph the anterior segment of a subject's eye E, and includes an illumination system 2, an imaging system 3, a moving mechanism 6, a control unit 7, a data processing unit 8, and a communication unit 9. Note that the symbol C denotes the cornea, and the symbol CL denotes the crystalline lens.

[0031] The slit lamp microscope 1 may be a single device or a system including two or more devices. As an example of the latter, the slit lamp microscope 1 includes a main body device including an illumination system 2, an imaging system 3, and a moving mechanism 6, a computer including a control unit 7, a data processing unit 8, and a communication unit 9, and a communication device that handles communication between the main body device and the computer. The computer may be installed, for example, together with the main body device or on a network.

[0032] <Lighting System 2> The illumination system 2 irradiates slit light onto the anterior segment of the subject's eye E. Reference symbol 2a indicates the optical axis (illumination optical axis) of the illumination system 2. The illumination system 2 may have a configuration similar to that of an illumination system of a conventional slit lamp microscope. For example, although not shown, the illumination system 2 includes, in order from the side farthest from the subject's eye E, an illumination light source, a positive lens, a slit forming unit, and an objective lens.

[0033] The illumination light source outputs illumination light. The illumination system 2 may include a plurality of illumination light sources. For example, the illumination system 2 may include an illumination light source that outputs continuous light and an illumination light source that outputs flash light. The illumination system 2 may also include an illumination light source for the anterior segment and an illumination light source for the posterior segment. The illumination system 2 may also include two or more illumination light sources with different output wavelengths. A typical illumination system 2 includes a visible light source as an illumination light source. The illumination system 2 may also include an infrared light source. The illumination light output from the illumination light source passes through a positive lens and is projected onto the slit forming portion.

[0034] The slit forming unit generates slit light by passing a part of the illumination light. A typical slit forming unit has a pair of slit blades. The width of the area (slit) through which the illumination light passes is changed by changing the interval (slit width) between these slit blades, and thus the width of the slit light is changed. The slit forming unit may be configured to change the length of the slit light. The length of the slit light is the cross-sectional dimension of the slit light in a direction perpendicular to the cross-sectional width direction of the slit light corresponding to the slit width. The width and length of the slit light are typically expressed as the dimension of the image of the slit light projected onto the anterior segment of the eye, but are not limited thereto. For example, they can be expressed as the dimension of the cross section of the slit light at any position, or as the dimension of the slit formed by the slit forming unit.

[0035] The slit light generated by the slit forming unit is refracted by the objective lens and irradiated onto the anterior segment of the subject's eye E.

[0036] The illumination system 2 may further include a focusing mechanism for changing the focus position of the slit light. The focusing mechanism, for example, moves an objective lens along the illumination optical axis 2a. The movement of the objective lens can be performed automatically and / or manually. Note that a focusing lens may be disposed at a position on the illumination optical axis 2a between the objective lens and the slit forming part, and the focus position of the slit light may be changed by moving this focusing lens along the illumination optical axis 2a.

[0037] 1 is a top view, and as shown in the figure, in this embodiment, the direction along the axis of the subject's eye E is the Z direction, the direction perpendicular to the Z direction from left to right with respect to the subject is the X direction, and the direction perpendicular to both the X direction and the Z direction is the Y direction. Typically, the X direction is the arrangement direction of the left eye and the right eye, and the Y direction is the direction along the subject's body axis (body axis direction).

[0038] <Photography 3> The imaging system 3 images the anterior eye segment illuminated with the slit light from the illumination system 2. Reference numeral 3a denotes the optical axis (imaging optical axis) of the imaging system 3. The imaging system 3 in this embodiment includes an optical system 4 and an image sensor 5.

[0039] The optical system 4 guides light from the anterior segment of the subject's eye E, which is irradiated with the slit light, to the imaging element 5. The imaging element 5 receives the light guided by the optical system 4 on an imaging surface.

[0040] The light guided by the optical system 4 (i.e., light from the anterior segment of the subject's eye E) includes returned light of the slit light irradiated on the anterior segment, and may further include other light. Examples of returned light include reflected light, scattered light, and fluorescence. Examples of other light include light from the installation environment of the slit lamp microscope 1 (room light, sunlight, etc.). When an anterior segment illumination system for illuminating the entire anterior segment is provided separately from the illumination system 2, the returned light of this anterior segment illumination light may be included in the light guided by the optical system 4.

[0041] The imaging element 5 is an area sensor having a two-dimensional imaging area, and may be, for example, a charge-coupled device (CCD) image sensor or a complementary metal-oxide semiconductor (CMOS) image sensor.

[0042] The optical system 4 may have a configuration similar to that of the imaging system of a conventional slit lamp microscope. For example, the optical system 4 includes, in order from the side closer to the subject's eye E, an objective lens, a variable magnification optical system, and an imaging lens. Light from the anterior segment of the subject's eye E irradiated with the slit light passes through the objective lens and the variable magnification optical system, and is imaged on the imaging surface of the image sensor 5 by the imaging lens.

[0043] The imaging system 3 may include, for example, a first imaging system and a second imaging system. Typically, the first imaging system and the second imaging system have the same configuration. The case where the imaging system 3 includes the first imaging system and the second imaging system will be described in another embodiment.

[0044] The imaging system 3 may further include a focusing mechanism for changing the focus position. The focusing mechanism, for example, moves an objective lens along the imaging optical axis 3a. The movement of the objective lens can be performed automatically and / or manually. Note that a focusing lens may be disposed at a position on the imaging optical axis 3a between the objective lens and the imaging lens, and the focus position may be changed by moving this focusing lens along the imaging optical axis 3a.

[0045] The illumination system 2 and the photographing system 3 function as a Scheimpflug camera. That is, the illumination system 2 and the photographing system 3 are configured so that the object plane along the illumination optical axis 2a, the optical system 4, and the imaging plane of the image sensor 5 satisfy the so-called Scheimpflug condition. More specifically, the YZ plane (including the object plane) passing through the illumination optical axis 2a, the principal plane of the optical system 4, and the imaging plane of the image sensor 5 intersect on the same straight line. This allows photographing to be performed by focusing on all positions in the object plane (all positions in the direction along the illumination optical axis 2a).

[0046] In this embodiment, for example, the illumination system 2 and the imaging system 3 are configured so that the imaging system 3 is focused on at least the range defined by the anterior surface of the cornea C and the posterior surface of the crystalline lens CL. In other words, imaging can be performed in a state in which the imaging system 3 is focused on the entire range from the apex (Z=Z1) of the anterior surface of the cornea C to the apex (Z=Z2) of the posterior surface of the crystalline lens CL shown in Fig. 1. Note that Z=Z0 indicates the Z coordinate of the intersection of the illumination optical axis 2a and the imaging optical axis 3a.

[0047] Such conditions are typically realized by the configuration and arrangement of elements included in the illumination system 2, the configuration and arrangement of elements included in the imaging system 3, and the relative positions of the illumination system 2 and the imaging system 3. The parameters indicating the relative positions of the illumination system 2 and the imaging system 3 include, for example, the angle θ between the illumination optical axis 2a and the imaging optical axis 3a. The angle θ is set to, for example, 17.5 degrees, 30 degrees, or 45 degrees. Note that the angle θ may be variable.

[0048] <Movement mechanism 6> The moving mechanism 6 moves the illumination system 2 and the imaging system 3. The moving mechanism 6 includes, for example, a movable stage on which the illumination system 2 and the imaging system 3 are mounted, an actuator that operates according to a control signal input from the control unit 7, and a mechanism that moves the movable stage based on a driving force generated by the actuator. In another example, the moving mechanism 6 includes a movable stage on which the illumination system 2 and the imaging system 3 are mounted, and a mechanism that moves the movable stage based on a force applied to an operating device (not shown). The operating device is, for example, a lever. The movable stage is movable at least in the X direction, and may further be movable in the Y direction and / or Z direction.

[0049] In this embodiment, the moving mechanism 6 moves the illumination system 2 and the imaging system 3 in the X direction, for example. That is, the moving mechanism 6 moves the illumination system 2 and the imaging system 3 in the X direction while maintaining a state in which the above-mentioned Scheimpflug condition is satisfied. In parallel with this movement, the imaging system 3 captures a moving image at a predetermined time interval (imaging rate), for example. As a result, a three-dimensional region of the anterior segment of the subject's eye E is scanned with slit light, and a plurality of images (a group of images) corresponding to a plurality of cross sections within this three-dimensional region are collected.

[0050] <Control unit 7> The control unit 7 controls each part of the slit lamp microscope 1. For example, the control unit 7 controls the elements of the illumination system 2 (illumination light source, slit forming unit, focusing mechanism, etc.), the elements of the imaging system 3 (focusing mechanism, image sensor, etc.), the movement mechanism 6, the data processing unit 8, the communication unit 9, etc. The control unit 7 may also be capable of executing control for changing the relative positions of the illumination system 2 and the imaging system 3.

[0051] The control unit 7 includes a processor, a main storage device, an auxiliary storage device, etc. The auxiliary storage device stores a control program, etc. The control program, etc. may be stored in a computer or storage device accessible to the slit lamp microscope 1. The functions of the control unit 7 are realized by cooperation between software such as the control program and hardware such as a processor.

[0052] In order to scan a three-dimensional area of ​​the anterior segment of the subject's eye E with slit light, the control unit 7 can apply the following controls to the illumination system 2, the imaging system 3, and the moving mechanism 6.

[0053] First, the control unit 7 controls the moving mechanism 6 to place the illumination system 2 and the imaging system 3 at a predetermined scan start position (alignment control). The scan start position is, for example, a position corresponding to an end (first end) of the cornea C in the X direction, or a position further away from the axis of the subject's eye E. Reference symbol X0 in FIG. 2A shows an example of a scan start position corresponding to the first end of the cornea C in the X direction. Reference symbol X0' in FIG. 2B shows an example of a scan start position further away from the axis EA of the subject's eye E than the position corresponding to the first end of the cornea C in the X direction.

[0054] The control unit 7 controls the illumination system 2 to start irradiating the anterior segment of the subject's eye E with slit light (slit light irradiation control). Incidentally, the slit light irradiation control may be performed before or during the execution of the alignment control. The slit light is typically continuous light, but may also be intermittent light (pulsed light). The lighting control of the pulsed light is synchronized with the imaging rate of the imaging system 3. Furthermore, the slit light is typically visible light, but may also be infrared light, or a mixture of visible light and infrared light.

[0055] The control unit 7 controls the imaging system 3 to start imaging of a moving image of the anterior segment of the subject's eye E (imaging control). Incidentally, the imaging control may be performed before or during the execution of the alignment control. Typically, the imaging control is performed simultaneously with or after the slit light irradiation control.

[0056] After performing the alignment control, the slit light irradiation control, and the photographing control, the control unit 7 controls the moving mechanism 6 to start moving the illumination system 2 and the photographing system 3 (movement control). The movement control moves the illumination system 2 and the photographing system 3 together. That is, the illumination system 2 and the photographing system 3 are moved while maintaining the relative position (angle θ, etc.) between the illumination system 2 and the photographing system 3. Typically, the illumination system 2 and the photographing system 3 are moved while maintaining a state in which the above-mentioned Scheimpflug condition is satisfied. The illumination system 2 and the photographing system 3 are moved from the above-mentioned scan start position to a predetermined scan end position. The scan end position is, for example, a position corresponding to the end (second end) of the cornea C on the opposite side of the first end in the X direction, similar to the scan start position, or a position further away from the axis of the subject's eye E. In such a case, the range from the scan start position to the scan end position is the scan range.

[0057] Typically, a slit light having the X direction as the width direction and the Y direction as the length direction is irradiated onto the anterior segment, and the illumination system 2 and the imaging system 3 are moved in the X direction, while the imaging system 3 captures a moving image.

[0058] Here, the length of the slit light (i.e., the dimension of the slit light in the Y direction) is set, for example, to be equal to or greater than the diameter of the cornea C on the surface of the subject's eye E. That is, the length of the slit light is set to be equal to or greater than the corneal diameter in the Y direction. Also, as described above, the movement distance of the illumination system 2 and the imaging system 3 by the movement mechanism 6 (i.e., the scanning range) is set to be equal to or greater than the corneal diameter in the X direction. This makes it possible to scan at least the entire cornea C with the slit light.

[0059] By such scanning, multiple anterior eye images with different irradiation positions of the slit light are obtained. In other words, a moving image is obtained that depicts the irradiation position of the slit light moving in the X direction. An example of such multiple anterior eye images (i.e., a group of frames constituting a moving image) is shown in FIG. 3.

[0060] FIG. 3 shows a plurality of anterior eye images (frame groups, image groups) F1, F2, F3, . . . , FN. The subscript n of these anterior eye images Fn (n=1, 2, . . . , N) represents the time series order. That is, the anterior eye image acquired n-th is represented by the symbol Fn. The anterior eye image Fn includes a slit light irradiation area An. As shown in FIG. 3, the slit light irradiation areas A1, A2, A3, . . . , AN move rightward along the time series. In the example shown in FIG. 3, the scan start position and the scan end position correspond to both ends of the cornea C in the X direction. Note that the scan start position and / or the scan end position are not limited to this example, and may be, for example, a position farther from the axis of the subject's eye E than the edge of the cornea. In addition, the direction and number of scans can also be set arbitrarily.

[0061] <Data Processing Unit 8> The data processing unit 8 executes various types of data processing. The data to be processed may be either data acquired by the slit lamp microscope 1 or data input from the outside. For example, the data processing unit 8 can process images acquired by the imaging system 3. The configuration and functions of the data processing unit 8 will be described in this embodiment as well as in other embodiments.

[0062] The data processing unit 8 includes a processor, a main storage device, an auxiliary storage device, etc. The auxiliary storage device stores a data processing program, etc. The data processing program, etc. may include a model (trained model, inference model, etc.) constructed by machine learning. The data processing program, etc. may be stored in a computer or storage device accessible to the slit lamp microscope 1. The functions of the data processing unit 8 are realized by cooperation between software such as the data processing program and hardware such as a processor.

[0063] Several examples of the data processing unit 8 will be described. Figures 4A, 4B, 4C, and 4D show data processing units 8A, 8B, 8C, and 8D, which are first, second, third, and fourth examples of the data processing unit 8, respectively. Note that the configuration of the data processing unit 8 is not limited to these. For example, the data processing unit 8 may include a combination of any two or more of the four data processing units 8A, 8B, 8C, and 8D. In addition, it is possible to provide the data processing unit 8 with any element for obtaining the same type of result or a similar type of result.

[0064] The slit lamp microscope 1 of this embodiment applies one or more scans to the anterior segment of the subject's eye E. Each scan results in a group of images such as the multiple anterior segment images F1 to FN shown in Fig. 3. The data processing unit 8 can process the group of one or more images thus obtained.

[0065] The slit lamp microscope 1 may be capable of applying two or more scans to the anterior segment. For example, if a good set of images is not obtained in the first scan, a second scan is performed. More generally, if a good set of images is not obtained in the first to v scans, a v+1 scan is performed (v is an integer equal to or greater than 1). In some exemplary aspects, if a good set of images (image set) is not obtained from the v image sets obtained in the first to v scans, a v+1 scan is performed. In some other exemplary aspects, if none of the v image sets obtained in the first to v scans is a good image set, a v+1 scan is performed. Note that the manner in which two or more scans are performed is not limited to these.

[0066] When two or more scans are performed, the start and end positions (i.e., the scan range) of these scans may be the same, or one or both of the start and end positions may be different. Typically, two or more scans are applied to the same scan range. As a result, a group of images such as the multiple anterior eye images F1 to FN shown in FIG. 3 is obtained in the same number as the number of scans.

[0067] Considering the eye movement of the subject's eye E, the scan range is typically defined not by the position or range in the anterior segment but by the operation of the slit lamp microscope 1. For example, the starting point of the movement of the illumination system 2 and the imaging system 3 during scanning is set as the scan start position, and the end point is set as the scan end position.

[0068] On the other hand, it is also possible to define the scan range according to the position or range in the anterior segment. In this case, an operation (tracking) is applied in which the illumination system 2 and the imaging system 3 are moved so as to follow the movement of the subject's eye E. The slit lamp microscope 1 of this example has, for example, a tracking function similar to that of a conventional one. The tracking function includes, for example, video capture of the anterior segment, extraction of landmarks from each frame, calculation of displacement between frames using the landmarks, and movement control of the illumination system 2 and the imaging system 3 to cancel the displacement.

[0069] As another example of the definition (setting) of the scan range, it is possible to set the scan range for an image that has already been collected. In other words, it is possible to configure the setting of the scan range to be performed after the fact (after scanning). Note that the scan range in this example represents the range of the image to be provided for later processing.

[0070] A first example of such a post-event scan range setting will be described. The slit lamp microscope of this example is capable of anterior eye photography, and is capable of recognizing the positional relationship between the anterior eye photography optical system and the scanning optical system (illumination system 2 and photography system 3). In this example, first, an image group (wide-area image group) is collected by scanning a sufficiently wide range (particularly in the X direction and Y direction) of the anterior eye, while performing anterior eye photography. Next, a scan range is set for the anterior eye image obtained by anterior eye photography. This setting may be performed either manually or automatically. The automatic setting includes, for example, a process of analyzing the anterior eye image to detect a landmark (for example, a pupil margin), and a process of setting the scan range by referring to the landmark. When the scan range is set, a partial area of ​​each wide-area image corresponding to this scan range is specified based on the positional relationship of the optical system described above. Finally, an image corresponding to the scan range is formed by cropping the specified partial area from the wide-area image. As a result, an image group corresponding to the set scan range is obtained. In this example, when the fixation of the subject's eye is stable (or when it is assumed or can be assumed), the anterior eye may be photographed as a still image. When the anterior eye is photographed as a moving image, the wide-area image group and the frame group in the moving image are associated with each other based on, for example, the control (synchronization information, etc.) between the scan and the moving image photographing, and the same processing as described above is performed for each pair of the associated wide-area image and frame.

[0071] A second example of the post-scan range setting will be described. In this example, the anterior eye does not need to be photographed in parallel, and a sufficiently wide range (particularly in the X-direction and Y-direction) of the anterior eye is scanned to collect a group of wide-area images. Next, a partial area corresponding to a desired scan range is specified for each wide-area image. This specification may be performed either manually or automatically. The automatic specification includes, for example, a process of analyzing the wide-area image to detect a landmark (for example, the corneal limbus or the iridocorneal angle) and a process of setting the scan range by referring to the landmark. Alternatively, a landmark may be manually specified for any wide-area image (reference wide-area image), and the reference wide-area image and an adjacent wide-area image may be analyzed to specify a landmark for the adjacent wide-area image. By sequentially applying such processing, it is possible to specify landmarks for all wide-area images. Furthermore, a partial area of ​​the wide-area image corresponding to the scan range can be specified based on the landmark, and an image corresponding to the scan range can be formed by cropping the specified partial area from the wide-area image. As a result, a group of images corresponding to the set scan range can be obtained.

[0072] A first example of the data processing unit 8 will be described. The data processing unit 8A shown in FIG. 4A includes an image group evaluation unit 81. The image group evaluation unit 81 evaluates the quality of an image group collected in one scan of the anterior segment of the subject's eye E. In some exemplary embodiments, the image group evaluation unit 81 is configured to determine whether the image group satisfies a predetermined condition, and typically is configured to determine whether each image included in the image group satisfies the predetermined condition.

[0073] The predetermined condition (image group evaluation condition) is, for example, a condition related to image quality required for effective interpretation and diagnosis, and may be, for example, a condition related to the subject's eye E, a condition related to the slit lamp microscope 1, a condition related to the environment, etc. The image group evaluation condition may include, for example, either or both of "the image to be evaluated is not affected by blinking (blink condition)" and "the image to be evaluated is not affected by eye movement (eye movement condition)". In addition, the image group evaluation condition may be any index related to image quality evaluation, and may include, for example, any of brightness, contrast, noise, signal-to-noise ratio (SNR), tone reproduction, dynamic range, sharpness, vignetting, aberration (spherical aberration, astigmatism, coma aberration, field curvature, distortion aberration, chromatic aberration, etc.), color accuracy, exposure accuracy, lens flare, moire, and artifacts. The image selection condition is not limited to these examples and may be selected arbitrarily.

[0074] The blinking condition will be described. For an image included in the image group, the image group evaluation unit 81 judges whether or not the image contains a reflected image of slit light irradiated to the anterior part of the subject's eye E. This judgment is a process that utilizes the fact that the reflected image of slit light is not depicted in an image taken during a blink and that the reflected image of slit light is expressed significantly brighter than other regions, and is performed, for example, based on the luminance distribution in the image.

[0075] As an example, the image group evaluation unit 81 creates a luminance histogram from the image and judges whether or not there are pixels with a luminance equal to or greater than a predetermined threshold. If it is judged that there is a luminance equal to or greater than the predetermined threshold, it is judged that the image contains a reflected image of slit light. This example has the advantage that the processing is extremely simple, but there is a risk of erroneously detecting high-luminance noise or reflections of external light.

[0076] As another example, the image group evaluation unit 81 creates a luminance histogram from the image and judges whether or not there are a predetermined number of pixels with a luminance equal to or greater than a predetermined threshold. If it is judged that there are a predetermined number of pixels with a luminance equal to or greater than a predetermined threshold, it is judged that the image contains a reflected image of slit light. This example has the advantage that the above-mentioned erroneous detection can be prevented by simple processing.

[0077] The eye movement conditions will now be described. For each image included in the image group, the image group evaluation unit 81 judges whether or not the image is affected by eye movement, for example, by comparing the image with its adjacent images. This judgment is a process that utilizes the fact that "image jumps" occur when eye movement occurs during video capture.

[0078] As an example, the image group evaluation unit 81 detects landmarks from the image and the adjacent images, calculates the amount of displacement of these landmarks, and judges whether the amount of displacement is equal to or greater than a predetermined threshold. If it is determined that the amount of displacement is equal to or greater than the predetermined threshold, it is determined that eye movement has occurred. Here, the landmarks may be, for example, the cornea, iris, pupil, or angle of the eye. The threshold is calculated based on predetermined scan conditions, such as the imaging rate of the imaging system 3 and the moving speed of the moving mechanism 6.

[0079] In another example, the image group evaluation unit 81 may be configured to determine the presence or absence of eye movement from one image. For example, when the shooting rate of the shooting system 3 is slow, if a fast eye movement occurs, a "blur" may occur in the image. The image group evaluation unit 81 can determine the presence or absence of eye movement by utilizing blur detection. Blur detection is typically performed using a known technique such as edge detection.

[0080] The image group evaluation unit 81 may include an artificial intelligence engine for determining whether the input image satisfies the above image selection conditions. This artificial intelligence engine typically includes a convolutional neural network (CNN), and this convolutional neural network is trained in advance using training data including a large number of images acquired by a slit lamp microscope and a determination result of whether each image satisfies the image selection conditions. Note that the images included in the training data are not limited to images acquired by a slit lamp microscope, and may be, for example, images acquired by other ophthalmic modalities (fundus cameras, OCT devices, SLOs, surgical microscopes, etc.), images acquired by image diagnostic modalities of other medical departments (ultrasound diagnostic devices, X-ray diagnostic devices, X-ray CT devices, magnetic resonance imaging (MRI) devices, etc.), images generated by processing actual images, pseudo images, etc. In addition, the method and technology (for example, hardware, software, machine learning methods, types of neural networks, etc.) used in the artificial intelligence engine are arbitrary.

[0081] A series of images constituting an image group collected in one scan may be associated with a plurality of positions in the scan range. This association is performed, for example, by the data processing unit 8. A specific example will be described. The scan range along the X direction is divided into (N-1) sections, the scan start position is set to the first position, and the scan end position is set to the Nth position (N is an integer equal to or greater than 2). As a result, N positions are set in the scan range. The N positions are represented by B1, B2, B3, . . . , BN. It is assumed that one scan is applied to the anterior segment of the subject's eye E, and an image group F1, F2, F3, . . . , FN shown in FIG. 3 is obtained. The data processing unit 8 can assign an image Fn to the position Bn. As a result, N images Fn (n=1, 2, . . . , N) corresponding to the N positions Bn (n=1, 2, . . . , N) are obtained.

[0082] The image group may include only a series of images, or may further include other information. Examples of information included in the image group together with the series of images include various types of additional information such as subject information, subject eye information, shooting date and time, and shooting conditions. Images obtained by other modalities and test data acquired by an examination device can also be included in the image group. Some examples of the configuration and operation of the image group evaluation unit 81 will be described later with reference to Figs. 5A to 5D.

[0083] A second example of the data processing unit 8 will be described. The effectiveness of this example is particularly demonstrated when two or more scans are applied to the anterior segment of the subject's eye E. The data processing unit 8B shown in FIG. 4B includes an image set creation unit 82 in addition to the image group evaluation unit 81 similar to that of FIG. 4A. Two or more image groups collected by two or more scans applied to the anterior segment of the subject's eye E are provided to the data processing unit 8B. The image group evaluation unit 81 of this example may perform quality evaluation on each of these image groups, or may perform quality evaluation only on the image group collected in the first scan (first scan). The image set creation unit 82 selects a series of images corresponding to the scan range from the two or more image groups input to the data processing unit 8B to create an image set.

[0084] A series of images constituting an image set may be, for example, a range (of the anterior eye) to which any one of the two or more scans is applied, or a range (of the anterior eye) set based on scan ranges in at least two of the two or more scans. As an example of the former, a maximum scan range or a minimum scan range among two or more scan ranges to which the two or more scans are applied, respectively, may be adopted. As an example of the latter, a union or intersection of at least two scan ranges may be adopted.

[0085] The image set may include only the series of images, or may further include other information. Examples of information included in the image set together with the series of images include various types of additional information such as subject information, subject eye information, shooting date and time, shooting conditions, etc. Images obtained by other modalities and test data obtained by an examination device may also be included in the image set.

[0086] A third example of the data processing unit 8 will be described. As with the second example, the effectiveness of this example is particularly evident when two or more scans are applied to the anterior segment of the subject's eye E. The data processing unit 8C shown in FIG. 4C includes an image set creation unit 82 similar to that in FIG. 4B in addition to an image group evaluation unit 81 similar to that in FIG. 4A, but the image set creation unit 82 of this example includes a selection unit 821.

[0087] The data processing unit 8C is provided with two or more image groups collected in two or more scans applied to the anterior segment of the subject's eye E. The image group evaluation unit 81 in this example may perform quality evaluation on each of these image groups, or may perform quality evaluation only on the image group collected in the initial scan (first scan). The image set creation unit 82 selects a series of images corresponding to the scan range from the two or more image groups input to the data processing unit 8B to create an image set. In creating this image set, the selection unit 821 selects images that satisfy predetermined conditions from the two or more image groups.

[0088] The predetermined condition (image selection condition) may be the same as the image group evaluation condition described above, or may be different. For example, when the image group evaluation unit 81 is configured to provide two or more image groups to the selection unit 821 after the quality evaluation of each image is performed by the image group evaluation unit 81, the selection unit 821 may be configured to select images taking into consideration conditions related to image arrangement (e.g., blink conditions, eye movement conditions, etc.). Note that the image selection conditions are not limited to these examples, and the relationship between the image group evaluation conditions and the image selection conditions is not limited to these examples.

[0089] The following describes the case where the selection unit 821 considers the blink condition and the eye movement condition. In addition, when considering conditions other than these, the selection unit 821 can execute the same process as a specific example (see Figs. 5A to 5D) related to the image group evaluation unit 81 described later.

[0090] Image selection taking blinking conditions into consideration will be described. The blinking conditions are conditions for confirming that "the image to be evaluated is not affected by blinking." For an image included in two or more image groups, the selection unit 821 judges, for example, whether the image contains a reflected image of slit light irradiated to the anterior segment of the subject's eye E. This judgment is a process that utilizes the fact that the reflected image of the slit light is not depicted in an image captured during a blink, and that the reflected image of the slit light is expressed significantly brighter than other regions, and is performed, for example, based on the luminance distribution in the image.

[0091] As an example, the selection unit 821 creates a luminance histogram from the image and judges whether or not there are pixels with a luminance equal to or greater than a predetermined threshold. If it is judged that there is a luminance equal to or greater than the predetermined threshold, it is judged that the image contains a reflected image of slit light. This example has the advantage that the processing is extremely simple, but there is a risk of erroneously detecting high-luminance noise or reflections of external light.

[0092] As another example, the selection unit 821 creates a luminance histogram from the image and determines whether there are a predetermined number of pixels with a luminance equal to or greater than a predetermined threshold. If it is determined that there are a predetermined number of pixels with a luminance equal to or greater than a predetermined threshold, it is determined that the image contains a reflected image of slit light. This example has the advantage that the above-mentioned erroneous detection can be prevented by simple processing.

[0093] Image selection taking into consideration eye movement conditions will be described. The eye movement conditions are conditions for confirming that "the image to be evaluated is not affected by eye movement." For an image included in two or more image groups, the selection unit 821 determines whether or not the image is affected by eye movement, for example, by comparing the image with its adjacent images. This determination is a process that utilizes the fact that "image jumps" occur when eye movement occurs during video capture.

[0094] As an example, the selection unit 821 detects landmarks from each of the image and the adjacent image, calculates the amount of displacement of these landmarks, and judges whether the amount of displacement is equal to or greater than a predetermined threshold. If it is determined that the amount of displacement is equal to or greater than the predetermined threshold, it is determined that eye movement has occurred. Here, the landmarks may be, for example, the cornea, iris, pupil, or angle of the eye. The threshold is calculated based on predetermined scan conditions, such as the imaging rate of the imaging system 3 and the moving speed of the moving mechanism 6.

[0095] In another example, the selection unit 821 may be configured to determine the presence or absence of eye movement from one image. For example, when the imaging rate of the imaging system 3 is slow, if a rapid eye movement occurs, a "blur" may occur in the image. The selection unit 821 can determine the presence or absence of eye movement by utilizing blur detection. The blur detection is typically performed using a known technique such as edge detection.

[0096] As with the image collection evaluator 81, the selector 821 may include an artificial intelligence engine for determining whether the input images satisfy the image selection criteria described above.

[0097] A series of images included in the image set created by the image set creation unit 82 may be associated with multiple positions in the scan range. For example, the selection unit 821 may be configured to select images such that one or more images are assigned to each of multiple positions in the scan range.

[0098] A specific example will be described. The scan range along the X direction is divided into (N-1) sections, the scan start position is set to the first position, and the scan end position is set to the Nth position (N is an integer equal to or greater than 2). In this way, N positions are set in the scan range. The N positions are represented as B1, B2, B3, ..., BN.

[0099] Assume that two or more scans are applied to the anterior segment of the subject's eye E, resulting in the acquisition of a group of images F1, F2, F3, ..., FN shown in Fig. 3. Note that, in the above description, the N images F1 to FN shown in Fig. 3 are considered to be a group of images acquired by one scan, but in this specification, for the sake of simplicity, the N images F1 to FN shown in Fig. 3 may be referred to as an arbitrary group of images (multiple images) (hereinafter the same). For example, in the description of this example, the N images F1 to FN are referred to as a series of images included in an image set.

[0100] For such N positions B1 to BN and N images F1 to FN, the selection unit 821 can assign an image Fn to the position Bn. This provides N images Fn (n=1, 2,...,N) corresponding to the N positions Bn (n=1, 2,...,N), and creates an image set in which the image group Fn is a "series of images."

[0101] In some exemplary embodiments, when the image group evaluation unit 81 evaluates that the quality of the image group obtained in the first scan is not good, a second scan is performed automatically or in response to a user's instruction. When the number of scans applied to the anterior segment is two or more, the slit lamp microscope 1 performs an operation of applying two or more scans to the anterior segment and an operation of selecting a series of images from two or more image groups collected thereby, but the manner of performing these operations may be arbitrary. As a first example, the slit lamp microscope 1 may be configured to alternately repeat a scan of the anterior segment and a selection of images from the image group acquired by this scan in response to the evaluation that the quality of the image group obtained in the first scan is not good. As a second example, the slit lamp microscope 1 may be configured to perform two or more scans together and then select a series of images from two or more image groups collected thereby in response to the evaluation that the quality of the image group obtained in the first scan is not good. These two examples will be described below. Note that instead of or in addition to the selection of images from the image group, the image group evaluation unit 81 may evaluate the image group. As previously discussed, image group evaluation and image selection may be similar, analogous, or interchangeable processes, and in some exemplary aspects, matters described as image group evaluation may be applied to image selection, and conversely, matters described as image selection may be applied to image group evaluation.

[0102] A first example of the mode of scanning and image selection executed after the image group evaluation unit 81 evaluates that the quality of the image group obtained by the first scan is not good is alternating repetition of scanning and image selection. More specifically, in the first example, for example, a set of application of a scan to the anterior eye segment and selection of an image from the image group obtained by this scan is repeatedly executed a predetermined number of times. That is, in the first example, U sets of operations are executed in the order of the first set of operations (scanning and image selection), the second set of operations (scanning and image selection), ..., the Uth set of operations (scanning and image selection) (U is an integer equal to or greater than 1). In addition, the set of the first scan performed before this alternating repetition and the quality evaluation (image group evaluation unit 81) of the image group obtained by the first scan will be called the 0th set of operations.

[0103] Here, the number of scans in the u-th set may be any number equal to or greater than 1 (u=0, 1,...,U). Also, the number of scans in the u1-th set and the number of scans in the u2-th set may be equal to or different from each other (u1=0, 1,...,U; u2=0, 1,...,U; u1≠u2).

[0104] In a first example, the selection unit 821 may be configured to select images from two or more image groups collected in two or more scans that have already been performed to create a provisional image set. That is, the slit lamp microscope 1 may be configured to create a provisional image set from two or more image groups obtained in two or more scans performed up to that point in time during the alternating repetition of scanning and image selection. For example, after the scan in the u-th set is performed, the selection unit 821 is configured to create a provisional image set from all images obtained in the 0-th set to the u-th set. With this configuration, a provisional image set can be created from the images obtained up to the current point in time to build a final image set.

[0105] When the above configuration for creating a provisional image set is applied, the following configuration can be combined. When a new scan is applied to the anterior segment of the eye E, the selection unit 821 first selects an image from a new image group collected in the new scan. Then, the selection unit 821 creates a new provisional image set by adding an image selected from the new image group to a provisional image set based on one or more scans performed before the new scan. For example, after a scan in the (u+1)th set is performed, the selection unit 821 can first select an image from the image group obtained in the (u+1)th set. Furthermore, the selection unit 821 can create a new provisional image set by adding an image selected from the image group obtained in the (u+1)th set to a provisional image set based on the image group obtained in the 0th set to the uth set. According to this configuration, each time a scan is applied to the anterior segment, the provisional image set can be sequentially updated based on the image group obtained in the scan. This makes it possible to reliably and efficiently construct a final image set.

[0106] When the above configuration for creating (and updating) a provisional image set is applied, the following configuration can be combined. The control unit 7 (or the image set creation unit 82 (selection unit 821)) includes an image number counter that counts the number of images included in the provisional image set. When the number of images included in the provisional image set reaches a predetermined number, the control unit 7 controls the scanning unit (illumination system 2, shooting system 3, movement mechanism 6) and the selection unit 821 to terminate the alternating repetition of the application of scan and the selection of images. Here, the predetermined number is the number of a series of images included in the final image set, and is set in advance or from the processing status. In addition, the control unit 7 determines whether the number of images included in the provisional image set has reached the predetermined number. This determination may be made only by comparing the numbers. Alternatively, when a series of images is associated with a plurality of positions in the scan range (as described above), it may be determined whether corresponding images have been assigned to all of the plurality of positions. According to this configuration, when the number required for the final image set is obtained, the alternating repetition of the scan and the selection of images can be automatically terminated.

[0107] When the above-mentioned configuration for creating (and updating) a tentative image set is applied, the following configuration can be further combined. The control unit 7 includes a repetition counter that counts the number of times the alternating repetition of scanning and image selection is performed. The number of times to be counted may be defined in units of pairs of scanning and image selection (the first pair to the Uth pair), or may be defined in units of the number of scans. When the number of repetitions reaches a predetermined number, the control unit 7 controls the scanning unit (illumination system 2, imaging system 3, moving mechanism 6) and the selection unit 821 so as to end the alternating repetition of the application of scanning and the selection of images. When the number of repetitions is defined in units of pairs of scanning and image selection, the predetermined number is equal to the total number of pairs (U+1) set in advance. When the number of repetitions is defined in units of the number of scans, the predetermined number is equal to the total number of scans set in advance. In addition, the control unit 7 determines whether the number of repetitions has reached the predetermined number. According to this configuration, it is possible to automatically end the repetition of scanning and image selection when it has been repeated a predetermined number of times. Without this configuration, scanning and image selection are repeated until the number of images required to construct the final image set is selected, which causes fatigue for the subject and reduces imaging efficiency, and significantly impairs imaging throughput, particularly when imaging multiple subjects in succession.

[0108] As described above, this embodiment may be configured to automatically end the alternating repetition of scanning and image selection. The condition for this automatic end is not limited to the above two examples, and may be, for example, an instruction input from the user. Alternatively, the elapsed time from the start of the alternating repetition of scanning and image selection may be measured, and the alternating repetition may be ended when a predetermined time is reached. In addition, when the repetition rate of scanning and image selection is constant, the automatic end control based on the elapsed time is equivalent to the automatic end control based on the number of repetitions described above. The image set creation unit 82 can create an image set based on a provisional image set when the alternating repetition of scanning and image selection is ended. The provisional image set is included in the image set as, for example, a series of images corresponding to the scanning range. Predetermined subject information, such as subject ID, official ID, name, age, and sex, is separately input to the slit lamp microscope 1. The image set creation unit 82 can create an image set by configuring such subject information, subject eye information (information indicating left eye / right eye, etc.), shooting date and time, shooting conditions, etc. as auxiliary information of a series of images. The image set may also include other images obtained by the slit lamp microscope 1, images obtained by other modalities, and examination data acquired by an examination device. This concludes the description of the first example of the execution mode of scanning and image selection. A specific example of the processing according to this example will be described later.

[0109] Next, a second example of the mode of scanning and image selection executed after the image group evaluation unit 81 evaluates that the quality of the image group obtained in the first scan is not good will be described. This example is configured to perform two or more scans together and then select a series of images from two or more image groups collected thereby. Then, images are selected from the two or more image groups collected in these two or more scans and the image group obtained in the first scan performed before that, and an image set including the selected series of images is created.

[0110] A specific example of such processing will be described. First, the selection unit 821 associates a group of images corresponding to each scan with a plurality of positions (described above) in the scan range. As a result, two or more images corresponding to different scans are assigned to each of the plurality of positions in the scan range.

[0111] Next, for each of the multiple positions in the scan range, the selection unit 821 selects one image from the two or more images assigned to that position. The image selection condition applied in this example may be, for example, the blink condition and eye movement condition described above. As a result, one image is assigned to each of the multiple positions in the scan range. The multiple images associated with the multiple positions in this way are adopted as a series of images included in the image set. This concludes the explanation of the second example of the execution mode of scanning and image selection.

[0112] A fourth example of the data processing unit 8 will be described. As with the second and third examples, the effectiveness of this example is particularly demonstrated when two or more scans are applied to the anterior segment of the subject's eye E. The data processing unit 8D shown in Fig. 4D includes an image set evaluation unit 83 in addition to an image group evaluation unit 81 similar to that in Fig. 4A and an image set creation unit 82 similar to that in Fig. 4B or 4C.

[0113] The image set evaluation unit 83 evaluates the quality of the image set created by the image set creation unit 82. This evaluation is to determine whether the image set has sufficient quality for effective diagnosis (image interpretation), and evaluation items and evaluation criteria are determined from this perspective. The evaluation items and evaluation criteria may be common to at least a part of the image group evaluation conditions and / or at least a part of the image selection conditions, but are not limited to them.

[0114] The image set evaluation unit 83 may be configured to perform different evaluations depending on the aspect of the image set. For example, the evaluation applied to an image set created in response to the number of images included in the provisional image set reaching a predetermined number may be different from the evaluation applied to an image set created in response to the number of alternating iterations of scanning and image selection reaching a predetermined number. Note that the same evaluation may be applied regardless of the aspect of the image set.

[0115] Examples of quality evaluation of an image set include evaluation of the quality of each image (which may be the same as the evaluation of an image group), as well as evaluation of the "arrangement order" of a series of images, evaluation of "image skips (missing images)", evaluation of "positional deviation", etc. Defects in an image set, such as rearrangement of the arrangement order, skipping of images, and positional deviation, occur due to eye movement, fixation deviation, etc.

[0116] The evaluation of the arrangement order of a series of images will be described. In some examples, the above-mentioned correspondence (one-to-one correspondence) is set between the series of images and multiple positions in the scan range. The image set evaluation unit 83 can evaluate the arrangement order by using this correspondence.

[0117] Here, the multiple positions in the scan range are ordered according to their positional relationships in real space. An example will be described. As described above, it is assumed that the scan range along the X direction is divided into (N-1) sections, and N positions B1, B2, B3, ..., BN are set in order from the scan start position to the scan end position. That is, the N positions B1 to BN are ordered according to their positional relationships in real space. It is also assumed that N images F1 to FN (a series of images) correspond to the N positions B1 to BN in a one-to-one relationship.

[0118] Under such a premise, the image set evaluation unit 83 arranges the N images F1 to FN according to the arrangement order (relative positional relationship) of the N positions B1 to BN, for example. This process is realized by setting the coordinates of the N positions B1 to BN in a certain three-dimensional coordinate system, and arranging (embedding) the N images F1 to FN according to the set N coordinates. More specifically, the image set evaluation unit 83 can extract slit light irradiation regions A1 to AN (two-dimensional cross-sectional images) from the N images F1 to FN, respectively, set the coordinates of the N positions B1 to BN in the three-dimensional coordinate system, and embed the N two-dimensional cross-sectional images A1 to AN according to the set N coordinates.

[0119] The image set evaluation unit 83 can evaluate whether the arrangement order is appropriate by analyzing the images F1 to FN (two-dimensional cross-sectional images A1 to AN) embedded in the three-dimensional coordinate system. For example, the image set evaluation unit 83 can detect regions of interest (image regions corresponding to sites of interest such as the anterior surface of the cornea, the posterior surface of the cornea, the iris, the pupil, the anterior surface of the lens, and the posterior surface of the lens) from the images F1 to FN (two-dimensional cross-sectional images A1 to AN), and perform evaluation based on the shape (connectivity, continuity, etc.) of the regions of interest in the arrangement direction (X direction in this example) of the images F1 to FN (two-dimensional cross-sectional images A1 to AN). For example, if a gap of a predetermined dimension or more exists in the region of interest, it is determined that the arrangement order is not appropriate (there is a change in the arrangement order).

[0120] In another example, the image set evaluation unit 83 constructs a cross-sectional image along the X direction from images F1 to FN (2D cross-sectional images A1 to AN) embedded in a 3D coordinate system. Furthermore, the image set evaluation unit 83 can perform evaluation based on the form (connectivity, continuity, etc.) of this cross-sectional image.

[0121] The evaluation of image skipping and positional deviation can also be performed in the same manner as the evaluation of the arrangement order.

[0122] As with the image group evaluation unit 81 and the selection unit 821, the image set evaluation unit 83 may include an artificial intelligence engine to evaluate whether the input image set has sufficient quality for effective diagnosis.

[0123] Further specific examples of evaluation of image sets are described in the second embodiment.

[0124] The control unit 7 may be configured to, when the image set evaluation unit 83 evaluates the quality of the image set to be good, perform control to cause the communication unit 9 to transmit this image set. For example, the control unit 7 prepares transmission information including this image set, and controls the communication unit 9 to transmit this transmission information to a predetermined external device.

[0125] The manner in which an image set or the like is output from the slit lamp microscope 1 is not limited to transmission. Examples of output manners other than transmission include saving in a storage device (such as a database), recording on a recording medium, and printing on a print medium.

[0126] The control unit may be configured to perform control to acquire a new image set when the image set evaluation unit 83 evaluates that the quality of the image set is not good. For example, the control unit 7 may be configured to display and / or output predetermined output information as sound. The predetermined output information has content to prompt the user to retake an image, such as a failure to take an image or a need to take an image again.

[0127] Alternatively, the control unit 7 may be configured to send commands to at least the scanning unit (illumination system 2, shooting system 3, movement mechanism 6) and the image set creation unit 82 to automatically start re-photographing (re-executing scanning and image set creation).

[0128] <Communications Division 9> The communication unit 9 performs data communication between the slit lamp microscope 1 and other devices. That is, the communication unit 9 transmits data to other devices and receives data transmitted from other devices.

[0129] The communication unit 9 may execute any data communication method. For example, the communication unit 9 includes one or more of various communication interfaces, such as a communication interface conforming to the Internet, a communication interface conforming to a dedicated line, a communication interface conforming to a LAN, a communication interface conforming to short-distance communication, etc. The data communication may be wired communication or wireless communication.

[0130] The data transmitted and received by the communication unit 9 may be encrypted. In that case, for example, the control unit 7 and / or the data processing unit 8 includes at least one of an encryption processing unit that encrypts the data transmitted by the communication unit 9 and a decryption processing unit that decrypts the data received by the communication unit 9.

[0131] <Image Group Evaluation Unit 81> Several examples of the image group evaluation unit 81 will be described with reference to Figures 5A to 5D. Note that the image group evaluation unit 81 is not limited to these examples, and any modifications (addition, substitution, omission, etc.) are possible. In addition, two or more of these examples and modifications can be at least partially combined.

[0132] A first example of the image group assessment unit 81 will be described with reference to Figures 5A and 5B. This example assesses the quality of an image group using artificial intelligence technology. The image group assessment unit 81A shown in Figure 5A includes an inference unit 811A that assesses the quality of an image group using an inference model 812A.

[0133] The inference model 812A is constructed in advance by machine learning using training data including a plurality of anterior segment images. An apparatus (inference model construction apparatus) for constructing the inference model 812A may be provided in the slit lamp microscope 1 (such as the data processing unit 8) or its peripheral device (such as a computer), or may be another computer.

[0134] 5B is an example of such an inference model construction device, and is assumed to be provided in the slit lamp microscope 1 or its peripheral devices. The model construction unit 90 includes a learning processing unit 91 and a neural network 92.

[0135] The neural network 92 typically includes a convolutional neural network (CNN). An example of the structure of a convolutional neural network is shown in FIG. 5B.

[0136] An image is input to the input layer. Behind the input layer, multiple pairs of convolution layers and pooling layers are arranged. In the example shown in Figure 5B, three pairs of convolution layers and pooling layers are provided, but the number of pairs is arbitrary.

[0137] In the convolution layer, a convolution operation is performed to grasp features (such as contours) from an image. The convolution operation is a multiplication and accumulation operation of a filter function (weighting factor, filter kernel) of the same dimension as the input image for the input image. In the convolution layer, the convolution operation is applied to each of multiple parts of the input image. More specifically, in the convolution layer, the value of each pixel of the partial image to which the filter function is applied is multiplied by the value (weight) of the filter function corresponding to that pixel to calculate the product, and the sum of the products is obtained over multiple pixels of this partial image. The product-accumulation value thus obtained is substituted for the corresponding pixel in the output image. The result of the convolution operation for the entire input image is obtained by performing the product-accumulation operation while moving the location (partial image) to which the filter function is applied. According to such a convolution operation, a large number of images in which various features are extracted using a large number of weighting factors are obtained. In other words, a large number of filtered images such as smoothed images and edge images are obtained. The large number of images generated by the convolution layer are called feature maps.

[0138] In the pooling layer, the feature map generated by the immediately preceding convolution layer is compressed (data thinning, etc.). More specifically, in the pooling layer, statistics of a predetermined neighborhood pixel of a pixel of interest in the feature map are calculated at a predetermined pixel interval, and an image with a smaller size than the input feature map is output. The statistics applied to the pooling operation are, for example, maximum value (max pooling) or average value (average pooling). The pixel interval applied to the pooling operation is called a stride.

[0139] A convolutional neural network can extract many features from an input image by performing processing using multiple pairs of convolutional layers and pooling layers.

[0140] A fully connected layer is provided after the last pair of convolutional and pooling layers. In the example shown in Figure 2, two fully connected layers are provided, but the number of fully connected layers is arbitrary. In the fully connected layer, image classification, image segmentation, regression, and other processes are performed using features compressed by a combination of convolution and pooling. After the last fully connected layer, an output layer is provided to provide the output results.

[0141] In some exemplary embodiments, the convolutional neural network may not include a fully connected layer (e.g., a fully layered convolutional network (FCN)), and may include a support vector machine, a recurrent neural network (RNN), or the like. Machine learning for the neural network 92 may be transfer learning. That is, the neural network 92 may include a neural network that has already been trained using other training data (training images) and has had its parameters adjusted. The model construction unit 90 (learning processing unit 91) may be configured to be capable of applying fine tuning to the trained neural network (92). The neural network 92 may include, for example, a known open source neural network architecture.

[0142] The learning processing unit 91 applies machine learning using training data to the neural network 92. When the neural network 92 includes a convolutional neural network, the parameters adjusted by the learning processing unit 91 include, for example, filter coefficients of the convolutional layer and connection weights and offsets of the fully connected layer.

[0143] As described above, the training data includes at least a plurality of anterior segment images. The plurality of anterior segment images are typically images acquired by a slit lamp microscope, but are not limited thereto, and may include, for example, images acquired by other ophthalmic modalities (fundus camera, OCT device, SLO, surgical microscope, etc.), images acquired by image diagnostic modalities of other medical departments (ultrasound diagnostic device, X-ray diagnostic device, X-ray CT device, magnetic resonance imaging (MRI) device, etc.), images generated by processing actual images, pseudo images, etc. In addition, the number of training data may be increased using techniques such as data expansion and data augmentation.

[0144] The training method for constructing an inference model may be any method, but may be, for example, any one of supervised learning, unsupervised learning, and reinforcement learning, or a combination of any two or more of them.

[0145] In some exemplary embodiments, supervised learning is performed using training data in which input images are labeled with a final output label. For example, a label indicating whether an image can be read or not is pre-assigned to each of a plurality of anterior eye images included in the training data. The label is generated, for example, by a doctor or another inference model. The learning processing unit 91 can build the inference model 812A by applying supervised learning using such training data to the neural network 92.

[0146] The inference model 812A of this example constructed in this way is a trained model that inputs an image obtained by scanning the anterior segment with a slit light and outputs the possibility of interpretation. Note that the possibility of interpretation, which is the output of the inference model 812A, may be any parameter indicating whether the image group to be evaluated is suitable for interpretation or not, and may be, for example, a determination of whether interpretation is possible or not, a probability that interpretation can be performed, or the validity of the result when interpretation is performed (for example, accuracy, precision), etc.

[0147] In order to prevent processing from concentrating on a particular unit of the neural network 92, the learning processing unit 91 may randomly select and disable some units and perform learning using the remaining units (dropout).

[0148] The method used to build the inference model is not limited to the examples shown here. For example, any method such as a support vector machine, a Bayesian classifier, boosting, a k-means algorithm, a kernel density estimation, a principal component analysis, an independent component analysis, a self-organizing map, a random forest, or a generative adversarial network (GAN) can be used to build the inference model.

[0149] 5A uses such an inference model 812A to evaluate the quality of a group of images collected in one scan of the anterior segment of the subject's eye E. More specifically, the inference unit 811A first inputs the group of images or each image contained therein to the inference model 812A. The inference model 812A derives information indicating the possibility of interpretation from the input group of images or images.

[0150] The inference unit 811A may use the image reading possibility information output from the inference model 812A as an inference result as it is, or may generate an inference result based on the image reading possibility information. As an example of the latter, the inference unit 811A may generate information for display from the image reading possibility information, or may execute a predetermined statistical process.

[0151] A second example of the image group evaluation unit 81 will be described with reference to Fig. 5C. In this example, an image group is converted into a 3D image and quality evaluation is performed. The image group evaluation unit 81B shown in Fig. 5C includes a 3D image construction unit 811B, a comparison unit 812B, and an evaluation processing unit 813B.

[0152] The three-dimensional image constructing section 811B constructs a three-dimensional image from a group of images acquired by one scan of the anterior segment of the subject's eye E. For example, the three-dimensional image constructing section 811B may be configured to construct stack data by embedding the group of images in a single three-dimensional coordinate system. This stack data is constructed, for example, by embedding N two-dimensional cross-sectional images A1 to AN shown in FIG. 3 in the three-dimensional coordinate system according to the arrangement order (relative positional relationship) of the N positions B1 to BN described above.

[0153] The three-dimensional image constructing unit 811B may be configured to construct volume data by applying a voxelization process to stack data constructed from the images. Furthermore, the three-dimensional image constructing unit 811B can apply a predetermined rendering to the stack data or the volume data. Examples of the rendering include volume rendering and surface rendering.

[0154] The comparison unit 812B compares the three-dimensional image (stack data, volume data, rendering image, etc.) constructed by the three-dimensional image construction unit 811B with a predetermined reference three-dimensional image. One or more arbitrary number of reference three-dimensional images are prepared.

[0155] The reference three-dimensional image may include one or more reference three-dimensional images corresponding to a normal eye. The reference three-dimensional image corresponding to a normal eye may be, for example, an image acquired by photographing a normal eye (an eye in which no disease is found (a healthy eye)). The photographing modality for acquiring this reference three-dimensional image may be any, but is typically a slit lamp microscope 1 or a similar slit lamp microscope. The reference three-dimensional image corresponding to a normal eye may also be an image acquired by photographing a model of a normal eye (such as a model eye), or an image created by computer graphics from a model of a normal eye or a clinical case.

[0156] The reference three-dimensional image may include one or more reference three-dimensional images corresponding to the diseased eye. The reference three-dimensional image corresponding to the diseased eye may be, for example, an image acquired by photographing an eye that has been definitively diagnosed with a specific disease. The imaging modality for acquiring the reference three-dimensional image may be any, but is typically a slit lamp microscope 1 or a similar slit lamp microscope. The reference three-dimensional image corresponding to the diseased eye may also be an image acquired by photographing a model of the diseased eye (such as a model eye), or an image created by computer graphics from a model or clinical case of the diseased eye.

[0157] The comparison unit 812B executes matching between the 3D image constructed by the 3D image construction unit 811B and a reference 3D image to calculate the value of a predetermined parameter. This image matching may utilize any method, such as image correlation, feature-based matching, region-based matching, machine learning (trained model), etc. The calculated parameters may be any parameters, such as correlation values, matching parameters (angle, scale, similarity, compatibility, etc.), and output parameters of trained models.

[0158] Such image matching is typically a process of quantitatively (i.e., numerically) expressing the extent to which the shape and structure of tissues and areas (such as the cornea, iris, pupil, and angle) depicted in a 3D image resembles the shape and structure of a standard normal eye and / or the shape and structure of a standard diseased eye.

[0159] The evaluation processing unit 813B evaluates the quality of the image group based on the parameter values ​​calculated by the comparison unit 812B. For example, the evaluation processing unit 813B may be configured to evaluate the quality of the image group by comparing the parameter values ​​calculated by the comparison unit 812B with a predetermined threshold value. Alternatively, the evaluation processing unit 813B may be configured to evaluate the quality of the image group by determining whether the parameter values ​​calculated by the comparison unit 812B are within a predetermined range. Note that the method of processing executed by the evaluation processing unit 813B is not limited to these, and may be any method that can be used to derive an evaluation result from the value of a certain parameter.

[0160] The quality evaluation of the image group using the 3D image construction is not limited to this example. For example, the image group evaluation unit 81 may be configured to be capable of performing one or more evaluation processes of evaluating the arrangement order of a series of images constituting the image group, evaluating skipping of images, and evaluating misalignment in the same manner as the image set evaluation unit 83.

[0161] A third example of the image group evaluation unit 81 will be described with reference to Fig. 5D. In this example, the quality of an image group is evaluated by quantitatively evaluating the image quality of the image group. The image group evaluation unit 81C shown in Fig. 5D includes an evaluation data generation unit 811C and an evaluation processing unit 812C.

[0162] The evaluation data generating unit 811C obtains image quality evaluation data from images included in an image group collected in one scan of the anterior segment of the subject's eye E. The image quality evaluation data quantitatively expresses the image quality of the image group.

[0163] Several examples of image quality evaluation processing executed by the evaluation data generating unit 811C will be described. In some exemplary embodiments, the image quality evaluation processing executed by the evaluation data generating unit 811C may be any processing, for example, processing using any known technology such as signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), root-mean-square (RMS) granularity, Wiener spectrum, modulation transfer function (MTF), quality index (QI), etc.

[0164] For example, the evaluation data generating unit 811C calculates a value of a predetermined image quality evaluation index (image quality evaluation value) as image quality evaluation data of the image group. The image quality evaluation value may be any parameter that quantitatively expresses the quality of the image, and typically, the higher the image quality evaluation value, the higher the image quality evaluation value.

[0165] As an example of a method for calculating an image quality evaluation value, a method for calculating an Image Quality value (IQ value) used in evaluating the image quality of an OCT image will be described below. First, the evaluation data generating unit 811C detects an image region (tissue image region) corresponding to the tissue (site) of the anterior eye and other image regions (background region, non-tissue image region) by applying a predetermined analysis process (e.g., segmentation) to an evaluation region set in an image to be evaluated. Next, the evaluation data generating unit 811C generates a histogram of brightness in the tissue image region and generates a histogram of brightness in the background region. Next, the evaluation data generating unit 811C calculates an image quality evaluation value (IQ value) from the degree of overlap of these two histograms. For example, the IQ value is defined in the range [0, 100] so that when both histograms completely overlap, the IQ value is 0, and when both histograms are completely separated, the IQ value is 100. This image quality evaluation calculation may include, for example, normalization of two histograms, generation of a probability distribution function, calculation of an IQ value using a predetermined calculation formula, and so on.

[0166] In this way, the evaluation data generation unit 811C may be configured to perform a process of identifying a tissue image region corresponding to the tissue of the anterior eye and a background region in the image to be evaluated, a process of creating a first histogram showing the frequency distribution of luminance in the tissue image region, a process of creating a second histogram showing the frequency distribution of luminance in the background region, and a process of calculating an image quality evaluation value (IQ value) as image quality evaluation data based on the first histogram and the second histogram.

[0167] The evaluation processing unit 812C evaluates the quality of the image group based on the image quality evaluation data generated by the evaluation data generating unit 811C. Several methods for quality evaluation based on image quality evaluation data are described below, but the quality evaluation method is not limited to these and may be any method.

[0168] A first example of quality evaluation based on image quality evaluation data will be described. When an IQ value is obtained for each image constituting an image group, the evaluation processing unit 812C compares each of the multiple IQ values ​​obtained for this image group with a predetermined threshold. If all of the multiple IQ values ​​are equal to or greater than the threshold, the evaluation processing unit 812C determines that the quality of this image group is good. On the other hand, if any of the multiple IQ values ​​is less than the threshold, the evaluation processing unit 812C determines that the quality of this image group is not good.

[0169] A second example of quality evaluation based on image quality evaluation data will be described. When an IQ value is obtained for each image constituting an image group, the evaluation processing unit 812C selects the lowest IQ value from among the multiple IQ values ​​obtained for this image group, and compares this minimum IQ value with a predetermined threshold. If the minimum IQ value is equal to or greater than the threshold, the evaluation processing unit 812C determines that the quality of this image group is good. On the other hand, if the minimum IQ value is less than the threshold, the evaluation processing unit 812C determines that the quality of this image group is not good.

[0170] A third example of quality evaluation based on image quality evaluation data will be described. When an IQ value is obtained for each image constituting an image group, the evaluation processing unit 812C applies a predetermined statistical operation to the multiple IQ values ​​obtained for this image group to calculate a statistical value. The type of this statistical value may be any, and may be, for example, an average value, a minimum value, a maximum value, a mode value, a median value, or the like. Note that the case where the statistical value is the minimum value corresponds to the above-mentioned second example. The evaluation processing unit 812C compares the calculated statistical value with a predetermined threshold value. If the statistical value is equal to or greater than the threshold value, the evaluation processing unit 812C determines that the quality of this image group is good. On the other hand, if the statistical value is less than the threshold value, the evaluation processing unit 812C determines that the quality of this image group is not good.

[0171] The processing executed by the evaluation processing unit 812C is not limited to processing based on the image quality evaluation data generated by the evaluation data generating unit 811C. For example, the evaluation processing unit 812C may be configured to be capable of executing one or more evaluation processes of evaluating the arrangement order of a series of images constituting an image group, evaluating skipping of images, and evaluating misalignment (all of the evaluation processes have been described above).

[0172] <Other elements> 1, the slit lamp microscope 1 may include a display device and an operating device. Alternatively, the display device and the operating device may be peripheral devices of the slit lamp microscope 1.

[0173] The display device displays various information under the control of the control unit 7. The display device may include a flat panel display such as a liquid crystal display (LCD).

[0174] The operation devices include devices for operating the slit lamp microscope 1 and devices for inputting information. The operation devices include, for example, buttons, switches, levers, dials, handles, knobs, mice, keyboards, trackballs, operation panels, and the like.

[0175] A device in which a display device and an operation device are integrated, such as a touch screen, may be used.

[0176] The subject and the assistant can operate the slit lamp microscope 1 by using the display device and the operation device.

[0177] <Alignment> The alignment of the slit lamp microscope 1 with respect to the subject's eye E will be described. In general, alignment is an operation for disposing the device optical system at a suitable position for photographing and measuring the subject's eye E. The alignment in this embodiment is an operation for disposing the illumination system 2 and the imaging system 3 at suitable positions for acquiring a plurality of anterior segment images (such as a series of images, a video image, an image group, or an image set) as shown in FIG.

[0178] There are various methods for aligning an ophthalmic apparatus. Some alignment methods are exemplified below, but the methods applicable to this embodiment are not limited to these.

[0179] An alignment method applicable to this embodiment is stereo alignment. Stereo alignment can be applied to an ophthalmic device capable of photographing the anterior eye from two or more different directions, and a specific method thereof is disclosed in JP 2013-248376 A by the present applicant. Stereo alignment includes, for example, the following steps: a step in which two or more anterior eye cameras photograph the anterior eye from different directions to obtain two or more photographed images; a step in which a processor analyzes these photographed images to obtain a three-dimensional position of the subject's eye; and a step in which the processor controls the movement of the optical system based on the obtained three-dimensional position. As a result, the optical system (illumination system 2 and photographing system 3 in this example) is placed at a suitable position relative to the subject's eye. In typical stereo alignment, the position of the pupil (center or center of gravity of the pupil) of the subject's eye is used as a reference.

[0180] In addition to such stereo alignment, any known alignment method can be adopted, such as a method using a Purkinje image obtained by alignment light, a method using an optical lever, a method using an alignment index, etc. In the method using a Purkinje image, or the method using an optical lever or an alignment index, the position of the corneal apex of the subject's eye is used as a reference.

[0181] Note that while typical conventional alignment methods, including the above examples, are performed with the aim of aligning the axis of the test eye with the optical axis of the optical system, in the present embodiment, alignment can be performed to position the illumination system 2 and the imaging system 3 at a position corresponding to the scan start position.

[0182] As a first example of alignment in this embodiment, after performing alignment based on the pupil or corneal vertex of the subject's eye E by applying any of the above-mentioned alignment methods, the illumination system 2 and the imaging system 3 can be moved (in the X direction) by a distance equivalent to a preset standard value of the corneal radius. Note that instead of using the standard value, a measured value of the corneal radius of the subject's eye E may be used.

[0183] As a second example, after performing alignment based on the pupil or corneal apex of the subject's eye E by applying any of the above-mentioned alignment methods, an image of the anterior segment of the subject's eye E is analyzed to measure the corneal radius, and the illumination system 2 and the imaging system 3 can be moved (in the X direction) by a distance corresponding to this measurement. The image of the anterior segment analyzed in this example is, for example, an anterior segment image obtained by the imaging system 3, or another image. The other image may be any image, such as an image obtained by an anterior segment camera, an image obtained by an anterior segment OCT, or the like.

[0184] As a third example, an image of the anterior segment obtained by an anterior segment camera or imaging system 3 for stereo alignment can be analyzed to determine a first end of the cornea, and stereo alignment can be applied to move the illumination system 2 and imaging system 3 to a position corresponding to this first end.

[0185] It is also possible to apply any of the above-mentioned alignment methods to perform alignment based on the pupil or corneal apex of the subject's eye E, and start scanning the anterior segment with the slit light from the position determined by this. Even in this case, the scan sequence can be set so as to scan the entire cornea C. For example, the scan sequence is set so as to scan to the left from the position determined by the alignment, and then scan to the right.

[0186] <Other matters> The slit lamp microscope 1 may include a fixation system that outputs light (fixation light) for fixating the subject's eye E. The fixation system typically includes at least one visible light source (fixation light source) or a display device that displays an image such as a landscape chart or a fixation target. The fixation system is disposed, for example, coaxially or non-coaxially with the illumination system 2 or the imaging system 3. The fixation system may include an internal fixation system that presents a fixation target to the subject through an optical path of the device optical system, and / or an external fixation system that presents a fixation target to the subject from outside the optical path.

[0187] The types of images that can be obtained by the slit lamp microscope 1 are not limited to a plurality of anterior eye images as shown in FIG. 3. For example, the slit lamp microscope 1 can obtain a three-dimensional image based on a plurality of anterior eye images, a rendering image based on the three-dimensional image, a transillumination image, a moving image showing the movement of a contact lens worn on the subject's eye, an image showing the gap between the contact lens and the corneal surface by applying a fluorescent agent, and the like. The rendering image will be described in another embodiment. The transillumination image is an image obtained by a transillumination method that uses the retinal reflection of illumination light to depict opacities or foreign bodies in the eye. It should be noted that fundus photography, corneal endothelial cell photography, meibomian gland photography, and the like may also be possible.

[0188] <Operation> Several examples of the operation of the slit lamp microscope 1 will now be described.

[0189] Although not shown in the drawings, at any stage, a user (subject, examiner, assistant, etc.) inputs subject information to the slit lamp microscope 1. The input subject information is stored in the control unit 7. The subject information typically includes identification information of the subject (subject ID).

[0190] Furthermore, background information can be input. The background information is any information related to the subject, and examples thereof include interview information of the subject, information written by the subject on a prescribed sheet, and information recorded in the subject's electronic medical record. Typically, the background information includes sex, age, height, weight, disease name, candidate disease name, test results (visual acuity value, eye refractive power value, intraocular pressure value, etc.), history of wearing and strength of refractive corrective devices (glasses, contact lenses, etc.), test history, treatment history, etc. These are merely examples, and the background information is not limited to these.

[0191] In preparation for imaging, the table on which the slit lamp microscope 1 is placed, the chair on which the subject sits, and the chin rest of the slit lamp microscope 1 are adjusted (all not shown). For example, the heights of the table, chair, and chin rest are adjusted. The chin rest is provided with a chin rest and a forehead rest to stably position the subject's face.

[0192] Once the preparation is complete, the subject sits on a chair, rests his / her chin on the chin rest, and places his / her forehead on the forehead rest. Before or after these actions, the user performs an instruction operation to start photographing the subject's eye. This operation may be, for example, pressing an imaging start trigger button (not shown) or inputting an instruction voice. Alternatively, the control unit 7 may detect the completion of the preparation phase and automatically transition to the photographing phase. Also, a fixation target (not shown) may be presented to the subject (subject's eye E or its fellow eye).

[0193] <First operation example> A first operation example that is executed after the above preparation will be described with reference to FIG.

[0194] (S1: Alignment) In response to the start of imaging, the slit lamp microscope 1 first aligns the illumination system 2 and the imaging system 3 with respect to the subject's eye E. Unlike general alignment for aligning the optical axis of the optical system with the corneal apex or pupil center of the subject's eye E, the alignment in step S1 is performed to place the illumination system 2 and the imaging system 3 at the starting position for the anterior segment scan performed in step S2.

[0195] The alignment mode in step S1 may be any mode, and may be, for example, any one of stereo alignment, manual or automatic alignment using Purkinje images, manual or automatic alignment using an optical lever, and manual or automatic alignment using an alignment index.

[0196] In some aspects, alignment is performed by targeting the corneal apex or the pupil center using such a conventional method. Furthermore, the control unit 7 further moves the illumination system 2 and the imaging system 3, which have been moved by the alignment targeting the corneal apex or the pupil center, to the scan start position (a position corresponding to this).

[0197] In some other aspects, alignment is performed with the scan start position as the target from the beginning. This alignment includes, for example, a process of analyzing an image of the anterior eye (for example, an image from the front or oblique direction) to specify a scan start position (for example, the first end of the cornea described above, or a position a predetermined distance away from the first end in a direction opposite to the axis of the subject's eye E), and a process of moving the illumination system 2 and the imaging system 3 to a position corresponding to the specified scan start position.

[0198] A predetermined operation may be performed before, during, and / or after the alignment is started. For example, adjustment of the image sensor 5 and focus adjustment may be performed.

[0199] (S2: Anterior segment scan) In the manner described above, the slit lamp microscope 1 scans the anterior segment of the subject's eye E by combining the irradiation of slit light by the illumination system 2, video capture by the imaging system 3, and movement of the illumination system 2 and the imaging system 3 by the movement mechanism 6. One scan (scan from the scan start position to the scan end position) can obtain, for example, a group of images (multiple anterior segment images) F1 to FN shown in FIG.

[0200] The data processing unit 8 may perform predetermined processing on the image obtained by scanning. For example, any signal processing or image processing such as noise removal, contrast adjustment, brightness adjustment, color correction, etc. may be applied.

[0201] (S3: Evaluate the quality of the images) The image group evaluation unit 81 evaluates the quality of the image group collected in the anterior segment scan in step S2.

[0202] (S4: Is the quality good?) If it is determined in step S3 that the quality of the image group is good (S4: Yes), the operation proceeds to step S5. On the other hand, if it is determined in step S3 that the quality of the image group is not good (S4: No), the operation proceeds to step S6.

[0203] (S5: Output a group of images) If it is determined in step S3 that the quality of the image group is good (S4: Yes), the control unit 7 performs control to output the image group. In this example, the control unit 7 controls the communication unit 9 to transmit the image group to another device.

[0204] Examples of devices to which the image group is sent include information processing devices and storage devices. Information processing devices are, for example, servers on a wide area line, servers on a LAN, computer terminals, etc. Storage devices are storage devices provided on a wide area line, storage devices provided on a LAN, etc.

[0205] The image group output in step S5 may include the background information described above. Alternatively, the background information may be supplementary information of the image group. In general, the data structure of the information output in step S5 is arbitrary.

[0206] Moreover, the image group transmitted in step S5 typically includes a series of images of the anterior segment of the right eye of the subject and a series of images of the anterior segment of the left eye. The series of images of the right eye and the series of images of the left eye are obtained by applying the operation of this example to the right eye and the left eye, respectively. The series of images of the right eye and the series of images of the left eye are each accompanied by the above-mentioned subject's eye information, whereby the series of images of the right eye and the series of images of the left eye are distinguished from each other.

[0207] Identification information of the subject is transmitted together with the image group. This identification information may be the subject ID input to the slit lamp microscope 1, or may be identification information generated based on the subject ID. For example, the subject ID (internal identification information) used for personal identification within the facility where the slit lamp microscope 1 is installed can be converted into external identification information to be used outside the facility. This can improve information security regarding personal information such as the image group and background information.

[0208] (S6: Prompt for reshooting) If it is determined in step S3 that the quality of the image group is not good (S4: No), the control unit 7 performs control to acquire a new image group. In this example, the control unit 7 executes control to display and / or output information to prompt the user to retake an image. The user performs an instruction operation to start retaking an image or an instruction operation not to retake an image.

[0209] When the user performs an instruction operation to start re-imaging, the control unit 7 performs control to execute the operations from step S1 (or step S2) again. Re-imaging is repeated, for example, up to a predetermined number of times.

[0210] On the other hand, when the user instructs not to retake the images, the control unit 7 may, for example, perform control to transmit the images determined to be of poor quality to another device, or may perform control to delete, save, or record the images determined to be of poor quality.

[0211] The image group transmitted from the slit lamp microscope 1 in step S5 (or S6) is sent directly or indirectly to an information processing device. A typical example of this information processing device is the image reading terminal used by a doctor (or optometrist).

[0212] A doctor can use the image interpretation terminal to interpret a series of images included in the image group (for example, a series of images F1 to FN shown in FIG. 3). It is also possible to construct a three-dimensional image based on the series of images, to display a rendering image of the three-dimensional image, and to display background information. Furthermore, it is possible to analyze any of the series of images, to analyze the three-dimensional image, to analyze the rendering image, and to analyze background information using the image interpretation terminal or another information processing device.

[0213] A doctor can use the image interpretation terminal to create a report (image interpretation report) in which information obtained from the image interpretation is recorded. The image interpretation report is provided to, for example, a facility where the slit lamp microscope 1 is installed, a medical institution designated by the subject, an information processing device used by a doctor designated by the subject, an address registered by the subject (email address, address, etc.), etc. Also, the image interpretation report may be sent to a specified database system to store and manage the image interpretation report.

[0214] Another example of an information processing device to which the image group transmitted from the slit lamp microscope 1 in step S5 (or S6) is a radiology device. The radiology device includes a radiology processor. The radiology processor operates, for example, according to a program for radiology interpretation, and analyzes a series of images included in the image group to obtain findings. Furthermore, the radiology processor creates a report based on the obtained findings.

[0215] The image interpretation processor may include an artificial intelligence engine that performs image interpretation using a trained model (inference model), which typically includes a convolutional neural network (CNN) trained using training data including a large number of images acquired by a slit lamp microscope and their image interpretation information.

[0216] When the image reading device includes an artificial intelligence engine and the slit lamp microscope 1 (data processing unit 8) includes an artificial intelligence engine, these artificial intelligence engines can be adjusted to have the same capabilities. In other words, adjustments can be made so that there is no difference (so that the difference is small) between the artificial intelligence engine of the image reading device and the artificial intelligence engine of the slit lamp microscope 1. In other words, the artificial intelligence engine provided in the image reading device may be at least partially the same as the aforementioned artificial intelligence engine provided in the slit lamp microscope 1.

[0217] For example, the same neural network model and the same parameters can be applied to both artificial intelligence engines, and updates of the model and parameters in both artificial intelligence engines can be synchronized.

[0218] Such unified adjustment of the artificial intelligence engines makes it possible to prevent the inconvenience of inconsistencies and errors occurring between the output by the artificial intelligence engine of the slit lamp microscope 1 and the output by the artificial intelligence engine of the image reading device. As described above, the quality evaluation performed before transmitting the image group from the slit lamp microscope 1 evaluates the image quality required for effective interpretation and diagnosis, so unified adjustment of the artificial intelligence engines makes it possible to appropriately evaluate the quality of the image group before transmission. In other words, the image group that can be read by the artificial intelligence engine of the image reading device can be evaluated with high accuracy as a "readable image group" before transmission from the slit lamp microscope 1, and the image group that cannot be read by the artificial intelligence engine of the image reading device can be evaluated with high accuracy as a "non-readable image group" before transmission from the slit lamp microscope 1.

[0219] In addition, in cases where such unified adjustment of artificial intelligence engines is not made or when interpretation is requested of another image reading device (including an artificial intelligence engine with a different model and / or parameters), it is possible to attach the processing conditions (representing the model, parameters, etc.) of the artificial intelligence engine of the slit lamp microscope 1 to the image set, or to attach the processing conditions of the artificial intelligence engine of the image reading device to the image set.

[0220] According to this first operation example, if an image group of good quality is obtained in the first scan, this image group can be provided for subsequent processing (image interpretation, etc.). Also, if an image group of good quality is not obtained in the first scan, the image group can be acquired again. More generally, shooting can be repeated until an image group of good quality is acquired. Therefore, whether or not an image group of good quality is obtained in the first scan, it is possible to provide the image group of good quality for subsequent processing. This concludes the explanation of the first operation example.

[0221] <Second operation example> A second operation example of the slit lamp microscope 1 will be described with reference to FIG.

[0222] In the first operation example, an anterior segment scan (photograph) is repeated until a good quality image group (a series of images collected in one scan) is obtained, whereas in this operation example, an anterior segment scan (photograph) is repeated until a good quality image set (e.g., a series of images arranged in a predetermined scan range) is obtained.

[0223] In other words, the data finally obtained in the first operation example is a series of images collected in one scan, whereas the data finally obtained in this operation example is, for example, a series of images arranged in a predetermined scan range. That is, the data finally obtained in this operation example may be a series of images obtained in one scan, as in the first operation example, or may be a series of images selected from two or more image groups obtained in two or more scans, unlike the first operation example.

[0224] Steps S11 to S13 of this operation example may be performed in the same manner as steps S1 to S3 of the first operation example. Therefore, the description of steps S11 to S13 will be omitted. In addition, unless otherwise specified, the matters described in the first operation example can be combined with this operation example.

[0225] (S14: Is the quality good?) If it is determined in step S13 that the quality of the image group is good (S14: Yes), the operation proceeds to step S15. On the other hand, if it is determined in step S13 that the quality of the image group is not good (S14: No), the operation proceeds to step S16.

[0226] (S15: Output image group) If it is determined in step S13 that the quality of the image group collected in the first scan is good (S14: Yes), the control unit 7 performs control to output the image group. As in the first operation example, the control unit 7 can transmit the image group to another device by controlling the communication unit 9.

[0227] (S16: Ask to take the photo again) If it is determined in step S13 that the quality of the image group is not good (S14: No), the control unit 7 performs control to prompt the user to retake the image. The user performs an instruction operation to start retaking the image or an instruction operation not to retake the image.

[0228] In this operation example, the case where re-imaging is performed will be described below. If re-imaging is not performed, the same operation as in the first operation example can be executed in this operation example as well.

[0229] (S17: Alignment) When the user performs an instruction operation to start re-imaging, the control unit 7 executes alignment in the same manner as in step S11.

[0230] (S18: Anterior segment scan) After the alignment in step S17 is completed, the slit lamp microscope 1 scans the anterior segment of the subject's eye E in the same manner as in step S12, thereby obtaining a new group of images.

[0231] (S19: Image selection) The selection unit 821 selects one or more images that satisfy a predetermined condition from a collection of images including the image group collected in the anterior eye scan in step S12 and the image group collected in the anterior eye scan in step S19. If "No" is determined two or more times in step S14, the selection unit 821 selects one or more images that satisfy a predetermined condition from a collection of images including the image group collected in the anterior eye scan in step S12 and two or more image groups collected in the anterior eye scan in step S19 that is performed two or more times.

[0232] In some exemplary embodiments, the operations of steps S18 and S19 may be either an operation of repeating scanning and image selection alternately, or an operation of performing image selection after all scanning has been performed, as described above. By such a coordinated operation of scanning and image selection, a series of images is selected from two or more image groups obtained by two or more scans. Note that a specific example of the alternating repetition of scanning and image selection will be described in the second embodiment described later.

[0233] (S20: Create image set) The image set creation unit 82 creates an image set including the series of images selected in step S19.

[0234] (S21: Evaluate the quality of the image set) The image set evaluation unit 83 evaluates the quality of the image set created in step S20.

[0235] (S14: Is the quality good?) If it is determined in step S21 that the quality of the image set is good (S14: Yes), the operation proceeds to step S15. On the other hand, if it is determined in step S21 that the quality of the image set is not good (S14: No), the operation proceeds to step S16, and steps S16 to S14 are executed again.

[0236] (S15: Output image set) If it is determined in step S14 that the quality of the image set is good (S14: Yes), the control unit 7 performs control to output the image set. In this example, the control unit 7 controls the communication unit 9 to transmit the image set to another device.

[0237] The destination of the image set may be the same as or different from the destination of the image group. The output image set may include the background information described above. Alternatively, the background information may be information accompanying the image set. As in the case of the image group, the image set to be transmitted may include a series of images of the anterior part of the subject's right eye and a series of images of the anterior part of the left eye, and further, subject information may be accompanied by the series of images of the right eye and the series of images of the left eye, respectively. In addition, identification information of the subject may be transmitted together with the image set.

[0238] The image set transmitted from the slit lamp microscope 1 in step S15 is sent directly or indirectly to an information processing device. A typical example of this information processing device is the image reading terminal used by a doctor (or optometrist) as described above.

[0239] A doctor can use the image reading terminal to read the series of images included in the image set, to construct a 3D image based on the series of images, to display a rendering of the 3D image, and to display background information, and to analyze any of the series of images, to analyze the 3D image, to analyze the rendering, and to analyze the background information.

[0240] A doctor can use the image interpretation terminal to create a report (image interpretation report) in which information obtained by interpretation of an image set is recorded. The image interpretation report is provided, for example, to a facility where the slit lamp microscope 1 is installed, a medical institution designated by the subject, an information processing device used by a doctor designated by the subject, an address registered by the subject (email address, address, etc.), etc. Also, the image interpretation report may be sent to a specified database system to store and manage the image interpretation report.

[0241] Another example of an information processing device to which the image set transmitted from the slit lamp microscope 1 in step S15 is a radiology device. The radiology device includes a radiology processor. The radiology processor operates, for example, according to a program for radiology interpretation, and analyzes a series of images included in the image set to obtain findings. Furthermore, the radiology processor creates a report based on the obtained findings.

[0242] The image interpretation processor may include an artificial intelligence engine that performs image interpretation using a trained model (inference model), which typically includes a convolutional neural network (CNN) trained using training data including a large number of images acquired by a slit lamp microscope and their image interpretation information.

[0243] When the image reading device includes an artificial intelligence engine and the slit lamp microscope 1 (data processing unit 8) includes an artificial intelligence engine, these artificial intelligence engines can be adjusted to have the same capabilities. That is, it is possible to adjust so that there is no difference (so that the difference is small) between the artificial intelligence engine of the image reading device and the artificial intelligence engine of the slit lamp microscope 1. In other words, the artificial intelligence engine provided in the image reading device may be at least partially the same as the artificial intelligence engine provided in the slit lamp microscope 1. For example, the same neural network model and the same parameters can be applied to both artificial intelligence engines. In addition, the updates of the models and parameters in both artificial intelligence engines can be synchronized. That is, it is possible to perform unified adjustment of the artificial intelligence engines between the slit lamp microscope 1 and the image reading device. When such unified adjustment of the artificial intelligence engines is not performed or when image reading is requested to another image reading device (including an artificial intelligence engine with a different model and / or parameter), it is possible to attach the processing conditions (representing the model, parameters, etc.) of the artificial intelligence engine of the slit lamp microscope 1 to the image set, or to attach the processing conditions of the artificial intelligence engine of the image reading device to the image set.

[0244] According to such a second operation example, if an image group of good quality is obtained in the first scan, this image group can be provided for subsequent processing (image interpretation, etc.). Also, if an image group of good quality is not obtained in the first scan, a new image group can be obtained and a series of images can be selected from two or more image groups obtained up to that point to create an image set of good quality. More generally, imaging can be repeated until an image set of good quality is obtained. Therefore, if an image group of good quality is obtained in the first scan, this image group can be provided for subsequent processing, and if an image group of good quality is not obtained in the first scan, it is possible to create an image set of good quality from two or more image groups obtained by two or more scans and provide it for subsequent processing. This concludes the explanation of the second operation example.

[0245] The operations that can be performed by the slit lamp microscope 1 of this embodiment are not limited to the above two examples. For example, the slit lamp microscope 1 can perform operations based on at least one or more of the matters described in this embodiment, modifications thereof, and any known technology.

[0246] <Effects etc.> Some features, some actions, and some effects of the slit lamp microscope 1 of this embodiment will be described.

[0247] The slit lamp microscope 1 includes a scanning unit (illumination system 2, imaging system 3, and movement mechanism 6), an image group evaluation unit 81 (first evaluation unit), and a control unit 7. The scanning unit collects an image group by scanning the anterior segment of the subject's eye E with slit light. The image group evaluation unit 81 evaluates the quality of the image group collected by the scanning unit. The control unit 7 selectively executes two or more predetermined controls depending on the quality evaluation result obtained by the first evaluation unit.

[0248] As described above, delicate and complicated operations are required to obtain good images using conventional slit lamp microscopes, while such difficult operations must be performed from a remote location with the remote operation technology disclosed in Patent Documents 3 and 4, making it extremely difficult to obtain images of a quality that can withstand diagnosis (image interpretation). In addition, although slit lamp microscopes are effective in screening and the like, considering the difficulty of remote operation, it can even be said that using slit lamp microscopes in remote medical care is practically impossible with conventional technology.

[0249] In contrast, the slit lamp microscope 1 of this embodiment is configured to first scan the anterior segment of the subject's eye E to collect a group of images, so there is no need for remote operation by a doctor as in the conventional case.

[0250] Furthermore, the quality of the images collected by scanning the anterior segment can be assessed by the slit lamp microscope itself, allowing the quality of the images to be verified before being provided to a physician or image reader.

[0251] Furthermore, since the control contents can be switched depending on the evaluation results of the image group quality by the image group evaluation unit 81, when an image group of good quality is obtained, suitable processing can be executed accordingly, and when an image group of good quality is not obtained, suitable processing can be executed accordingly.

[0252] In this way, according to the slit lamp microscope 1 of this embodiment, a group of high-quality images obtained by scanning the anterior segment can be provided to doctors and image reading devices, making it possible to widely provide high-quality slit lamp microscope examinations.

[0253] Furthermore, while conventionally, doctors perform examinations while operating the device from a remote location, in this embodiment, the doctor only needs to interpret the images acquired in advance. In other words, this embodiment frees the doctor from the time and effort required for image capture, allowing him or her to concentrate on image interpretation. Therefore, this embodiment contributes to expanding the scope of high-quality slit lamp microscopy examinations.

[0254] The slit lamp microscope 1 of this embodiment may further include a communication unit 9 for performing data communication. Moreover, the control unit 7 may be configured to control the communication unit 9 so as to transmit the image group collected by the scanning unit to a predetermined external device (e.g., an image reading terminal, an image reading device) when the image group evaluation unit 81 evaluates that the quality of the image group collected by the scanning unit is good.

[0255] According to this configuration, once a group of images of good quality has been acquired, the group of images can be provided to, for example, a doctor in a remote location or to an image reading device in a remote location.

[0256] In this embodiment, the control unit 7 may be configured to execute control for causing the scanning unit to execute a new scan of the anterior segment of the subject's eye E when the image group evaluation unit 81 evaluates that the quality of the image group is not good. This is one example of selectively executable control content. Note that the control for causing the scanning unit to execute a new scan may be, for example, control for prompting the user to retake an image, or control for automatically taking a retake (control for automatically shifting to retake an image).

[0257] According to this configuration, if a group of images of good quality cannot be obtained in one scan, a smooth transition to re-imaging (rescanning) can be made.

[0258] The slit lamp microscope 1 of this embodiment may further include an image set creation unit 82. The image set creation unit 82 is configured to create an image set by selecting a series of images corresponding to a predetermined scan range from two or more image groups including an image group already acquired by scanning the anterior segment and a new image group collected in a new scan.

[0259] According to this configuration, when two or more scans are applied to the anterior segment, an image set can be created by a series of images selected from two or more image groups obtained thereby, and the possibility of obtaining a good quality image set is improved. For example, even if a good image is not obtained due to blinking or eye movement during the execution of a certain scan, it is possible to supplement it with images obtained by other scans.

[0260] The image set creation unit 82 may include a selection unit 821 that selects images that satisfy a predetermined condition from two or more image groups acquired by two or more scans.

[0261] According to this configuration, good images that satisfy predetermined conditions can be selected for creating an image set.

[0262] For example, the selection unit 821 may be configured to select, from two or more image groups, an image including a reflected image of slit light irradiated onto the anterior segment of the subject's eye E. This makes it possible to create an image set by excluding images captured when a blink occurs (images in which the anterior segment is not depicted).

[0263] The selection unit 821 may be configured to select an image from a set of two or more images by comparison of adjacent images.

[0264] According to this configuration, it is possible to create an image set by excluding images captured when eye movement occurs (images in which positional inconsistency occurs).

[0265] In this embodiment, a plurality of positions (B1 to BN) may be set for the scan range. Furthermore, the selection unit 821 may be configured to select images so as to assign one or more images to each of the plurality of positions (B1 to BN).

[0266] According to this configuration, a plurality of images (F1 to FN) corresponding to a plurality of positions (B1 to BN) in the scan range, respectively, are obtained.

[0267] Typically, one image is assigned to each of the multiple positions (B1 to BN), but two or more images can be assigned to any of the multiple positions (B1 to BN). For example, noise can be reduced by averaging two or more images assigned to a certain position Bn. A doctor can compare two or more images assigned to a certain position Bn and make a selection. Two or more images taken under different conditions (wavelength, presence or absence of administration of fluorescent agent, etc.) can be assigned to a certain position Bn.

[0268] In this embodiment, the control unit 7 may be configured to control the scanning unit and the selection unit 821 to alternately repeat application of one (or more) scans to the anterior segment of the subject's eye E and selection of images from one (or more) image groups collected in the one (or more) scans. That is, the slit lamp microscope 1 may be configured to perform alternating repetition of scanning and image selection to obtain a series of images included in an image set.

[0269] In the alternating repetition of scanning and image selection, the scanning period and the image selection period may overlap or may be separated from each other, i.e., the scanning and image selection may be performed partially in parallel or may be performed exclusively in time.

[0270] The selection unit 821 may be configured to select images from one or more images acquired in one or more scans that have already been performed in an alternating iteration of scanning and image selection to create a provisional image set, i.e., the selection unit 821 may be configured to create a provisional image set from one or more images acquired in one or more scans that have been performed up to the current time.

[0271] Furthermore, in the alternating repetition of scanning and image selection, when a new scan is applied to the anterior segment of the subject's eye E, the selection unit 821 may be configured to execute a process of selecting an image from a new image group collected by the new scan, and a process of creating a new provisional image set by adding an image selected from the new image group to a provisional image set based on one or more scans performed before the new scan. That is, the selection unit 821 may be configured to sequentially select images from the image group collected by the scan and add them to the provisional image set every time a scan is applied to the anterior segment. More generally, when a new scan is applied to the anterior segment after one or more scans, the selection unit 821 may be configured to select an image from the new image group collected by the new scan to update the provisional image set. Here, "update" is not limited to "addition" and may be, for example, "replacement", "combination (e.g., averaging)", etc.

[0272] According to the above-described configuration of alternately repeating scanning and image selection, a series of images corresponding to the scan range can be obtained by sequentially selecting images from the image group obtained in each scan, and an image set can be created. Furthermore, it is possible to sequentially select good images from the image group obtained in each scan.

[0273] The control unit 7 may be configured to control the scanning and selecting unit 821 to terminate the alternating repetition of applying the scan and selecting the images when the number of images included in the tentative image set reaches a predetermined number. Furthermore, the image set creating unit 82 may be configured to create an image set based on the tentative image set when the alternating repetition of applying the scan and selecting the images is terminated.

[0274] With this configuration, the alternating repetition of scanning and image selection can be automatically terminated when a sufficient number of images are obtained to create an image set, making it possible to efficiently perform photography and create an image set.

[0275] The control unit 7 may be configured to control the scanning and selecting unit 821 to terminate the alternating repetition of applying the scan and selecting the image when the number of alternating repetitions of applying the scan and selecting the image reaches a predetermined number. Furthermore, the image set creating unit 82 may be configured to create an image set based on the tentative image set when the alternating repetition of applying the scan and selecting the image is terminated.

[0276] According to this configuration, it is possible to prevent a situation in which photographing for creating an image set takes a long time.

[0277] Instead of the above-described alternating repetition of scanning and image selection, the slit lamp microscope 1 may be configured to perform image selection after multiple scans are performed together. Specifically, when the quality of the image group collected by the scanning unit is evaluated to be poor, the control unit 7 may be configured to control the scanning unit to apply two or more scans to the anterior segment, and then control the selection unit 821 to select images from three or more image groups including the two or more image groups collected in the two or more scans and the image group collected in the scan performed before that. Furthermore, the image set creation unit 82 may be configured to create an image set including a series of images selected by the selection unit 821.

[0278] The slit lamp microscope 1 of this embodiment may further include an image set evaluation unit 83 that evaluates the quality of the image set created by the image set creation unit 82. For example, the image set evaluation unit 83 can evaluate a series of images included in the image set with respect to any of the arrangement order, missing images, and positional deviation. This evaluation is performed by analyzing a series of images in the image set. For example, the quality of the series of images is evaluated based on landmarks (image regions corresponding to tissues such as the cornea, iris, and pupil) in a series of frames.

[0279] By performing such image set evaluation, it is possible to prepare an image set of good quality that enables effective interpretation by a doctor or an image reading device.

[0280] The slit lamp microscope 1 of this embodiment may further include an output unit. Furthermore, the control unit 7 may be configured to perform control to cause the output unit to output the image set when the image set evaluation unit 83 evaluates that the quality of the image set is good. Here, the output unit may be, for example, any one of a communication unit 9 for transmitting the image set to an external device, a storage device (such as a database) for saving the image set, a recording device (such as a data writer or drive device) for writing the image set to a recording medium, and a printer for recording information included in the image set on a print medium.

[0281] This configuration makes it possible to output a good quality image set for presentation to a doctor or image reading device.

[0282] In this embodiment, the control unit 7 may be configured to perform control for acquiring a new image set when the image set evaluation unit 83 evaluates that the quality of the image set is not good. This control may be, for example, control for prompting the user to retake images, or control for automatically starting retake images (re-execution of scanning and image set creation).

[0283] The slit lamp microscope 1 of this embodiment is configured as follows to realize scanning of the anterior segment with slit light. That is, the scanning unit includes an illumination system 2, an imaging system 3, and a movement mechanism 6. The illumination system 2 irradiates the anterior segment of the subject's eye E with slit light. The imaging system 3 images the anterior segment from a direction different from that of the illumination system 2. The movement mechanism 6 moves the illumination system 2 and the imaging system 3. The imaging system 3 repeatedly captures images in parallel with the movement of the illumination system 2 and the imaging system 3 by the movement mechanism 6. This repeated capture is, for example, video capture at a predetermined capture rate.

[0284] In this embodiment, the moving mechanism 6 moves the illumination system 2 and the imaging system 3 in the X direction when scanning the anterior segment with slit light. The moving mechanism 6 may also be capable of moving the illumination system 2 and the imaging system 3 three-dimensionally during alignment.

[0285] Furthermore, the slit lamp microscope 1 of this embodiment may have a function as a Scheimpflug camera in order to capture an image of the range from the anterior surface of the cornea to the posterior surface of the crystalline lens at once. For this purpose, the imaging system 3 may include an optical system 4 that guides light from the anterior segment irradiated with slit light, and an image sensor 5 that receives the light guided by the optical system 4 on an imaging surface. Furthermore, the slit lamp microscope 1 may be configured such that the object surface along the optical axis of the illumination system 2, the optical system 4, and the image sensor 5 (imaging surface) satisfy the Scheimpflug condition.

[0286] In this embodiment, the image group evaluation section 81 may have any configuration and operate in any manner. The image group evaluation section 81 may include any one or more of the following preferred examples (image group evaluation sections 81A, 81B, and 81C).

[0287] The image group evaluation unit 81A is configured to evaluate the quality of the image group collected by the scanning unit using an inference model 812A constructed by machine learning using training data including multiple anterior eye images.

[0288] Here, each of the multiple anterior eye images included in the training data may be labeled with a label indicating whether it is readable or not. Furthermore, the inference model 812A may be constructed by supervised learning using the training data, so that the image obtained by scanning the anterior eye with a slit light is used as an input and the readability is used as an output.

[0289] When a group of images (or a set of images) acquired by the slit lamp microscope 1 is provided to an image reading device equipped with an artificial intelligence engine that performs image reading using an inference model, the inference model 812A provided in the slit lamp microscope 1 may be the same as or similar to the inference model provided in the image reading device.

[0290] The image group evaluation unit 81B includes a three-dimensional image construction unit 811B, a comparison unit 812B, and an evaluation processing unit 813B. The three-dimensional image construction unit 811B is configured to construct a three-dimensional image from the image group collected by the scanning unit. The comparison unit 812B is configured to compare the three-dimensional image constructed by the three-dimensional image construction unit 811B with one or more predetermined reference three-dimensional images. The evaluation processing unit 813B is configured to evaluate the quality of the image group based on the comparison result obtained by the comparison unit 812B.

[0291] Here, the one or more predetermined reference three-dimensional images referred to by the comparison unit 812B may include one or both of one or more reference three-dimensional images corresponding to a normal eye and one or more reference three-dimensional images corresponding to an affected eye. In other words, the one or more predetermined reference three-dimensional images referred to by the comparison unit 812B may include only one or more reference three-dimensional images corresponding to a normal eye, may include only one or more reference three-dimensional images corresponding to an affected eye, or may include both one or more reference three-dimensional images corresponding to a normal eye and one or more reference three-dimensional images corresponding to an affected eye.

[0292] Moreover, the comparison unit 812B may be configured to calculate a value of a predetermined parameter by executing matching between the 3D image constructed by the 3D image construction unit 811B and each of one or more predetermined reference 3D images. Furthermore, the evaluation processing unit 813B may be configured to evaluate the quality of the image group based on the value calculated by the comparison unit 812B.

[0293] The image group evaluation unit 81C includes an evaluation data generation unit 811C and an evaluation processing unit 812C. The evaluation data generation unit 811C is configured to obtain image quality evaluation data from images included in the image group collected by the scanning unit. The evaluation processing unit 812C is configured to evaluate the quality of the image group based on the image quality evaluation data generated by the evaluation data generation unit 811C.

[0294] Here, the evaluation data generation unit 811C may be configured to execute a process of identifying a tissue image area corresponding to the tissue of the anterior eye and a background area in an image included in the image group collected by the scanning unit, a process of creating a first histogram showing the frequency distribution of luminance in the tissue image area, a process of creating a second histogram showing the frequency distribution of luminance in the background area, and a process of calculating an image quality evaluation value as image quality evaluation data based on the first histogram and the second histogram.

[0295] <Second aspect> In this embodiment, the configuration of an optical system applicable to the slit lamp microscope 1 of the first embodiment and its applications will be described. An example of the configuration of the optical system of this embodiment is shown in FIG. 8. In addition to the group of elements shown in FIG. 8, elements shown in other embodiments may be provided. For example, the control unit 7, data processing unit 8, communication unit 9, etc. of the first embodiment may be provided. Unless otherwise stated, the matters related to the first embodiment are applicable.

[0296] The illumination system 20 shown in FIG. 8 is an example of the illumination system 2 in the first embodiment, and the left imaging system 30L and the right imaging system 30R are examples of the imaging system 3. Hereinafter, the illumination system 20 may be described as the illumination system 2, and the left imaging system 30L and / or the right imaging system 30R may be described as the imaging system 3. In some embodiments, only one of the left imaging system 30L and the right imaging system 30R may be provided and configured as an example of the imaging system 3. Reference numeral 20a indicates the optical axis (illumination optical axis) of the illumination system 20, reference numeral 30La indicates the optical axis (left imaging optical axis) of the left imaging system 30L, and reference numeral 30Ra indicates the optical axis (right imaging optical axis) of the right imaging system 30R. The left imaging optical axis 30La and the right imaging optical axis 30Ra are arranged in different directions. The angle between the illumination optical axis 20a and the left imaging optical axis 30La is indicated by θL, and the angle between the illumination optical axis 20a and the right imaging optical axis 30Ra is indicated by θR. The angles θL and θR may be equal to or different from each other. The illumination optical axis 20a, the left photographing optical axis 30La, and the right photographing optical axis 30Ra intersect at a single point. As in FIG. 1, the Z coordinate of this intersection is indicated as Z0.

[0297] The moving mechanism 6 can move the illumination system 20, the left imaging system 30L, and the right imaging system 30R in a direction indicated by an arrow 49 (X direction). Typically, the illumination system 20, the left imaging system 30L, and the right imaging system 30R are placed on a stage that is movable at least in the X direction, and the moving mechanism 6 moves this movable stage in accordance with a control signal from the control unit 7.

[0298] The illumination system 20 irradiates slit light onto the anterior segment of the subject's eye E. Similar to the illumination system of a conventional slit lamp microscope, the illumination system 20 includes, in order from the side farthest from the subject's eye E, an illumination light source 21, a positive lens 22, a slit forming unit 23, and objective lens groups 24 and 25.

[0299] Illumination light (typically visible light) output from the illumination light source 21 is refracted by the positive lens 22 and projected onto the slit forming unit 23. A part of the projected illumination light passes through the slit formed by the slit forming unit 23 to become slit light. The generated slit light is refracted by the objective lens groups 24 and 25, and then reflected by the beam splitter 47 to be irradiated onto the anterior segment of the subject's eye E.

[0300] The left imaging system 30L includes a reflector 31L, an imaging lens 32L, and an image sensor 33L. The reflector 31L and the imaging lens 32L guide light from the anterior segment irradiated with slit light by the illumination system 20 (light traveling in the direction of the left imaging system 30L) to the image sensor 33L.

[0301] The light traveling from the anterior eye to the left imaging system 30L is the light from the anterior eye irradiated with the slit light, and travels in a direction away from the illumination optical axis 20a. The reflector 31L reflects the light in a direction approaching the illumination optical axis 20a. The imaging lens 32L refracts the light reflected by the reflector 31L to form an image on the imaging surface 34L of the imaging element 33L. The imaging element 33L receives the light on the imaging surface 34L.

[0302] As in the first embodiment, the left imaging system 30L repeatedly captures images in parallel with the movement of the illumination system 20, the left imaging system 30L, and the right imaging system 30R by the movement mechanism 6. This allows a plurality of anterior eye images (image group) to be obtained.

[0303] As in the first embodiment, the object plane along the illumination optical axis 20a, the optical system including the reflector 31L and the imaging lens 32L, and the imaging plane 34L satisfy the Scheimpflug condition. More specifically, when the deflection of the optical path of the imaging system 30L by the reflector 31L is taken into consideration, the YZ plane (including the object plane) passing through the illumination optical axis 20a, the principal plane of the imaging lens 32L, and the imaging plane 34L intersect on the same straight line. This allows the left imaging system 30L to focus on all positions in the object plane (for example, in the range from the anterior surface of the cornea to the posterior surface of the crystalline lens) and perform imaging.

[0304] The right photographing system 30R includes a reflector 31R, an imaging lens 32R, and an image sensor 33R. As with the left photographing system 30L, the right photographing system 30R guides light from the anterior eye, which is irradiated with slit light by the illumination system 20, to an imaging surface 34R of the image sensor 33R by the reflector 31R and the imaging lens 32R. Furthermore, as with the left photographing system 30L, the right photographing system 30R obtains a plurality of anterior eye images (image group) by repeatedly photographing in parallel with the movement of the illumination system 20, the left photographing system 30L, and the right photographing system 30R by the moving mechanism 6. As with the left photographing system 30L, the object surface along the illumination optical axis 20a, the optical system including the reflector 31R and the imaging lens 32R, and the imaging surface 34R satisfy the Scheimpflug condition.

[0305] The control unit 7 can synchronize the repeated photographing by the left photographing system 30L and the repeated photographing by the right photographing system 30R. This allows a correspondence relationship to be obtained between a plurality of anterior eye images obtained by the left photographing system 30L and a plurality of anterior eye images obtained by the right photographing system 30R. This correspondence relationship is a temporal correspondence relationship, and more specifically, images acquired substantially simultaneously are paired with each other.

[0306] Alternatively, the control unit 7 or the data processing unit 8 can execute a process for determining a correspondence relationship between a plurality of anterior-segment images obtained by the left photographing system 30L and a plurality of anterior-segment images obtained by the right photographing system 30R. For example, the control unit 7 or the data processing unit 8 can pair the anterior-segment images sequentially input from the left photographing system 30L and the anterior-segment images sequentially input from the right photographing system 30R based on their input timing.

[0307] This embodiment further includes a video capture unit 40. The video capture unit 40 captures video of the anterior segment of the subject's eye E from a fixed position in parallel with the capture by the left and right imaging systems 30L and 30R. "Video capture from a fixed position" means that the video capture unit 40 does not move when the illumination system 20, the left and right imaging systems 30L and 30R are moved to scan the anterior segment. The video capture unit 40 can also capture still images.

[0308] In this embodiment, the video capture unit 40 is disposed coaxially with the illumination system 20, but the arrangement is not limited thereto. For example, the video capture unit may be disposed non-coaxially with the illumination system 20. Also, an optical system may be provided that illuminates the anterior eye segment with illumination light in a band to which the video capture unit 40 has sensitivity.

[0309] The light transmitted through the beam splitter 47 is reflected by a reflector 48 and enters the video shooting unit 40. The light entering the video shooting unit 40 is refracted by an objective lens 41, and then an image is formed on the imaging surface of an image sensor 43 by an imaging lens 42. The image sensor 43 is an area sensor. The image sensor 43 has sensitivity to, for example, either one or both of the visible light and infrared light bands.

[0310] When the moving image capturing unit 40 is provided, it is possible to monitor and track the movement of the subject's eye E. In addition, it is also possible to perform alignment using the moving image capturing unit 40.

[0311] Depending on the output wavelength of the illumination system 20 and the detection wavelength of the moving image capturing unit 40, the beam splitter 47 is, for example, a dichroic mirror or a half mirror.

[0312] Some features, some actions, and some effects of the slit lamp microscope of this embodiment will be described.

[0313] This embodiment provides a left photographing system 30L and a right photographing system 30R as an example of the photographing system 3 of the first embodiment. The left photographing system 30L and the right photographing system 30R are examples of the first photographing system and the second photographing system described above. The left photographing system 30L includes a reflector 31L and an imaging lens 32L (first optical system) that guide light from the anterior eye segment irradiated with slit light, and an imaging element 33L (first imaging element) that receives the guided light on an imaging surface 34L (first imaging surface). Similarly, the right photographing system 30R includes a reflector 31R and an imaging lens 32R (second optical system) that guide light from the anterior eye segment irradiated with slit light, and an imaging element 33R (second imaging element) that receives the guided light on an imaging surface 34R (second imaging surface).

[0314] The optical axis (left photographing optical axis 30La) of the left photographing system 30L and the optical axis (right photographing optical axis 30Ra) of the right photographing system 30R are arranged in different directions. Furthermore, the object surface along the optical axis (illumination optical axis 20a) of the illumination system 20, the reflector 31L, the imaging lens 32L, and the imaging surface 34L satisfy the Scheimpflug condition. Similarly, the object surface, the reflector 31L, the imaging lens 32L, and the imaging surface 34L satisfy the Scheimpflug condition.

[0315] The left imaging system 30L acquires a first group of images by repeatedly capturing images in parallel with the movement of the illumination system 20, the left imaging system 30L, and the right imaging system 30R by the movement mechanism 6. Similarly, the right imaging system 30R acquires a second group of images by repeatedly capturing images in parallel with the movement of the illumination system 20, the left imaging system 30L, and the right imaging system 30R by the movement mechanism 6.

[0316] The control unit 7 can control the illumination system 20, the left imaging system 30L, the right imaging system 30R, and the moving mechanism 6, for example, so that the left imaging system 30L and the right imaging system 30R apply one scan to the anterior part of the subject's eye E in parallel. The data processing unit 8 can select a series of images corresponding to the scan range from the pair of left and right image groups collected in this scan, and create one image group. The image group evaluation unit 81 can evaluate the quality of the created image group. If the quality of this image group is evaluated to be good, the control unit 7 can control the communication unit 9 to transmit this image group to a predetermined external device.

[0317] If the quality of this image group is evaluated to be poor, the control unit 7 can control the illumination system 20, the left imaging system 30L, the right imaging system 30R, and the moving mechanism 6 to apply a new scan (one or more scans) to the anterior segment of the subject's eye E. By this additional scan, the left imaging system 30L collects one or more first image groups corresponding to one or more scans, and the right imaging system 30R collects one or more second image groups corresponding to one or more scans. The image set creation unit 82 can, for example, select a first series of images corresponding to the scan range from two or more first image groups obtained by the first scan and the additional scan to create a first image set, and select a second series of images corresponding to the scan range from two or more second image groups obtained by the first scan and the additional scan to create a second image set. Alternatively, the image set creation unit 82 may be configured to select a series of images corresponding to the scan range from two or more first image groups and two or more second image groups obtained by the first scan and the additional scan to create an image set. Any aspect (configuration, control, processing, action, function, etc.) in the first aspect described above can be combined with such a second aspect.

[0318] According to the second aspect, the anterior eye part irradiated with the slit light can be captured in a moving image from different directions. Even if an image acquired by one imaging system includes an artifact, an image acquired by the other imaging system substantially simultaneously with the image may not include the artifact. In addition, even if both of a pair of images acquired by both imaging systems substantially simultaneously include an artifact, and an artifact in one image overlaps a region of interest (e.g., a region irradiated with the slit light), an artifact in the other image may not overlap the region of interest. Therefore, the possibility of acquiring a suitable image is increased. Therefore, it is possible to further improve the probability of acquiring a group of images and / or a set of images with good quality.

[0319] In addition to the first and second imaging systems, the imaging system 3 may include a third, ..., Kth imaging system (K is an integer equal to or greater than 3) having a similar configuration. This complicates the structure of the optical system, but it is possible to further improve the probability of obtaining a group of images and / or a set of images of good quality. Those who wish to implement this embodiment can design a slit lamp microscope taking into account these trade-offs (complexity of the optical system and probability of obtaining a high-quality set of images).

[0320] The left imaging system 30L in this embodiment includes a reflector 31L and an imaging lens 32L. The reflector 31L reflects light from the anterior segment irradiated with the slit light, traveling in a direction away from the illumination optical axis 20a, in a direction toward the illumination optical axis 20a. Furthermore, the imaging lens 32L forms an image of the light reflected by the reflector 31L on the imaging surface 34L. Here, the imaging lens 32L includes one or more lenses.

[0321] Similarly, the right imaging system 30R includes a reflector 31R and an imaging lens 32R. The reflector 31R reflects light from the anterior eye segment irradiated with the slit light, traveling in a direction away from the illumination optical axis 20a, toward the illumination optical axis 20a. Furthermore, the imaging lens 32R forms an image of the light reflected by the reflector 31R on an imaging surface 34R. Here, the imaging lens 32R includes one or more lenses.

[0322] According to this configuration, it is possible to miniaturize the device. That is, the image acquired by the image pickup element 33L (33R) is output through a cable extending from the surface opposite to the image pickup surface 34L (34R), but according to this configuration, the cable can be arranged from the back surface of the image pickup element 33L (33R) located relatively close to the illumination optical axis 20a toward the opposite direction to the eye E. Therefore, the cable can be preferably routed, and the device can be miniaturized.

[0323] Moreover, according to this configuration, since the angles θL and θR can be set large, when an image acquired by one imaging system includes an artifact, the possibility that the image acquired by the other imaging system does not include the artifact can be increased. Furthermore, when an artifact is included in both of a pair of images acquired by both imaging systems substantially simultaneously, and an artifact in one image overlaps with a region of interest (e.g., a region irradiated with slit light), the possibility that the artifact in the other image overlaps with the region of interest can be reduced.

[0324] This embodiment includes a video shooting unit 40. The left and right shooting systems 30L and 30R repeatedly shoot images of the anterior segment in parallel with the movement of the illumination system 20, the left and right shooting systems 30L and 30R by the moving mechanism 6. In parallel with this repeated shooting, the video shooting unit 40 shoots video images of the anterior segment from a fixed position.

[0325] With this configuration, by capturing video from a fixed position (e.g., the front) in parallel with scanning the anterior segment with slit light, it is possible to grasp the condition of the test eye E during scanning and to perform control according to the condition of the test eye E.

[0326] For example, when the slit lamp microscope of this embodiment includes the image group evaluation unit 81 (image set evaluation unit 83) of the first embodiment, the slit lamp microscope of this embodiment can evaluate whether the image group (image set) acquired by the left imaging system 30L and / or the right imaging system 30R is of a quality suitable for interpretation. The image group evaluation unit 81 (image set evaluation unit 83) of this embodiment can perform the following operations in combination with the video shooting unit 40. Note that similar operations can also be performed using two or more anterior eye cameras for stereo alignment or similar shooting means.

[0327] The video capture unit 40 captures video of the anterior segment from a fixed position in parallel with the application of a scan to the anterior segment of the subject's eye E. This video capture is performed, for example, under the control of the control unit 7. That is, the control unit 7 can control the illumination system 2 (illumination system 20), the imaging system 3 (left imaging system 30L and / or right imaging system 30R), the moving mechanism 6, and the video capture unit 40 in a coordinated manner in order to scan the anterior segment of the subject's eye E.

[0328] In this linked control, the control unit 7 can synchronize the shooting rate of the imaging system 3 with the shooting rate of the video shooting unit 40. For example, the shooting rate of the imaging system 3 and the shooting rate of the video shooting unit 40 are set to be equal, and the shooting timing of the imaging system 3 and the shooting timing of the video shooting unit 40 are matched. This allows the frame group acquired by the imaging system 3 during scanning to correspond in time with the frame group acquired by the video shooting unit 40.

[0329] Even if one or both of the shooting rate and shooting timing are different, it is possible to temporally correspond the frame group acquired by the shooting system 3 during scanning with the frame group acquired by the video shooting unit 40, for example, by allowing a time difference within a predetermined range.

[0330] A pair of frames associated in time (a pair of a frame obtained by the imaging system 3 and a frame obtained by the video imaging unit 40) can be considered to have been acquired substantially simultaneously. Therefore, when considering a pair of frames associated in time, positional deviation due to eye movement can be ignored.

[0331] Under such a premise, the image group evaluation unit 81 (image set evaluation unit 83) may be configured to evaluate the quality of a series of images included in the image group (image set) based on the moving images (frame group) acquired by the video shooting unit 40.

[0332] At this time, the image group evaluation unit 81 (image set evaluation unit 83) may be configured to evaluate the quality of the image group (image set) based on a correspondence relationship between a series of images included in the image group (image set) and a series of frames included in the moving image captured by the video shooting unit 40. That is, the image group evaluation unit 81 (image set evaluation unit 83) may be configured to evaluate the quality of the image group (image set) based on a temporal correspondence relationship between a frame group acquired by the imaging system 3 and a frame group acquired by the video shooting unit 40. Furthermore, the image group evaluation unit 81 (image set evaluation unit 83) may be configured to evaluate the quality of the image group (image set) based on landmarks in the series of frames and the correspondence relationship.

[0333] A specific example will be described below. In this example, it is assumed that, for a group of frames F1 to FN (described above) acquired by the imaging system 3 and a group of frames D1 to DN acquired in parallel by the video imaging unit 40, a frame Fn corresponds to a frame Dn (n=1, 2, . . . , N).

[0334] The image group evaluation unit 81 (image set evaluation unit 83) identifies landmarks in each of the frame groups D1 to DN. The landmarks may be, for example, image regions corresponding to the iris (iris regions).

[0335] Next, the image group evaluation unit 81 (image set evaluation unit 83) determines the arrangement order (spatial arrangement order) of the frame groups D1 to DN based on the positions (e.g., changes in spatial position) of the N landmarks respectively identified in the frame groups D1 to DN.

[0336] As described above, the frame groups D1 to DN are arranged in this order (temporal arrangement order). If the temporal arrangement order and the spatial arrangement order differ, the order may be changed or the position may be shifted due to the influence of eye movement. Also, a skip of a frame may occur due to the influence of a blink.

[0337] If such a defect is detected, that is, if the temporal arrangement order and the spatial arrangement order differ, the image group evaluation unit 81 (image set evaluation unit 83) determines that the quality of this image group (image set) is not good.

[0338] According to this configuration, a moving image that depicts a wide area of ​​the anterior segment is acquired (at least in the X-direction and Y-direction) in parallel with scanning the anterior segment with the slit light, and defects in a series of images included in an image group (image set) can be detected using the moving image.

[0339] Each of the series of images included in the image group (image set) is an image extending in the depth direction (Z direction). Therefore, in order to recognize the arrangement and alignment of the series of images in the X direction and Y direction perpendicular to the Z direction from the series of images themselves, it is necessary to use image processing (image analysis) such as the first aspect.

[0340] In this example, the quality evaluation of a series of images is realized by using a video image acquired separately in parallel with the scan, without applying image analysis to the series of images. Note that the evaluation items in this example may be any of the arrangement order of the series of images, missing images, and positional deviation, as in the first embodiment. It is also possible to combine the evaluation process of this example with any evaluation process described in the first embodiment.

[0341] Another application of the video capturing unit 40 will be described. In this example, when two or more scans are performed after the quality of the image group collected in the first scan is evaluated to be poor, the purpose is to prevent the positional deviation of the subject's eye E between scans by adjusting the start timing of the two or more additional scans.

[0342] Also in this example, the video capturing unit 40 captures an image of the anterior segment of the subject's eye E from a fixed position. The control unit 7 is configured to cause the scanning unit to start a second scan in response to the video capturing unit 40 acquiring an image that is substantially the same as a reference image acquired by the video capturing unit 40 in response to the start of a first scan among two or more additional scans.

[0343] More specifically, any one of two or more scans applied to the anterior segment of the eye to create an image set is called a first scan. In this example, the first scan among the two or more scans is the first scan.

[0344] First, the control unit 7 stores an image (reference image) acquired by the video capturing unit 40 at the start timing of the first scan. The reference image may be, for example, a frame acquired closest to the start of the first scan among a group of video frames whose capture was started before the start of the first scan. Alternatively, the video capturing unit 40 may acquire the reference image by capturing a still image of the anterior eye segment at the start of the first scan (immediately before, simultaneously with, or immediately after) the first scan under the control of the control unit 7.

[0345] Any scan (second scan) performed after the first scan is started while the video capture unit 40 is capturing video of the anterior segment. The control unit 7 (or the data processing unit 8) compares frames sequentially acquired by the video capture unit 40 with a reference image. This comparison may include any image processing, such as segmentation to identify landmarks, image matching, image correlation, etc.

[0346] When it is determined that a frame substantially identical to the reference image has been obtained, the control unit 7 controls the illumination system 2, the imaging system 3, and the moving mechanism 6 to start a second scan.

[0347] The slit lamp microscope of this embodiment repeatedly executes the above-mentioned series of processes until the number of scans applied to the anterior segment reaches the number of scan repetitions required to create an image set. This reduces the positional deviation of the subject's eye E between scans, and prevents deterioration of the image set quality due to eye movement. It also makes it possible to make image selection more efficient and easier.

[0348] The operation and usage of the slit lamp microscope of this embodiment will be described. An example of the operation and usage is shown in Fig. 9A and Fig. 9B. It is assumed that the various preparations (operations performed before step S1 in Fig. 5) described in the operation example of the first embodiment have already been performed. It is further assumed that the alignment (step S1 or S11), the first scan (step S2 or S12), and the evaluation of the quality of the image group (step S3 or S13) in the operation example of the first embodiment have already been performed. It is also assumed that the quality of the image group is not good in step S3 or S13 (step S4 "No" or S14 "No"). Step S31 in Fig. 9A is started, for example, when it is judged "No" in step S4 or when it is judged "No" in step S14.

[0349] In this example, it is assumed that any of the images in the series of images collected in step S2 or S12 are evaluated to be of poor quality, and therefore the image set created by the operations of Figures 9A and 9B does not include any images obtained in step S2 or S12.

[0350] If the quality of any of the images in the series of images included in the image group collected in step S2 or S12 is evaluated to be good, an image set including that image may be created. The operations to be performed in this case will be apparent to those skilled in the art from the operations shown in Figures 9A and 9B.

[0351] In this example, the number of images obtained in one anterior segment scan (the number of images included in the image group) is 256, and the number of images in the image set is also 256. The upper limit of the number of scan repetitions is 5. Unless otherwise specified, the matters related to the first aspect are applicable.

[0352] (S31: Alignment) First, alignment is performed in the same manner as in step S1 of the first embodiment. As a result, the illumination system 2 (illumination system 20) and the imaging system 3 (either the left imaging system 30L or the right imaging system 30R, or both) are disposed at a position corresponding to the scan start position, and the video imaging unit 40 is disposed, for example, in front of the subject's eye E.

[0353] (S32: Start video recording) After the alignment is completed, video capture of the anterior segment is started by the video capture unit 40. The capture rate of this video capture may be equal to the capture rate of the imaging system 3 in the anterior segment scan described below.

[0354] Although the illumination system 2 and the imaging system 3 are moved for the anterior eye scan, the video capture unit 40 captures video of the anterior eye from a fixed position. When the anterior eye scan is performed in parallel with the video capture, the above-mentioned correspondence is assigned between the image group collected by the anterior eye scan and the frame group collected by the video capture unit 40.

[0355] The timing for starting video capture may be arbitrary, for example, before or during alignment.

[0356] (S33: Anterior segment scan) As in the first aspect, the slit lamp microscope 1 scans the anterior segment of the subject's eye E by combining the irradiation of slit light by the illumination system 2, video capture by the imaging system 3, and movement of the illumination system 2 and the imaging system 3 by the movement mechanism 6.

[0357] In this example, 256 images are acquired in one scan (scan from the scan start position to the scan end position). The 256 images (image group) acquired in the first scan (the first scan executed in step S33; the same applies below) are indicated as H1(1) to H1(256). The 256 images H1(1) to H1(256) are assigned to 256 ordered positions in the scan range (the range in the X direction with the scan start position and scan end position at both ends).

[0358] (S34: Select a good quality image) The selection unit 821 selects images that satisfy a predetermined image selection condition from among the 256 images H1(1) to H1(256) acquired in step S33 in the same manner as in the first aspect.

[0359] In this example, as shown in Fig. 10A, it is assumed that the 71st to 170th 100 images H1(71) to H1(170) do not satisfy the image selection condition, and the 1st to 70th and 171st to 256th 156 images H1(1) to H1(70) and H1(171) to H1(256) satisfy the image selection condition. In this case, the selection unit 821 selects the 1st to 70th and 171st to 256th 156 images H1(1) to H1(70) and H1(171) to H1(256) from among the 256 images H1(1) to H1(256).

[0360] (S35: Create a provisional image set) The selection unit 821 creates a tentative image set based on the (one or more) images acquired up to the present time. At the present stage of this example, a tentative image set including 156 images H1(1) to H1(70) and H1(171) to H1(256) is created.

[0361] (S36: Number of selected images = predetermined number?) The selection unit 821 compares the number of images included in the temporary image set created in the immediately preceding step S35 (the number of selected images) with a predetermined number. As described above, in this example, the predetermined number (the number of images included in the image set) is set to 256.

[0362] If the number of images included in the provisional image set reaches 256 (S36: Yes), the operation proceeds to step S38. If the number of images included in the provisional image set does not reach 256 (S36: No), the operation proceeds to step S37.

[0363] (S37: Number of scans = predetermined number?) The selection unit 821 compares the number of scans performed up to the current stage (the number of times that scanning and image selection have been alternately repeated) with a predetermined number of times. As described above, in this example, the predetermined number of times is set to 5 times.

[0364] If the number of scans performed up to the current stage reaches 5 (S37: Yes), the operation proceeds to step S18. If the number of scans performed up to the current stage does not reach 5 (S17: No), the operation returns to step S33.

[0365] Steps S33 to S37 are repeated until step S36 or S37 returns "Yes." A "Yes" determination in step S36 corresponds to the number of images (256) required to create an image set being prepared. A "Yes" determination in step S37 corresponds to the scanning and image selection being repeated the maximum number of times.

[0366] (S38: Create image set) If the determination in step S36 or S37 is "Yes", the image set creation unit 82 creates an image set based on the provisional image set created in the immediately preceding step S35 in the same manner as in the first aspect.

[0367] Here, a specific example of steps S33 to S38 will be described. As described above, in the example of Fig. 10A, a tentative image set including 156 images H1(1) to H1(70) and H1(171) to H1(256) is created (step S35) based on the first scan (step S33) and the first image selection (step S34).

[0368] At this stage, the number of images included in the provisional image set has not reached 256 (S36: No), so the operation proceeds to step S37. At this stage, only one scan has been performed (S37: No), so the operation returns to step S33.

[0369] When the operation returns to step S33, a second anterior eye scan is performed. The 256 images obtained in the second scan are denoted by H2(1) to H2(256). As with the image group obtained in the first scan, the 256 images H2(1) to H2(256) are assigned to 256 ordered positions in the scan range. Furthermore, the image H1(n) obtained in the first scan and the image H2(n) obtained in the second scan are associated with each other via the 256 positions (n=1, 2, . . . , 256). The same applies to the image groups obtained in subsequent scans.

[0370] In step S34, the selection unit 821 selects images that satisfy a predetermined image selection condition from among the 256 images H2(1) to H2(256) acquired in step S33. In this example, as shown in FIG. 10B, it is assumed that the 100 images H2(1) to H2(100) from the 1st to 100th do not satisfy the image selection condition, and the 156 images H2(101) to H2(256) from the 101st to 256th satisfy the image selection condition. In this case, the selection unit 821 selects the 156 images H2(101) to H2(256) from the 256 images H2(1) to H2(256).

[0371] Here, when two or more images corresponding to a certain position in the scan range are selected, the selection unit 821 can select one of the images. In the example shown in Figures 10A and 10B, for example, image H1 (256) and image H2 (256) are selected as images corresponding to the 256th position.

[0372] The selection unit 821 may be configured to execute a process of selecting one of the image H1(256) and the image H2(256). For example, the selection unit 821 may be configured to select the image H1(256) acquired first, or to select the image H2(256) acquired later. In this example, it is assumed that the selection unit 821 is configured to select the image acquired first.

[0373] As another example, the selection unit 821 may be configured to select one of the images H1(256) and H2(256) by comparing them. For example, the selection unit 821 may be configured to calculate an image quality evaluation value for the image H1(256) and an image quality evaluation value for the image H2(256), compare these image quality evaluation values, and select one of the images H1(256) and H2(256). Typically, the image with the highest image quality evaluation value is selected.

[0374] This process assigns at most one image to each of the 256 positions in the scan range, meaning that the interim image set contains a maximum of 256 images.

[0375] The selection unit 821 creates a tentative image set based on two image groups obtained up to this point (image group H1(1) to H1(256) and image group H2(1) to H2(256)). At the current stage of this example, a tentative image set is created that includes 156 images H1(1) to H1(70) and H1(171) to H1(256) based on the first scan and 70 images H2(101) to H2(170) based on the second scan.

[0376] In other words, the provisional image set obtained at this stage includes 226 images H1(1) to H1(70), H2(101) to H2(170), and H1(171) to H1(256), corresponding to the 1st to 70th and 101st to 256th positions (226 positions) out of the 256 positions in the scan range.

[0377] At this stage, the number of images included in the provisional image set has not reached 256 (S36: No), so the operation proceeds to step S37. At this stage, only two scans have been performed (S37: No), so the operation returns to step S33 again.

[0378] When the operation returns to step S33, a third anterior segment scan is performed. The 256 images obtained in the third scan are denoted by H3(1) to H3(256).

[0379] In step S34, the selection unit 821 selects images that satisfy a predetermined image selection condition from among the 256 images H3(1) to H3(256) acquired in step S33. In this example, as shown in FIG. 10C, it is assumed that the 227th to 256th 30 images H3(227) to H3(256) do not satisfy the image selection condition, and the 1st to 226th 226 images H3(1) to H3(226) satisfy the image selection condition. In this case, the selection unit 821 selects the 1st to 226th 226 images H3(1) to H3(226) from among the 256 images H3(1) to H3(256).

[0380] Furthermore, the selection unit 821 selects 30 images H3(71) to H3(100) from among the 226 images H3(1) to H3(226), corresponding to the 71st to 100th positions, excluding the 1st to 70th and 101st to 256th positions of the 256 positions in the scan range to which images have already been assigned.

[0381] The selection unit 821 creates a tentative image set based on three image groups obtained up to this point (image group H1(1) to H1(256), image group H2(1) to H2(256), and image group H3(1) to H3(256)). At the current stage of this example, a tentative image set is created that includes 156 images H1(1) to H1(70) and H1(171) to H1(256) based on the first scan, 70 images H2(101) to H2(170) based on the second scan, and 30 images H3(71) to H3(100) based on the third scan.

[0382] That is, the tentative image set obtained at this stage includes 256 images H1(1) to H1(70), H3(71) to H3(100), H2(101) to H2(170), and H1(171) to H1(256) corresponding to all 256 positions in the scan range. Therefore, the answer in step S36 is "Yes," and the operation proceeds to step S38.

[0383] In step S38, the image set creation unit 82 creates an image set based on the 256 images H1(1) to H1(70), H3(71) to H3(100), H2(101) to H2(170), and H1(171) to H1(256) included in the temporary image set created in the previous step S35.

[0384] The acts shown in Figure 9A provide an example of a process for creating an image set using alternating iterations of scanning and image selection, followed by the acts shown in Figure 9B.

[0385] (S39: Evaluate the quality of the image set) The image set evaluation unit 83 evaluates the quality of the image set created in step S38 in the same manner as in the first embodiment.

[0386] Alternatively, in this embodiment, the quality of the image set may be evaluated using the video captured by the video capture unit 40 in the manner described above. For example, the image set based on the example shown in Figures 10A to 10C includes a series of 256 images H1(1) to H1(70), H3(71) to H3(100), H2(101) to H2(170), and H1(171) to H1(256). Three scans were performed to collect these images, and video capture was performed by the video capture unit 40 in parallel with each scan.

[0387] Among the frames obtained by video shooting in parallel with the first scan, the frame corresponding to image H1(n) is denoted by J1(n) (n = 1, 2, ..., 256). Among the frames obtained by video shooting in parallel with the second scan, the frame corresponding to image H2(n) is denoted by J2(n) (n = 1, 2, ..., 256). Among the frames obtained by video shooting in parallel with the third scan, the frame corresponding to image H3(n) is denoted by J3(n) (n = 1, 2, ..., 256).

[0388] Therefore, the 256 images included in the image set in this example, H1(1) through H1(70), H3(71) through H3(100), H2(101) through H2(170), and H1(171) through H1(256), are associated with frames J1(1) through J1(70), J3(71) through J3(100), J2(101) through J2(170), and J1(171) through J1(256), respectively.

[0389] The image set evaluation unit 83 can evaluate this image set by applying processing similar to the processing performed on the frame groups D1 to DN described above to the 256 frames corresponding to the 256 images contained in the image set in this example.

[0390] (S40: Good quality?) If it is determined in step S39 that the quality of the image set is good (S40: Yes), the operation proceeds to step S43. On the other hand, if it is determined in step S39 that the quality of the image set is not good (S40: No), the operation proceeds to step S41.

[0391] (S41: Prompt for reshooting) If it is determined in step S39 that the quality of the image set is not good (S40: No), the control unit 7 displays and / or outputs audio information to encourage the user to retake the image, in the same manner as in the first aspect.

[0392] (S42: Should we retake the photo?) In response to the information displayed and / or output as sound in step S41, the user performs an instruction operation to start re-imaging or an instruction operation to not perform re-imaging.

[0393] If the user performs an instruction operation to start re-imaging (S42: Yes), the operation returns to step S31. In this case, the slit lamp microscope executes the series of processes up to this point again. Note that re-imaging is repeated, for example, up to a predetermined number of times.

[0394] On the other hand, if the user performs an instruction operation not to perform re-shooting (S42: No), the operation proceeds to step S43.

[0395] (S43: Send image set) If the quality of the image set is determined to be good in step S39 (S40: Yes), or if the user performs an instruction operation in step S42 not to re-take the image (S42: No), the control unit 7 controls the communication unit 9 to send the image set to the reading terminal and / or reading device.

[0396] (S44: Reading the image set) The image interpretation terminal is a computer used by a doctor to perform image interpretation, and the image interpretation device is a computer having an image interpretation function.

[0397] When the image set is transmitted to the image reading terminal in step S43, the doctor uses the image reading terminal to read a series of images included in the image set. When the image set is transmitted to the image reading device in step S43, the image reading device reads the image set.

[0398] (S45: Create image interpretation report) If the doctor has performed the image interpretation in step S44, the doctor inputs the results of the image interpretation into a predetermined report template. If the image interpretation device has performed the image interpretation in step S44, the image interpretation device inputs the results of the image interpretation into a predetermined report template. This creates an image interpretation report for the image set.

[0399] (S46: Save image set and image reading report) The image set and the image interpretation report are provided to and stored in, for example, a facility where the slit lamp microscope 1 is installed, a medical institution designated by the subject, etc., an information processing device used by a doctor designated by the subject, etc., an address registered by the subject (email address, address, etc.), etc. Also, the image set and the image interpretation report may be sent to a specified database system for storage and management.

[0400] This concludes the explanation of the operation according to this example.

[0401] Several advantages of the slit lamp microscope of this embodiment will be described below. Note that, unless otherwise specified, explanations of advantages common to the slit lamp microscope 1 of the first embodiment will be omitted.

[0402] The slit lamp microscope of this embodiment can prevent the subject's eye from shifting between scans by matching the anterior segment image at the start of the scan in the alternating repetition of scanning and image selection. To this end, the slit lamp microscope of this embodiment includes, in addition to the elements of the first embodiment, a video capture unit 40 (capture unit) that captures the anterior segment of the subject's eye from a fixed position. The control unit 7 is configured to cause the scanning unit to start a second scan in response to the video capture unit 40 acquiring an image that is substantially the same as the reference image acquired by the video capture unit 40 in response to the start of the first scan.

[0403] The slit lamp microscope of this embodiment can use moving images (frame group) acquired by the moving image capturing unit 40 in evaluating the quality of the image set. To that end, the slit lamp microscope of this embodiment includes, in addition to the elements of the first embodiment, a moving image capturing unit 40 that captures moving images of the anterior segment of the subject's eye from a fixed position in parallel with application of a scan to the anterior segment of the subject's eye. The image set evaluation unit 83 is configured to evaluate the image set quality based on the moving images acquired by the moving image capturing unit 40.

[0404] In this case, the image set evaluation unit 83 may be configured to evaluate the quality of the image set based on the correspondence between the series of images included in the image set and the series of frames included in the video captured by the video shooting unit 40.

[0405] Furthermore, the image set evaluation unit 83 may be configured to evaluate the quality of the image set based on landmarks in a series of frames included in a video captured by the video shooting unit 40 and a correspondence between a series of images included in the image set and a series of frames included in the video.

[0406] According to the image set evaluation of this embodiment, it is possible to provide an evaluation method different from that of the first embodiment. It is optional whether to adopt the evaluation method of this embodiment or the evaluation method of the first embodiment.

[0407] <Third aspect> In this embodiment, an ophthalmic system including an ophthalmic imaging apparatus, an information processing apparatus, and an image interpretation terminal will be described. The ophthalmic imaging apparatus has at least a function as a slit lamp microscope. The slit lamp microscope included in the ophthalmic imaging apparatus may be the slit lamp microscope of the first or second embodiment. In the following, the elements, configurations, and symbols of the first and / or second embodiment will be appropriately applied mutatis mutandis.

[0408] The ophthalmologic system 1000 illustrated in FIG. 11 is constructed using a communication path (communication line) 1100 connecting each of T facilities (facility 1 to facility T) where ophthalmologic photography is performed, a server 4000, and an image reading terminal 5000m.

[0409] Here, the ophthalmic photography includes at least anterior ocular photography using a slit lamp microscope. This anterior ocular photography includes at least anterior ocular scanning using a slit light as described in the first and / or second aspects.

[0410] Each facility (the tth facility: t=1 to T, T is an integer equal to or greater than 1) includes an ophthalmic imaging device 2000-i t (i t =1~K t , K t In other words, each facility (the tth facility) is equipped with one or more ophthalmic imaging devices 2000-i t The Ophthalmic Imaging Device 2000-i is installed. t constitute a part of the ophthalmologic system 1000. Note that the ophthalmologic system 1000 may also include an examination device capable of performing examinations other than ophthalmologic examinations.

[0411] The ophthalmic imaging device 2000-i in this example thas both the function of an "imaging device" that takes an image of the subject's eye and the function of a "computer" that processes various data and communicates with external devices. In another example, the imaging device and the computer can be provided separately. In this case, the imaging device and the computer can be configured to be able to communicate with each other. Furthermore, the number of imaging devices and the number of computers are each arbitrary, and for example, a single computer and a plurality of imaging devices can be provided.

[0412] Ophthalmic Imaging Device 2000-i t The "imaging device" in the above includes at least a slit lamp microscope. This slit lamp microscope may be the slit lamp microscope of the first or second aspect.

[0413] Furthermore, each facility (t-th facility) is provided with an information processing device (terminal 3000-t) that can be used by assistants and subjects. The terminal 3000-t is a computer used in the facility, and may be, for example, a mobile terminal such as a tablet terminal or a smartphone, or a server installed in the facility. Furthermore, the terminal 3000-t may include a wearable device such as wireless earphones. Note that the terminal 3000-t is sufficient as long as it is a computer whose functions can be used in the facility, and may be, for example, a computer (cloud server, etc.) installed outside the facility.

[0414] Ophthalmic Imaging Device 2000-i t The terminal 3000-t may be configured to be able to communicate with the terminal 3000-t using a network (such as an in-facility LAN) established within the t-th facility, a wide area network (such as the Internet), or a short-range communication technology.

[0415] Ophthalmic Imaging Device 2000-i t may have a function as a communication device such as a server. t The server 4000 and the terminal 3000-t can be configured to directly communicate with each other.t Since the communication can be performed via the above, it is not necessary to provide a function for communicating between the terminal 3000-t and the server 4000.

[0416] The server 4000 is typically installed in a facility different from any of the first to Tth facilities, for example, in a management center. The server 4000 can communicate with an image interpretation terminal 5000m (m=1 to M, M is an integer equal to or greater than 1) via a network (LAN, wide area network, etc.). Furthermore, the server 4000 can communicate with the ophthalmic imaging apparatuses 2000-i installed in the first to Tth facilities. t and at least some of the above may be communicated with each other via a wide area network.

[0417] The server 4000 includes, for example, an ophthalmic imaging device 2000-i t and the image reading terminal 5000m, and a function for recording the contents of this communication; t and a function of storing data and information acquired by the image interpretation terminal 5000m. The server 4000 may be provided with a data processing function.

[0418] The image reading terminal 5000m is the ophthalmic imaging device 2000-i t The image interpretation terminal 5000m may include a computer that can be used to interpret images of the subject's eye acquired by the image interpretation terminal 5000m (e.g., a plurality of anterior eye images included in an image group or an image set, or a three-dimensional rendering image based on these images) and to generate a report.

[0419] The following describes the server 4000. The server 4000 illustrated in FIG.

[0420] The control unit 4010 executes control of each unit of the server 4000. The control unit 4010 may be capable of executing other arithmetic processing. The control unit 4010 includes a processor. The control unit 4010 may further include a RAM, a ROM, a hard disk drive, a solid state drive, etc.

[0421] The control unit 4010 includes a communication control unit 4011 and a transfer control unit 4012 .

[0422] The communication control unit 4011 controls the plurality of ophthalmic photographing apparatuses 2000-i t The communication control unit 4011 executes control regarding the establishment of communication between a plurality of devices including the plurality of terminals 3000-t and a plurality of image interpretation terminals 5000m. For example, the communication control unit 4011 transmits a control signal for establishing communication to each of two or more devices selected by a selection unit 4120 described later from among the plurality of devices included in the ophthalmologic system 1000.

[0423] The transfer control unit 4012 controls the exchange of information between two or more devices with which communication has been established by the communication establishment unit 4100 (and the communication control unit 4011). For example, the transfer control unit 4012 functions to transfer information transmitted from one of at least two devices with which communication has been established by the communication establishment unit 4100 (and the communication control unit 4011) to the other device.

[0424] As a specific example, the ophthalmic imaging device 2000-i t When communication between the ophthalmic imaging apparatus 2000-i and the image interpretation terminal 5000m is established, the transfer control unit 4012 t The transfer control unit 4012 can transfer information (e.g., a group of images or a set of images) transmitted from the image interpretation terminal 5000m to the image interpretation terminal 5000m. t instructions to the ophthalmic imaging device 2000-i t can be forwarded to.

[0425] The transfer control unit 4012 may have a function of processing information received from a source device. In this case, the transfer control unit 4012 can transmit at least one of the received information and information obtained by processing to a destination device.

[0426] For example, the transfer control unit 4012 transfers the ophthalmic imaging apparatus 2000-i t For example, the transfer control unit 4012 can extract a part of the information transmitted from the ophthalmologic imaging apparatus 2000-i, and transmit it to the image interpretation terminal 5000m. t A series of images can be extracted from the image group or image set transmitted from and transmitted to the image reading terminal 5000m.

[0427] In addition, Ophthalmic Imaging Device 2000-i t The server 4000 or other device may analyze information (e.g., a group of images or a set of images) transmitted from the ophthalmologic imaging device 2000-i, and transmit the analysis results (and the original information) to the image interpretation terminal 5000m, etc. t The interpretation of the image group or image set transmitted from can be performed using an artificial intelligence engine or the like, and the results can be transmitted to the interpretation terminal 5000m together with the image group or image set.

[0428] Ophthalmic Imaging Device 2000-i t When an image group or image set is transmitted from the server 4000 or other device, the server 4000 or other device can be configured to construct a three-dimensional image (e.g., stack data or volume data) from a series of images contained in the image group or image set, and the transfer control unit 4012 can be configured to transmit the constructed three-dimensional image to the reading terminal 5000m.

[0429] Ophthalmic Imaging Device 2000-i t When stack data is transmitted from the server 4000 or other device, the server 4000 or other device can be configured to construct volume data from this stack data, and the transfer control unit 4012 can be configured to transmit the constructed volume data to the reading terminal 5000m.

[0430] The data processing that can be performed by the server 4000 or other devices is not limited to the above examples and may include any data processing. For example, the server 4000 or other devices may be capable of rendering three-dimensional images, removing artifacts, correcting distortion, measuring, etc.

[0431] The communication establishment unit 4100 is a communication unit for establishing a communication between the plurality of ophthalmic photographing apparatuses 2000-i t and executes a process for establishing communication between at least two devices selected from a plurality of devices including the plurality of terminals 3000-t and a plurality of image reading terminals 5000m. In this aspect, "establishing communication" is a concept including at least one of, for example, (1) establishing one-way communication from a disconnected state, (2) establishing two-way communication from a disconnected state, (3) switching from a state in which only reception is possible to a state in which transmission is also possible, and (4) switching from a state in which only transmission is possible to a state in which reception is also possible.

[0432] Furthermore, the communication establishment unit 4100 can execute a process of disconnecting established communication. In this aspect, "disconnecting communication" is a concept including at least one of, for example, (1) disconnecting communication from a state in which one-way communication is established, (2) disconnecting communication from a state in which two-way communication is established, (3) switching from a state in which two-way communication is established to a state in which one-way communication is established, (4) switching from a state in which both transmission and reception are possible to a state in which only reception is possible, and (5) switching from a state in which both transmission and reception are possible to a state in which only transmission is possible.

[0433] Ophthalmic Imaging Device 2000-i t Each of the terminals 3000-i, 3000-t, and 5000m can transmit at least one of a communication request (call request) for calling another device (its user) and a communication request (interrupt request) for interrupting communication between the other two devices to the server 4000. The call request and the interrupt request are sent manually or automatically. The server 4000 (communication unit 4200) transmits the ophthalmic imaging devices 2000-i, 2000-t, and 5000m to the server 4000. t, receives a communication request transmitted from the terminal 3000-t or the image interpretation terminal 5000m.

[0434] In this embodiment, the communication establishment unit 4100 may include a selection unit 4120. The selection unit 4120 may select, for example, the ophthalmic imaging device 2000-i t , the ophthalmologic imaging device 2000-i based on a communication request transmitted from the terminal 3000-t or the image reading terminal 5000m. t , the terminal 3000-t, and the image interpretation terminal 5000m, one or more devices other than the device that transmitted the communication request are selected.

[0435] A specific example of the process executed by the selection unit 4120 will be described. t Or a communication request from the terminal 3000-t (e.g., the ophthalmic imaging device 2000-i t When a request for interpretation of an image acquired by the ophthalmologic imaging device 2000-i is received, the selection unit 4120 selects, for example, one of the multiple image interpretation terminals 5000m. The communication establishment unit 4100 establishes a connection between the selected image interpretation terminal 5000m and the ophthalmologic imaging device 2000-i. t and establish communication between at least one of the terminals 3000-t and 3000-i.

[0436] The selection of a device in response to a communication request is performed, for example, based on a preset attribute. Examples of the attribute include the type of examination (e.g., the type of imaging modality, the type of image, the type of disease, the type of candidate disease, etc.), the required expertise and proficiency, and the type of language. In this example, for example, the specialty and proficiency of the image reader are referenced. In order to realize the process according to this example, the communication establishment unit 4100 may include a storage unit 4110 in which attribute information created in advance is stored. The attribute information records attributes of the image reading terminal 5000m and / or its user (doctor, optometrist, etc.).

[0437] A user is identified by a pre-assigned user ID. An image reading terminal 5000m is identified by, for example, a pre-assigned device ID or network address. In a typical example, the attribute information includes, as attributes of each user, a specialty (e.g., medical department, specialized disease, etc.), a degree of specialty / proficiency, a type of language that can be used, etc.

[0438] When the selection unit 4120 refers to the attribute information, the ophthalmic photographing apparatus 2000-i t The communication request transmitted from the terminal 3000-t or the image reading terminal 5000m may include information on the attributes. t The image interpretation request (diagnosis request) transmitted from the ophthalmologic imaging device 2000-i may include any of the following information: (1) information indicating the type of imaging modality; (2) information indicating the type of image; (3) information indicating the name of a disease or a candidate disease; (4) information indicating the difficulty of image interpretation; and (5) information indicating the type of image interpretation. t and / or information indicating the language used by the user of terminal 3000-t.

[0439] When such an interpretation request is received, the selection unit 4120 can select one of the interpretation terminals 5000m based on the interpretation request and the attribute information stored in the storage unit 4110. At this time, the selection unit 4120 compares the information on the attributes included in the interpretation request with the information recorded in the attribute information stored in the storage unit 4110. As a result, the selection unit 4120 selects the interpretation terminal 5000m corresponding to a doctor (or optometrist) that corresponds to any of the following attributes, for example: (1) a doctor who specializes in the imaging modality; (2) a doctor who specializes in the image type; (3) a doctor who specializes in the disease (candidate disease); (4) a doctor who can perform interpretation of the difficulty level; (5) a doctor who can use the language.

[0440] The doctor or optometrist may be associated with the image interpretation terminal 5000m by a user ID entered when logging in to the image interpretation terminal 5000m (or the ophthalmology system 1000), for example.

[0441] The communication unit 4200 communicates with other devices (e.g., the ophthalmic imaging device 2000-i t , the terminal 3000-t, and the image reading terminal 5000m). The data communication method and encryption are t The communication unit may be the same as the communication unit (communication unit 9 in the first embodiment) provided in the

[0442] The server 4000 includes a data processing unit 4300. The data processing unit 4300 executes various data processing operations. t (particularly, a slit lamp microscope) can process multiple anterior eye images or three-dimensional images acquired by the microscope. The data processing unit 4300 includes a processor, a main storage device, an auxiliary storage device, etc. The auxiliary storage device stores a data processing program, etc. The functions of the data processing unit 4300 are realized by cooperation between software such as a data processing program and hardware such as a processor.

[0443] The server 4000 can provide the data obtained by the data processing unit 4300 to other devices. For example, the data processing unit 4300 can t When a three-dimensional image is constructed from a plurality of anterior eye images acquired by the ophthalmic imaging apparatus 2000-i, the server 4000 can transmit the three-dimensional image to the image interpretation terminal 5000m via the communication unit 4200. t Or, when the data processing unit 4300 renders the constructed three-dimensional image, the server 4000 can transmit the constructed rendering image to the image reading terminal 5000m via the communication unit 4200. When the data processing unit 4300 applies measurement processing to one or more anterior eye images or three-dimensional images, the server 4000 can transmit the obtained measurement data to the image reading terminal 5000m via the communication unit 4200. When the data processing unit 4300 applies distortion correction to one or more anterior eye images or three-dimensional images, the server 4000 can transmit the corrected images to the image reading terminal 5000m via the communication unit 4200.

[0444] Next, the interpretation terminal 5000m will be described. The interpretation terminal 5000m illustrated in FIG.

[0445] The control unit 5010 executes control of each unit of the image interpretation terminal 5000m. The control unit 5010 may be capable of executing other arithmetic processing. The control unit 5010 includes a processor, a RAM, a ROM, a hard disk drive, a solid state drive, and the like.

[0446] The control unit 5010 includes a display control unit 5011. The display control unit 5011 controls the display device 6000m. The display device 6000m may be included in the image interpretation terminal 5000m, or may be a peripheral device connected to the image interpretation terminal 5000m. The display control unit 5011 causes the display device 6000m to display an image of the anterior segment of the subject's eye E (for example, a series of images included in an image group or an image set).

[0447] The control unit 5010 includes a report creation control unit 5012. The report creation control unit 5012 executes various controls for creating a report regarding the information displayed by the display control unit 5011. For example, the report creation control unit 5012 causes the display device 6000m to display a screen or a graphical user interface (GUI) for creating a report. In addition, the report creation control unit 5012 inputs information input by the user, an image of the anterior segment, measurement data, analysis data, and the like into a predetermined report template.

[0448] The data processing unit 5100 executes various data processing operations. t(particularly, a slit lamp microscope). The data processing unit 5100 can process three-dimensional images or rendering images constructed by another information processing device such as the server 4000. The data processing unit 5100 includes a processor, a main storage device, an auxiliary storage device, etc. The auxiliary storage device stores a data processing program, etc. The functions of the data processing unit 5100 are realized by cooperation between software such as a data processing program and hardware such as a processor.

[0449] The communication unit 5200 communicates with other devices (e.g., the ophthalmic imaging device 2000-i t The data communication method and encryption method are described in detail below with respect to the ophthalmic photographing apparatus 2000-i. t The communication unit may be the same as that of the communication unit of the first embodiment.

[0450] The operation unit 5300 is used for operating the image reading terminal 5000m and inputting information to the image reading terminal 5000m. In this embodiment, the operation unit 5300 is used for creating a report. The operation unit 5300 includes an operation device and an input device. The operation unit 5300 includes, for example, a mouse, a keyboard, a trackball, an operation panel, a switch, a button, a dial, and the like. The operation unit 5300 may include a touch screen.

[0451] The ophthalmologic system 1000 of this embodiment is capable of executing the following operations.

[0452] First, the Ophthalmic Imaging Device 2000-i t The ophthalmic imaging device 2000-i (slit lamp microscope) scans the anterior segment of the subject's eye with a slit light to collect a group of images. If the quality of the group of images is evaluated to be good, t The ophthalmic photographing apparatus 2000-i transmits the first transmission information including the image group to the server 4000 through the communication line 1100. On the other hand, if the image group is evaluated to be of poor quality, t For example, the ophthalmic imaging device 2000-i applies two or more scans to the anterior segment of the eye.t selects a series of images corresponding to the scan range from the two or more image groups collected corresponding to the two or more scans (and the image group evaluated as having poor quality) to create an image set. t The ophthalmologic imaging apparatus 2000-i transmits the first transmission information including the created image set to the server 4000 through the communication line 1100. t The operation may be performed in the same manner as in the first or second embodiment. t may be capable of executing any process in the first or second aspect.

[0453] The server 4000 is an ophthalmic imaging device 2000-i t The server 4000 receives the first transmission information transmitted from the server 4000 via the communication unit 4200 (receiving unit) and stores the first transmission information in the storage unit 4110. Furthermore, the server 4000 transmits second transmission information including at least the image group or image set included in the first transmission information to the image interpretation terminal 5000m via the communication line 1100 using the communication unit 4200 (transmitting unit).

[0454] The image interpretation terminal 5000m receives the second transmission information transmitted from the server 4000 by the communication unit 5200 (receiving unit). A user (image reader) of the image interpretation terminal 5000m uses a user interface (operation unit 5300, display device 6000m, report creation control unit 5012, etc.) to interpret an image group or an image set. The image interpretation terminal 5000m transmits the third transmission information including information (image interpretation report, etc.) input using the user interface to the server 4000 by the communication unit 5200 (transmitting unit) through the communication line 1100.

[0455] The server 4000 receives the third transmission information transmitted from the image interpretation terminal 5000m by the communication unit 4200 (receiving unit), and stores this third transmission information in the storage unit 4110 in association with the first transmission information.

[0456] Such an ophthalmologic system 1000 can provide a doctor (radiographer) with a group or set of high-quality images created based on one or more scans of the anterior segment of the subject's eye, thereby making it possible to widely provide high-quality slit lamp microscopy examinations.

[0457] Furthermore, while conventionally, doctors perform examinations while operating the device from a remote location, in this aspect, the doctor only needs to interpret a group or set of images acquired in advance. In other words, this aspect frees the doctor from the time and effort required for photographing, allowing him or her to concentrate on interpreting images. Therefore, this aspect contributes to expanding the scope of provision of high-quality slit lamp microscopy.

[0458] <Fourth aspect> In this embodiment, an ophthalmic system including an ophthalmic imaging device, an information processing device, and an image reading device will be described. The difference from the third embodiment is that an image reading device is provided instead of an image reading terminal. It is also possible to combine the third embodiment with the fourth embodiment to construct an ophthalmic system including both an image reading terminal and an image reading device. The following description will be given by applying the elements, configurations, and symbols of any of the first, second, and third embodiments as appropriate.

[0459] As described above, the ophthalmologic system 1000A illustrated in FIG. 14 is obtained by replacing the image reading terminal 5000m of the ophthalmologic system 1000 according to the third embodiment with an image reading device 7000m. The image reading device 7000m uses, for example, an image processor or an artificial intelligence engine to read the images of the ophthalmologic imaging devices 2000-i t A computer that interprets a group or set of images acquired by a slit lamp microscope.

[0460] An example of the configuration of the image reading apparatus 7000m is shown in FIG. 15. The image reading apparatus 7000m of this example includes an image reading processing unit 7100 and a communication unit 7200. The communication unit 7200 communicates with other apparatuses (e.g., the ophthalmic imaging apparatus 2000-i t , terminal 3000-t, and server 4000).

[0461] The image interpretation processing unit 7100 includes, for example, an image interpretation processor that operates according to an image interpretation program, and analyzes a series of images included in an image group or image set to obtain findings. In some embodiments, the image interpretation processing unit 7100 may include an artificial intelligence engine in the first embodiment to obtain findings from a series of images included in an image group or image set. Furthermore, the image interpretation processing unit 7100 creates a report based on the obtained findings.

[0462] The ophthalmologic system 1000A of this embodiment is capable of executing the following operations.

[0463] First, the Ophthalmic Imaging Device 2000-i t A slit lamp microscope (microscope) scans the anterior segment of the subject's eye with a slit light to collect a group of images. In this example, as in the third embodiment, one or more scans are applied to the anterior segment to obtain a group or set of images. Then, first transmission information including the group or set of images is transmitted to the server 4000 via the communication line 1100.

[0464] The server 4000 is an ophthalmic imaging device 2000-i t The server 4000 receives the first transmission information transmitted from the server 4000 via the communication unit 4200 (receiving unit) and stores the first transmission information in the storage unit 4110. Furthermore, the server 4000 transmits second transmission information including at least the image group or image set included in the first transmission information to the image interpretation terminal 5000m via the communication line 1100 using the communication unit 4200 (transmitting unit).

[0465] The image reading apparatus 7000m receives the second transmission information transmitted from the server 4000 by the communication unit 7200 (receiving unit). The image reading apparatus 7000m performs image reading of a group of images or an image set by the image reading processing unit 7100. The image reading apparatus 7000m transmits fourth transmission information including the information acquired by the image reading processing unit 7100 to the server 4000 via the communication line 1100 by the communication unit 7200 (transmitting unit).

[0466] The server 4000 receives the fourth transmission information transmitted from the image reading apparatus 7000m by the communication unit 4200 (receiving unit), and stores this fourth transmission information in the storage unit 4110 in association with the first transmission information.

[0467] According to such an ophthalmic system 1000A, a group or set of good quality images created based on one or more scans of the anterior segment of the subject's eye can be provided to the image reading device, making it possible to widely provide high-quality slit lamp microscopy examinations.

[0468] Furthermore, since the work of a doctor (image reader) can be supported by automatic image interpretation using the image interpretation device, this embodiment contributes to expanding the scope of provision of high-quality slit lamp microscopy examinations.

[0469] The image reading device 7000m includes an artificial intelligence engine, and the ophthalmic imaging device 2000-i t If the ophthalmic imaging device 2000-i (slit lamp microscope) includes an artificial intelligence engine, the image group or image set was created. t A reading device 7000m equipped with an artificial intelligence engine with capabilities equivalent to that of a slit lamp microscope may be selected to provide a group or set of images.

[0470] This allows the ophthalmic imaging device 2000-i t This makes it possible to prevent the inconvenience of inconsistencies or errors occurring between the output from the artificial intelligence engine of the (slit lamp microscope) and the output from the artificial intelligence engine of the image reading device 7000m.

[0471] In addition, in an ophthalmic system including both the image reading terminal 5000m and the image reading device 7000m, the ophthalmic imaging device 2000-i that created the image group or image set t In the absence of an image reading device 7000m equipped with an artificial intelligence engine with capabilities equivalent to that of a slit lamp microscope, the image group or image set may be transmitted to the image reading terminal 5000m.

[0472] <Other matters> The above-described embodiments are merely examples of the present invention, and therefore any modifications (omissions, substitutions, additions, etc.) may be made to the above-described embodiments as appropriate within the scope of the present invention.

[0473] A method for controlling a slit lamp microscope according to any of the aspects can be provided. The slit lamp microscope includes a processor and a scanning unit that scans an anterior segment of a subject's eye with slit light to collect a group of images. The control method first causes the processor to control the scanning unit to apply a scan to the anterior segment of the subject's eye. The control method further causes the processor to evaluate the quality of the group of images collected by the scanning unit. In addition, the control method causes the processor to selectively perform two or more predetermined controls depending on a result of evaluating the quality of the group of images.

[0474] It is possible to configure a program for causing a computer to execute this control method. Furthermore, it is possible to create a computer-readable non-transitory recording medium on which this program is recorded. This non-transitory recording medium may be in any form, and examples of such a recording medium include a magnetic disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

[0475] Similarly, any of the control methods disclosed in any of the first to fourth aspects can be provided. Also, any of the processing methods (calculation methods, image processing methods, image analysis methods, etc.) disclosed in any of the first to fourth aspects can be provided. Furthermore, it is possible to configure a program that causes a computer to execute this processing method. In addition, it is possible to create a computer-readable non-transitory recording medium on which this program is recorded. [Explanation of symbols]

[0476] 1. Slit lamp microscope 2. Lighting system 3. Photography 4 Optical system 5. Image sensor 6 Moving mechanism 7. Control System 8 Data Processing Section 81 Image Group Evaluation Unit 811A Reasoning part 812A Inference Model 811B 3D Image Construction Unit 812B Comparison section 813B Evaluation Processing Unit 811C Evaluation Data Generation Unit 812C Evaluation Processing Unit 82 Image Set Creation Department 821 Selection Section 83 Image Set Evaluation Unit 9. Communications Department

Claims

1. a scanning unit that scans an anterior segment of a subject's eye with a slit light to collect a group of images; a first evaluation unit that evaluates the readability of the image group collected by the scanning unit using an inference model constructed by machine learning; Including, a slit lamp microscope.

2. The machine learning includes supervised learning and is performed using training data including a plurality of anterior segment images.

2. The slit lamp microscope of claim 1.

3. Each of the plurality of anterior eye images included in the training data of the supervised learning is labeled with a label indicating whether the image is readable or unreadable; The inference model is constructed by the supervised learning using the training data so that an image obtained by scanning the anterior segment with a slit light is input, and image reading possibility information indicating image reading possibility is output.

3. The slit lamp microscope of claim 2.

4. The inference model outputs a parameter indicating whether the image group is suitable for interpretation as the interpretation possibility information.

4. The slit lamp microscope of claim 3.

5. The inference model outputs, as the parameter of the image reading possibility information, at least one of a result of determining whether an image reading is possible or not, a probability that an image reading can be performed, and a validity of a result when an image reading is performed.

5. The slit lamp microscope of claim 4.

6. The first evaluation unit generates display information from the image interpretation possibility information. The slit lamp microscope according to any one of claims 3 to 5.

7. The first evaluation unit executes a predetermined statistical process based on the image reading possibility information. The slit lamp microscope according to any one of claims 3 to 7.

8. The machine learning includes unsupervised learning and is performed using training data including a plurality of anterior segment images.

2. The slit lamp microscope of claim 1.

9. a control unit that executes control to cause the scanning unit to execute a new scan of the anterior segment when the first evaluation unit evaluates that the image group is not readable, The slit lamp microscope according to any one of claims 1 to 8.

10. An image set creation unit that creates an image set by selecting a series of images corresponding to a scan range from two or more image groups including the image group and a new image group acquired by the new scan.

10. The slit lamp microscope of claim 9.

11. The image set creation unit includes: a selection unit that selects an image that satisfies a predetermined condition from the two or more image groups; creating the image set including the images selected by the selection unit; 11. The slit lamp microscope of claim 10.

12. Further comprising a second evaluation unit that evaluates the quality of the image set created by the image set creation unit.

12. A slit lamp microscope according to claim 10 or 11.

Images