Ophthalmologic apparatus

Abstract

To provide an ophthalmologic apparatus capable of providing information useful for diagnosis of a dry eye.SOLUTION: An ophthalmologic apparatus includes: a projection part for projecting pattern light on the corneal surface of an eye to be examined; an imaging part for capturing a reflection image from the corneal surface of the pattern light projected by the projection part; a control part for causing the imaging part to repeatedly capture the reflection image from the corneal surface at a predetermined time interval; a calculation part for calculating a sum total of the absolute values of the shake amount of the line of a pattern image by the pattern light of the captured image for each of the images captured by the imaging part; a graph creation part for creating a graph showing a time-series change in the sum total of the absolute values of the shake amount calculated by the calculation part; and a determination part for determining which of a plurality of change patterns set beforehand the graph created by the graph creation part corresponds to.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] The present disclosure relates to an ophthalmic apparatus, and more particularly to an ophthalmic apparatus that captures an image of a pattern of light projected onto a corneal surface and reflected from the corneal surface. [Background technology]

[0002] In order to measure the state of dry eye, an ophthalmic device has been developed that projects a pattern of light onto the corneal surface and captures an image of the pattern of light reflected from the corneal surface (for example, Patent Document 1). The ophthalmic device disclosed in Patent Document 1 projects a ring-shaped pattern of light onto the corneal surface and measures the state of the corneal surface and corneal shape by observing, capturing, and analyzing the image of the pattern of light reflected from the corneal surface. The state of dry eye in the examinee's eye is then determined by comparing the change over time in the corneal shape after the start of measurement. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2004-321508 A Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, when ophthalmologists diagnose dry eye, they perform the examination according to the dry eye treatment guidelines (dry eye diagnosis flow). Specifically, the cornea is stained with fluorescein, the eyelids are closed and then quickly opened, and the breakup time of the corneal stain is measured. However, it has become clear that there are multiple types of dry eye, and the type of dry eye cannot be diagnosed by the breakup time of the corneal stain alone. Conventional ophthalmic devices can only obtain limited information through measurements, and do not provide sufficient information to diagnose dry eye.

[0005] The present specification discloses an ophthalmic device that can provide information useful in diagnosing dry eye. [Means for solving the problem]

[0006] In a first aspect of the technology disclosed in this specification, an ophthalmic apparatus includes a projection unit that projects pattern light onto a corneal surface of a test eye, an imaging unit that captures a reflected image of the pattern light projected by the projection unit from the corneal surface, a control unit that causes the imaging unit to repeatedly capture the reflected image from the corneal surface at a predetermined time interval, a calculation unit that calculates, for each captured image captured by the imaging unit, a sum of absolute values ​​of the amount of blur of lines in the pattern image caused by the pattern light of the captured image, a graph generation unit that generates a graph showing time-series changes in the sum of the absolute values ​​of the amount of blur calculated by the calculation unit, and a determination unit that determines which of a plurality of pre-set change patterns the graph generated by the graph generation unit corresponds to.

[0007] In the above ophthalmic device, for each captured image, the sum of absolute values ​​of the blurring amount of the pattern image caused by the pattern light of the captured image is calculated. That is, the sum of absolute values ​​of the blurring amount of the pattern image of each captured image is calculated without using the image captured immediately after opening the eyelids as a reference. Therefore, the change in the tear film on the corneal surface can be correctly evaluated. Furthermore, the inventors have studied and found that by determining the change pattern of the graph showing the time series change in the sum of the absolute values ​​of the blurring amount, the type of dry eye can be identified or narrowed down from the change pattern. Therefore, even if it is difficult to determine the type of dry eye only from the breakup time of the corneal staining, it becomes easier to determine the type of dry eye by determining the change pattern of the graph. [Brief description of the drawings]

[0008] [Figure 1] 1 is a schematic diagram showing an overall configuration of an optical system of an ophthalmic apparatus according to an embodiment; [Diagram 2] FIG. 2 is a block diagram showing the configuration of a control system of the ophthalmic apparatus according to the embodiment. [Diagram 3] 11 is a flowchart showing an example of a process for measuring a temporal change in the tear film formed on the corneal surface. [Figure 4] FIG. 2 is a diagram showing an example of an image (original image) obtained by capturing a reflected image from the surface of the cornea. [Diagram 5] 10 is a flowchart showing an example of a process for calculating the sum of absolute values ​​of the amounts of blur of a ring image. [Figure 6] 5 is a diagram showing an example of a polar coordinate transformed image obtained by polar coordinate transformation of the original image shown in FIG. 4. [Figure 7] 7 is a diagram showing an example of a corrected image obtained by correcting the distortion of the polar coordinate converted image shown in FIG. 6. [Figure 8] 11 is a flowchart showing an example of a process for generating a display image. [Figure 9] FIG. 4 is a diagram showing an example of a modulation portion superimposed image in which a modulation portion is superimposed on a captured original image. [Figure 10] 1 is a table showing the relationship between corneal staining breakdown time, subtype, condition of the test eye, and treatment method. [Figure 11] 5A and 5B are schematic diagrams showing a change pattern of the amount of blur. [Figure 12] 10 is a flowchart showing an example of a process for determining a change pattern of the amount of shaking. [Figure 13] 11 is a flowchart showing an example of a process for generating an SPK map. [Figure 14] FIG. 4 is a diagram showing an example of regions into which the subject's eye is divided. [Figure 15] FIG. 13 is a diagram showing an example of an SPK map. [Figure 16] FIG. 4 is a diagram showing an example of a display screen. [Figure 17] 11A and 11B are diagrams showing examples of display of determination results of a change pattern of the amount of blur. [Figure 18] 11A and 11B are diagrams showing graphs of the amount of shaking, in which (a) shows a state without offset, and (b) shows a state in which the graph of the amount of shaking is offset. [Figure 19] 11A and 11B are diagrams showing graphs of the amount of shaking, in which (a) shows a state without offsetting, and (b) shows a state with a first predetermined value (threshold value) offset. [Figure 20]FIG. 13 is a diagram showing another example of the SPK map. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] The main features of the embodiments described below are listed below. Note that the technical elements described below are independent technical elements that exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing.

[0010] In a second aspect of the technology disclosed in the present specification, the ophthalmologic apparatus in the first aspect may further include a display unit that displays the change pattern of the graph determined by the determination unit. With this configuration, it is possible to visually inform the examiner of which change pattern the graph corresponds to.

[0011] In a third aspect of the technology disclosed in the present specification, in the second aspect, the display unit may display a plurality of schematic diagrams each showing a plurality of change patterns, and may display the change pattern of the graph determined by the determination unit in a manner different from the other change patterns. According to this configuration, by displaying the schematic diagrams of the plurality of change patterns, it is possible to inform the examiner of what change patterns are available. In addition, by displaying the change pattern to which the graph corresponds in a manner different from the other change patterns, it is possible to visually determine which of the plurality of change patterns the graph corresponds to.

[0012] In a fourth aspect of the technology disclosed in the present specification, in the second or third aspect, the display unit may display the graph generated by the graph generating unit and the change pattern of the graph determined by the determining unit on the same screen. With this configuration, the examiner can visually grasp the actually generated graph and which change pattern the graph corresponds to.

[0013] In a fifth aspect of the technique disclosed in the present specification, in any one of the second to fourth aspects, the ophthalmologic apparatus may further include an eyelid opening detection unit that detects that the eyelid of the subject's eye is opened. The display unit may display a threshold value, in which the sum of the absolute values ​​of the amount of blur indicates the destruction of the tear film of the subject's eye, superimposed on the graph. The graph generating unit may correct the sum of the absolute values ​​of the amount of blur calculated for each of the images captured by the photographing unit or the threshold value so that the sum of the absolute values ​​of the amount of blur calculated for the images captured by the photographing unit immediately after the eyelid opening detection unit detects the opening of the eyelid of the subject's eye becomes a predetermined value. With this configuration, it is possible to easily grasp how the sum of the absolute values ​​of the amount of blur calculated for the images captured thereafter has changed with respect to the sum of the absolute values ​​of the amount of blur calculated for the images captured immediately after the eyelid opening is detected. EXAMPLES

[0014] Hereinafter, an ophthalmic device according to an embodiment will be described. As shown in FIG. 1, the ophthalmic device of this embodiment includes a cone 1 positioned facing the cornea of ​​the subject's eye C, and an illumination device 2 (e.g., LED, etc.) arranged on the back of the cone 1. The cone 1 is a hollow cone-shaped cylinder made of transparent resin. A transparent film with a concentric pattern printed thereon is attached to the inner wall surface of the cone 1, while a light-reflecting paint is applied to the outer wall surface of the cone 1. Therefore, the illumination light from the illumination device 2 arranged on the back of the cone 1 is scattered within the cone 1, and part of the light is blocked by the transparent film attached to the inner wall surface, and is projected onto the cornea of ​​the subject's eye C. As a result, a concentric pattern light (ring light) as shown in FIG. 4 is projected onto the cornea of ​​the subject's eye. As a method for projecting a desired pattern using the cone 1, a method of directly cutting a desired pattern inside the cone 1 and applying a light-shielding paint to that part can also be adopted.

[0015] A lens 3 and a half mirror 4 are arranged on the back of the illumination device 2. The light of a fixation light point light source 6 is incident on the half mirror 4 via a lens 5. The light of the fixation light point light source 6 is reflected by the half mirror 4, passes through the lens 3, and enters the cone 1. The light that enters the cone 1 is irradiated onto the cornea of ​​the subject's eye C from the central opening of the cone 1. The light of the fixation light point light source 6 is used as a fixation light during corneal observation, and is adjusted to be located at the center of the concentric pattern formed by the cone 1. In addition, the light reflected on the corneal surface of the subject's eye C (i.e., the reflected image (ring image) of the concentric pattern of light projected on the corneal surface and the reflected image of the light of the fixation light) passes through the cone 1, the lens 3, and the half mirror 4, and is observed by a CCD camera 8 after focus adjustment by the lens 7.

[0016] In order to accurately measure the corneal shape with the above-mentioned ophthalmic device, it is preferable to project the ring pattern in a concentric shape centered on the corneal apex of the subject's eye C. To achieve this, the fixation light point light source 6 is turned on during corneal shape measurement, and the subject is made to fixate on the light. The controller 12 (shown in FIG. 2) adjusts the position of the optical system with respect to the subject's eye C so that the light from the fixation light point light source 6 observed by the CCD camera 8 is at the center of the image. This makes it possible to project the light of the ring pattern (ring light) in a concentric shape centered on the corneal apex of the subject's eye C.

[0017] The configuration of the control system of the above-mentioned ophthalmic apparatus will be described. As shown in FIG. 2, the ophthalmic apparatus is controlled by a controller 12. The controller 12 can be configured by a computer equipped with a CPU, a ROM, a RAM, and the like. The controller 12 is connected to a CCD camera 8, an image capture start switch 11, an input device 13, and a memory 10. The image capture start switch 11 and the input device 13 are operated by an examiner. When the examiner operates the image capture start switch 11, the controller 12 starts a predetermined program and measures the amount of change over time of the tear film formed on the corneal surface. The image data captured by the CCD camera 8 is input to the controller 12 and stored in the memory 10, etc. The image data captured by the CCD camera 8 is subjected to image processing by the controller 12. The controller 12 processes the image data captured by the CCD camera 8 to provide information useful for diagnosing dry eye. The image processing performed by the controller 12 will be described in detail later.

[0018] In addition, the controller 12 is connected to the illumination device 2, the fixation light spot light source 6, and the display device 9. The controller 12 turns on / off the illumination device 2 and the fixation light spot light source 6, and also displays images captured by the CCD camera 8, etc. on the display device 9.

[0019] Next, a procedure for measuring the temporal change of the tear film formed on the corneal surface using the above-mentioned ophthalmic apparatus will be described. As shown in FIG. 3, first, the ophthalmic apparatus captures an image of the anterior part of the subject's eye (S10). Specifically, the examiner turns on the fixation light point light source 6 and instructs the subject to fixate the fixation light. When the fixation light point light source 6 is turned on, an image observed by the CCD camera 8 is displayed on the display device 9. The controller 12 adjusts the position of the optical system with respect to the subject's eye C so that the anterior part of the subject's eye displayed on the display device 9 is located at the center of the screen (i.e., so that the reflected light by the fixation light point light source 6 is positioned at the center of the cornea C). After the position adjustment of the optical system is completed, the controller 12 turns on the illumination device 2 and projects a concentric ring light on the corneal surface of the subject's eye C. This completes preparations for capturing an image of the anterior part of the subject's eye C. Next, the examiner operates the imaging start switch 11 and has the examinee close his / her eyes to form a homogeneous tear film on the corneal surface, and then has the examinee open his / her eyes. When the examiner operates the imaging start switch 11, the controller 12 operates the CCD camera 8 to capture an image reflected from the corneal surface.

[0020] Next, the controller 12 performs a process of analyzing the images captured by the CCD camera 8 (S12 to S16). That is, in this embodiment, the reflected images from the corneal surface are captured at a frame rate of 15 frames per second until a set time (e.g., 10 seconds) has elapsed since the capture start switch 11 was operated. Then, the process of steps S12 to S16 is executed for each captured image. FIG. 4 shows an example of an image (i.e., an original image) captured by the CCD camera 8. Note that the examiner is instructed not to blink until a set time has elapsed since the capture start switch 11 was operated (i.e., the capture process is completed).

[0021] The process of steps S12 to S16 will be described in detail below. First, the controller 12 calculates the sum of the absolute values ​​of the blurring amounts of the ring images (described in detail later) (S12). The process of calculating the sum of the absolute values ​​of the blurring amounts of the ring images is executed in the following procedure. As shown in FIG. 5, first, the controller 12 recognizes the center of the ring image in the captured image (S100), and performs polar coordinate conversion with the center as the origin (S110). In this embodiment, the image is converted to an image of 240 pixels in the radial direction at 1° intervals in the circumferential direction. As shown in FIG. 4, a plurality of ring images are captured in a concentric manner in the acquired image. Therefore, when the center of the ring image is converted to the origin and polar coordinate conversion is performed, the image shown in FIG. 6 (polar coordinate converted image) is obtained. As is clear from FIG. 6, in the polar coordinate converted image, the ring image is converted to a straight line extending in the up and down direction (y direction). Since the ring image is converted to an image extending in a straight line (hereinafter, sometimes referred to as a straight line image), the process of step S120 and subsequent steps can be easily performed. However, since this image is a reflected image from the corneal surface, it may become a curved shape with a major axis and a minor axis in a certain direction, as represented by a trigonometric function, depending on the degree of astigmatism, the axial angle, and the deviation of the center recognition. Note that the center recognition of the ring image in step S100 can be performed, for example, by using the ring image located at the most center (i.e., the ring image with the smallest radius). Alternatively, if the light of a fixation lamp is reflected in the original image, the light of the fixation lamp may be used as the center. Note that the center of the ring image in step S100 does not need to coincide with the center for performing the known corneal shape analysis (corneal topography).

[0022] Next, the controller 12 corrects the image converted into polar coordinates in step S110 (S120). In the image captured in step S10 (the original image shown in FIG. 4), the image (ring image) is distorted due to an unstable tear film, and the ring image is partially missing due to the inclusion of eyelash shadows. Therefore, the linear image of the polar coordinate converted image obtained in step S110 is also distorted and partially missing. For example, the image shown in FIG. 6 includes an area where the image is distorted due to an unstable tear film, and an area where the ring image is partially missing due to the inclusion of eyelash shadows. In step S120, artifacts such as the distorted image and partial missing that occur in the linear image of the polar coordinate converted image are corrected (removed).

[0023] That is, the image is restored by taking the average of the brightness of the pixel group in the averaging area set in the neighborhood area expanded in the radial direction and the circumferential direction from the pixel in the polar coordinate conversion image. In this embodiment, a plurality of ring lights arranged in a concentric circle are irradiated on the corneal surface of the eye C to be examined. Therefore, if there is no disturbance or partial loss in the ring image, the linear image converted into polar coordinates becomes an image of a substantially linear shape extending in the vertical direction (y direction). Therefore, for example, when a certain pixel is used as a reference, even if the angle from the reference pixel changes, it is considered that the distance between the rings does not change significantly at the angle near the reference pixel. That is, it is considered that the radial position of the ring image does not change significantly. Also, even if the eye to be examined has astigmatism, it is considered that the distance between the rings does not change significantly in the angular direction. Therefore, in this embodiment, by utilizing the characteristic that the radial position of the ring image hardly changes when there is no distortion or loss in the ring image, the distortion of the ring image is corrected and the loss of the ring image is compensated for for the part where the ring image has distortion or loss.

[0024] The luminance of each pixel in step S120 can be calculated in various ways. For example, a linear area extending a preset range in the circumferential direction from the pixel for which the luminance is to be calculated may be used as the averaging area, and the luminance of each pixel in the averaging area may be averaged. Alternatively, a rectangular area simply extending a preset range in the radial and circumferential directions from the pixel for which the luminance is to be calculated may be used as the averaging area, and the luminance of each pixel in the averaging area may be averaged.

[0025] Alternatively, the range in which the average value is calculated in the vicinity of the pixel for which the luminance is to be calculated may be changed in the radial direction to identify an averaging area with the smallest amount of error, and the luminance of each pixel in the identified averaging area may be averaged. An example of the adoption of such a method will be specifically described using the formulas 1 and 2. Here, the position of the pixel for which the luminance is to be calculated is set to an angle a in the circumferential direction and a radius r, and its luminance is set to P(a,r). The range in which the average value is calculated is set to a calculation range in the radial direction of ±r w The calculation range in the circumferential direction is ±a w As described above, if the ring image is free of distortion or loss, the radial position of the ring image will hardly change, whereas if the ring image is distorted or lost, the radial position of the ring image will shift. The radial shift of the ring image varies depending on the circumferential position (angle a) and the radial position r. Therefore, the shift of this ring image is taken as h(a,r). As described above, the shift h(a,r) is a function whose value changes depending on the angle a and the radius r. In order to identify the averaging region with the smallest error, the angle a' and the shift h(a',r) that minimizes S(a',r) shown in Equation 1 are found. Here, a' is an angle near the angle a.

[0026]

number

[0027] Once the angle a' and the amount of deviation h(a',r) are determined, the average brightness P'(a,r) of pixel (a,r) is calculated using equation 2. In this way, the distortion and defects of the ring image are corrected by averaging the brightness of each pixel in the polar coordinate transformed image.

[0028]

number

[0029] FIG. 7 shows a corrected image corrected using the formulas 1 and 2. As is clear from a comparison of FIG. 6 and FIG. 7, the distortion of the straight line image has been corrected in the corrected image, and the partial loss of the straight line image due to the shadow of the eyelashes has also disappeared. Note that even in the corrected image, distortion of the straight line image remains at the bottom of the image. This is thought to be due to the fact that the straight line image was significantly distorted in the image before correction (FIG. 6), but it has been corrected to a level that is practically problem-free. Note that in the corrected image shown in FIG. 7, the notched parts of the ring at the angle directions of 0° and 180° are slightly shifted and filled in, but this is because there is no data outside the position where the loss begins. Also, since there is no need to repair the areas where data is missing due to the upper and lower eyelids, this part is not repaired.

[0030] Next, the controller 12 performs brightness correction on the corrected image corrected in step S120, and also performs brightness correction on the polar coordinate converted image before being corrected in step S120 (S130). In this embodiment, in order to detect a decrease in brightness or a positional displacement of the ring image due to instability of the tear film, the corrected image corrected in step S120 is compared with the polar coordinate converted image before being corrected in step S120 (hereinafter, sometimes referred to as the image before correction). In order to properly compare the two images, first, the brightness of each image is corrected. Note that the processing after step S130 is performed on the region that does not cover the upper and lower eyelids, which is cut out from the photographed image and the corrected image corrected in step S120.

[0031] Specifically, the controller 12 first obtains a luminance waveform showing the relationship between "radial position" and "luminance" for each angle for the corrected image corrected in step S120, and calculates an inclusive line that includes the maximum and minimum values. Here, it is assumed that the inclusive line calculated from the corrected image includes the upper and lower limit values ​​of the luminance of the pre-corrected image, and creates an inclusive line for the pre-corrected image from the corrected image. That is, if the upper and lower limit values ​​of the pre-corrected image exceed the inclusive line of the corrected image, the upper and lower limit values ​​of the pre-corrected image are adopted to correct the inclusive line of the corrected image, and create an inclusive line for the pre-corrected image. In addition, the position of the ring image in the corrected image corrected in step S120 is corrected to the correct position. If there is no distortion of the ring image due to disturbance of the tear film, it is considered that the position of the ring image in the pre-corrected image and the position of the ring image in the corrected image coincide with each other and are not significantly different. For this reason, the peak position of the ring image in the pre-corrected image is detected in the vicinity of the peak position of the corrected image.

[0032] When the inclusive line is calculated in this manner, the controller 12 uses the created inclusive line to expand the luminance waveform so that it falls within a preset luminance range (for example, gray level 20 to 200 (full span: gray level 0 to 255)). It is preferable to expand the luminance waveform in a range narrower than the full span. By expanding in a range narrower than the full span, it is possible to prevent the luminance from exceeding the upper and lower limits in the calculation process. By expanding the luminance waveform of the corrected image and the luminance waveform of the uncorrected image, the peak heights of each ring image are normalized to be approximately the same. The luminance correction performed in step S130 is disclosed in detail in JP 2020-10986 A.

[0033] Next, the controller 12 calculates the sum of absolute values ​​of the blurring amounts of the ring image (straight line image) by subtracting the luminance-corrected corrected image from the uncorrected image that has been luminance-corrected by the above process (S140). Specifically, the controller 12 first calculates, for each pixel of the uncorrected image, the luminance difference between that pixel and the pixel of the corrected image corresponding to that pixel. For example, if there is no distortion or loss in the ring image, when a pixel of the uncorrected image is bright or dark (i.e., on or off the line image), the corresponding pixel of the corrected image is bright or dark (i.e., on or off the line image). Therefore, the luminance difference between the corresponding pixels of the uncorrected image and the corrected image is small. On the other hand, if there is distortion or loss in the ring image, when a pixel of the uncorrected image is bright or dark (i.e., on or off the line image), the corresponding pixel of the corrected image is dark or bright (i.e., off the line image or on the line image). Therefore, the luminance difference between the corresponding pixels of the uncorrected image and the corrected image is large. Furthermore, as distortion or loss in the ring image increases, the number of pixels with a large luminance difference between corresponding pixels in the uncorrected image and the corrected image increases. In this embodiment, for each pixel in the uncorrected image, the luminance difference between that pixel and its corresponding pixel in the corrected image is calculated, and the sum of the absolute values ​​of the calculated luminance differences is calculated to obtain the total absolute value of the amount of blur.

[0034] Here, if the ring image is distorted or missing, when a pixel in the pre-correction image becomes bright (i.e., the pixel is a pixel on the ring image in the pre-correction image), the corresponding pixel in the corrected image becomes dark (i.e., because the ring image is distorted, the corresponding pixel in the corrected image shifts from its position on the ring image and becomes dark). Since the ring images are arranged at a predetermined interval in the radial direction, when a pixel in the pre-correction image becomes bright, a pixel shifted by 1 / 2 the predetermined interval in the radial direction from the pixel becomes dark. Therefore, a combination in which a pixel in the pre-correction image is bright and a corresponding pixel in the corrected image is dark, and a combination in which a pixel in the pre-correction image is dark and a corresponding pixel in the corrected image is bright, form a pair. Therefore, when evaluating the sum of the absolute values ​​of the amount of blur, it is not necessary to count both. In this embodiment, only when a pixel in the pre-correction image is bright and a pixel in the corrected image is dark, they are counted when calculating the sum of the absolute values ​​of the amount of blur. As a result, if there is a defect in the ring image due to eyelashes or the like (i.e., if the pixels in the pre-correction image are dark and the pixels in the corrected image are bright), it is not counted when calculating the sum of the absolute values ​​of the amount of blur. This allows only the amount of blur of the ring image (straight line image) to be properly evaluated.

[0035] The calculation of the sum of the absolute values ​​of the amount of blur is not limited to the above method, and various methods can be used. For example, the ring image (bright part) of the pre-correction image is thinned into a line, and the ring image (bright part) of the corrected image is thinned into a line, and the deviation and brightness difference of the two lines are quantified. Then, the deviation amount (absolute value) of the two lines and the brightness difference (absolute value) of the two lines may be weighted by a predetermined amount, and the sum of the absolute values ​​of the amount of blur may be calculated by adding them up. After that, the process proceeds to step S14 in FIG. 3.

[0036] When the process of calculating the absolute value of the amount of blur of the ring image in step S12 is completed, the controller 12 calculates the eyelid opening length of the subject's eye C from the eye image captured in step S10 (S14). Specifically, the controller 12 detects the upper and lower parts of the eye based on the luminance of the eye image captured in step S10, and calculates the distance between them.

[0037] Next, the controller 12 generates a display image (S16). The display image generated in step S16 is an image showing the modulation location, specifically, a pseudo-fluorescein image described later. The process of generating the display image is executed in the following procedure. As shown in FIG. 8, first, the controller 12 generates a modulation location extraction image in which the modulation location is extracted by subtracting the brightness-corrected corrected image from the uncorrected image whose brightness has been corrected by the above process (S200). That is, if there is no change at all in the state of the tear film, the brightness and position of the ring image in the brightness-corrected uncorrected image will be the same as the brightness and position of the ring image in the brightness-corrected corrected image. Conversely, if the brightness and position of the ring image change due to instability of the tear film, they become the modulation location of the two images. In step S200, the difference between the two images is taken to extract the changed portion of the tear film (the modulation location). Note that the difference image obtained by taking the difference between the two images is an image that has been polar coordinate converted, and it is difficult to grasp the position on the cornea where the modulation occurs. Therefore, a difference image obtained by taking the difference between the two images may be converted into an orthogonal coordinate system and plotted on the eye image. FIG. 9 shows an image in which the modulation location is plotted on the eye image. As shown in FIG. 9, by plotting the modulation location on the eye image, it is possible to easily grasp where on the eye the tear film is destabilized. Note that the image shown in FIG. 9 does not have the ring image removed, but an image in which the ring image is removed may be generated. That is, the ring image may be removed from the captured image, and the extracted modulation location may be superimposed on the image from which the ring image is removed. Alternatively, the subject's eye C may be photographed without projecting the ring light, and the extracted modulation location may be superimposed on the captured image. When the ring image is removed, it becomes easier to recognize the boundaries (edges) of the iris and pupil, and therefore it becomes easier to recognize where on the cornea the modulation location is occurring.

[0038] Next, the controller 12 converts the image in which the modulation portion is extracted by the process of step S200 into a pseudo-fluorescein image in which the degree of modulation is expressed by brightness (S210). That is, the modulation portion extraction image in which the modulation portion of the tear film (i.e., the modulation portion between the uncorrected image and the corrected image) is specified is generated by the process of step S200. In the modulation portion extraction image, if the brightness difference between the uncorrected image and the corrected image is large, the degree of modulation of the tear film is large, and if the brightness difference between the uncorrected image and the corrected image is small, the degree of modulation of the tear film is small. Here, the fluorescein image is generally used in dry eye examinations, and the brightness of the stained portion changes in a direction decreasing according to the degree of breakdown of the tear film. In this embodiment, the modulation portion extraction image is converted into a pseudo-fluorescein image by decreasing the brightness as the degree of modulation portion increases in the modulation portion extraction image. That is, in step S210, the modulation portion extraction image is converted into a pseudo-fluorescein image by changing the gradation with a color tone similar to fluorescein according to the degree of modulation.

[0039] Next, the controller 12 superimposes the pseudo fluorescein image generated in step S210 on the eye image from which the ring image has been removed (S220). The pseudo fluorescein image generated in step S210 is an image converted into polar coordinates, and it is difficult to grasp the position on the cornea where the modulation occurs. Therefore, the pseudo fluorescein image generated in step S210 is converted into Cartesian coordinates and superimposed on the eye image. This makes it easier to recognize the position on the cornea where the modulation occurs, and also results in an image that is similar to a fluorescein image obtained in a dry eye test. Therefore, an image useful for diagnosing dry eye can be provided. Note that the eye image on which the pseudo fluorescein image is superimposed may be an image from which the ring image has been removed from a captured image, or may be an image of the subject's eye C captured without projecting the ring light. The image on the left side of FIG. 16 is an example of an image generated by step S220. By superimposing an image with a color tone similar to fluorescein and a brightness change at the modulation point on the eye image, an image similar to an image obtained in a dry eye test is obtained, and it is easy to understand where in the eye the tear film is destabilized. The generation of the pseudo-fluorescein image performed in steps S210 and 220 is disclosed in detail in JP 2020-10986 A. Then, proceed to the processing of step S18 in FIG. 3.

[0040] When the display image generation process of step S16 ends, the controller 12 judges whether or not a set time (e.g., 10 seconds) has elapsed since the photographing start switch 11 was operated (S18). If 10 seconds have elapsed since the photographing start (YES in S18), the process proceeds to step S20. If 10 seconds have not elapsed since the photographing start (NO in S18), the process returns to step S10 and repeats the process from step S10. As a result, the CCD camera 8 photographs a reflected image from the corneal surface at a predetermined time interval from the start of photographing until the set time has elapsed, and the sum of the absolute values ​​of the blur amounts and the eyelid opening length of the subject's eye C are calculated for the photographed images, and a display image (pseudo fluorescein image) is created.

[0041] If the answer is YES in step S18, the controller 12 determines the change pattern of the amount of blur (S20). In step S20, the controller 12 determines the change pattern over time of the sum of the absolute values ​​of the amount of blur of the captured images (hereinafter, simply referred to as the "amount of blur"). There are multiple types of dry eye, which can be classified based on the change over time of the breakdown of corneal staining. For example, as shown in FIG. 10, dry eye can be classified into seven types according to the breakup pattern of corneal staining. Dry eye can also be classified into five types according to subtypes. The state of dry eye differs depending on the breakup pattern of corneal staining and subtype, and the treatment method (for example, the type of drug administered during treatment) also differs accordingly. It may be difficult to identify the type of dry eye based only on the breakup time of corneal staining. Therefore, in step S20, the change pattern over time of the amount of blur after opening the eyelids is determined from the amount of blur of the multiple images. By presenting the determined change pattern of the amount of shaking, the examiner can easily diagnose the type of dry eye of the examinee's eye.

[0042] As shown in FIG. 11, in this embodiment, the change pattern of the amount of blur is classified into five types, patterns a to e. Pattern a is a pattern in which the amount of blur hardly changes. Pattern a can be said to be a normal type (non-dry eye type). Pattern b is a pattern in which the amount of blur is small immediately after the eyelids are opened, and gradually increases. Pattern b can be said to be a standard break-up type. Pattern c is a pattern in which the amount of blur is large immediately after the eyelids are opened, and the amount of blur further increases with time. When the amount of blur changes in this way, there is a possibility that SPK (Superficial punctate keratopathy) has occurred in the subject's eye C. That is, pattern c can be said to be an SPK type. Pattern d is a pattern in which the amount of blur is large immediately after the eyelids are opened, and gradually decreases. Pattern d can be said to be an early break-up type. Pattern e is a pattern in which the amount of blur is small for a while after the eyelids are opened, but increases after a while. Pattern e can be said to be a late break-up type.

[0043] For each of the corneal staining break time and subtypes shown in Fig. 10, the change pattern of the amount of blur corresponds to one of the above-mentioned five patterns a to e. For example, Fig. 10 shows which of the five patterns a to e each of the corneal staining break time and subtypes corresponds to. By presenting the change pattern of the amount of blur of the subject's eye C, the examiner can identify or narrow down the type of dry eye of the subject's eye C.

[0044] The process of determining the change pattern of the amount of blur in step S20 is executed in the following procedure. As shown in Fig. 12, first, the controller 12 identifies the eyelid-opening start point (S300). The eyelid-opening start point is the point at which the subject's eye C starts to open its eyelid. While the eyelid-opening length of the subject's eye C calculated in step S14 is "0", it can be determined that the eyelid is in a closed state. Therefore, the controller 12 identifies the last point at which the eyelid-opening length becomes "0" as the eyelid-opening start point.

[0045] Next, the controller 12 calculates the amount of change in the amount of blur (hereinafter also referred to as the gradient value of the amount of blur) from the eyelid opening start point identified in step S300 until a predetermined time (hereinafter also referred to as the first predetermined time) has elapsed (S310). The first predetermined time is a time that indicates the period of early eyelid opening. The length of the first predetermined time is not particularly limited, but is, for example, 2 seconds.

[0046] Next, the controller 12 judges whether the gradient value of the amount of blur calculated in step S310 is negative (S320). That is, the controller 12 judges whether the amount of blur when a first predetermined time has passed since the eyelid opening start point is smaller than the amount of blur at the eyelid opening start point. As shown in FIG. 11, of the five types of patterns a to e, only pattern d has a small amount of blur at the beginning of eyelid opening. If the gradient value of the amount of blur is negative (YES in step S320), the controller 12 determines that the change pattern of the amount of blur is pattern d (S330).

[0047] On the other hand, if the slope value of the amount of blur is not negative (NO in step S320), controller 12 calculates an offset amount (S340). The offset amount is the amount by which a graph of the amount of blur (described later) is offset when the graph is displayed on display device 9. For example, controller 12 calculates the offset amount so that the amount of blur at the eyelid opening start point is "0". If the amount of blur is large immediately after the eyelid opens and continues to increase thereafter (for example, pattern c in FIG. 11), the graph of the amount of blur may tend to stay at high values ​​overall, making it difficult to recognize the change in the amount of blur after the eyelid opens. In such a case, offsetting the entire graph of the amount of blur downward makes it easier to grasp the change in the amount of blur after the eyelid opens.

[0048] Next, the controller 350 calculates the time when the tear film breaks down (so-called Breakup time) (S350). For example, the controller 12 can determine the time when the amount of blur at the eyelid opening start point becomes greater than a predetermined value (hereinafter, also referred to as a first predetermined value) as the Breakup time.

[0049] Next, the controller 12 judges whether the breakup time calculated in step S350 exceeds a predetermined time (hereinafter, also referred to as the second predetermined time) (S360). The second predetermined time is a threshold (length of the breakup time) for judging whether or not it is dry eye. The length of the second predetermined time is, for example, 5 seconds. If the breakup time exceeds the second predetermined time (YES in step S360), the amount of blur is small even if the second predetermined time has passed since the start of eyelid opening. Therefore, the controller 12 determines that the change pattern of the amount of blur is pattern a (S370).

[0050] On the other hand, if the Breakup time is equal to or less than the second predetermined time (NO in step S360), the controller 12 judges whether or not the amount of blur at the eyelid opening start point is equal to or more than a predetermined value (hereinafter also referred to as the second predetermined value) (S380). The second predetermined value is the amount of blur when it is judged that the amount of blur is large immediately after the eyelid opening. If the amount of blur at the eyelid opening start point is equal to or more than the second predetermined value (YES in step S380), the controller 12 determines that the change pattern of the amount of blur is pattern c (S390). Note that, among the five types of patterns a to e, patterns c and d are patterns in which the amount of blur is large immediately after the eyelid opening. The process of step S380 is executed when NO in step S320. That is, the process of step S380 is not executed when YES is obtained in step S320 and pattern d is determined in step S330. Therefore, when YES is determined in step S380, the change pattern of the amount of blur can be determined to be pattern c.

[0051] On the other hand, if the amount of blur at the eyelid opening start point is less than the second predetermined value (NO in step S380), the controller 12 judges whether the gradient value of the amount of blur calculated in step S310 is smaller than a predetermined value (hereinafter, also referred to as the third predetermined value) (S400). The third predetermined value is a value at which it is judged that the amount of blur increases from the eyelid opening start point until the first predetermined time has elapsed. If the gradient value of the amount of blur is smaller than the third predetermined value (YES in step S400), it can be judged that the amount of blur has hardly changed during the first predetermined time from the eyelid opening start point. For this reason, the controller 12 determines that the change pattern of the amount of blur is pattern e (S410). Among the five types of patterns a to e, there are patterns a and e as patterns in which the amount of blur hardly changes at the beginning of eyelid opening. The process of step S400 is executed when NO is found in step S360. That is, the process of step S400 is not executed if the answer is YES in step S360 and pattern a is determined in step S370. Therefore, if the answer is YES in step S400, the shake amount change pattern can be determined to be pattern e.

[0052] On the other hand, if the gradient value of the amount of blur is equal to or greater than the third predetermined value (NO in step S400), it can be determined that the amount of blur has increased from the eyelid opening start point until the first predetermined time has elapsed. There are two patterns in which the amount of blur increases at the beginning of eyelid opening: pattern b and pattern c. The process of step S400 is executed when NO is determined in step S380. That is, the process of step S400 is not executed when YES is determined in step S380 and pattern c is determined in step S390. Therefore, when NO is determined in step S400, the controller 12 determines that the change pattern of the amount of blur is pattern b (S420). By executing the above process, it is determined which of the five patterns a to e the change pattern of the amount of blur corresponds to. Thereafter, the process proceeds to step S22 in FIG. 3.

[0053] When the process of determining the change pattern of the amount of blur in step S20 is completed, the controller 12 determines whether the change pattern of the amount of blur determined in step S20 is pattern c (S22). As described above, pattern c is the SPK type. In the case of the SPK type, if it is possible to grasp whether the SPK occurs in the whole or part of the eye C to be examined, it becomes easier to determine a treatment method. There is a high possibility that SPK has occurred in a part where the amount of blur is large. Therefore, if the change pattern of the amount of blur is pattern c (YES in step S22), the controller 12 generates a map (hereinafter also referred to as an SPK map) indicating an area where the change in the amount of blur is large and where the occurrence of SPK is estimated (S24). On the other hand, if the change pattern of the amount of blur is not pattern c (NO in step S22), the process of step S24 is skipped and the process proceeds to step S26.

[0054] The process of generating the SPK map in step S24 is executed in the following procedure. As shown in FIG. 13, first, the controller 12 divides the subject's eye C into predetermined regions for the image at the eyelid opening start point (S500). Specifically, as shown in FIG. 14(a), the controller 12 specifies the center 20 of the ring image and the range 22 of the iris of the subject's eye C for the image at the eyelid opening start point. Then, as shown in FIG. 14(b), the controller 12 draws a horizontal line 24 above the center 20 of the ring image and draws a horizontal line 24 below the center 20 of the ring image. As a result, the range 22 of the iris of the subject's eye C is divided into three parts: an upper part, a central part (a part including the center 20 of the ring image), and a lower part. The number of regions into which the iris of the subject's eye C is divided is not particularly limited. For example, the controller 12 may divide the area 22 of the iris of the test eye C into four or more regions by drawing three or more horizontal lines 24 in the area 22 of the iris of the test eye C. Furthermore, the areas dividing the iris of the test eye C may be of substantially equal size or may be of different sizes.

[0055] Next, the controller 12 calculates the total amount of blur for each area divided in step S500 (i.e., each area of ​​the image at the eyelid opening start point) (S510). Specifically, the controller 12 selects one of the areas divided in step S500 and adds up the amount of blur for each pixel in the selected area. The controller 12 also adds up the amount of blur for each pixel in the other areas divided in step S500.

[0056] Next, the controller 12 divides the subject's eye C into predetermined regions for the image captured next to the image processed immediately before (S520). In the process immediately after starting the process of generating the SPK map in step S24, the image at the eyelid opening start point is divided into regions, and the total amount of blurring for each region is calculated. As described above, the controller 12 divides the iris of the subject's eye C into predetermined regions for the image captured next to the image at the eyelid opening start point. For this reason, in S520, the controller 12 divides the subject's eye C into predetermined regions in the same manner as when the iris of the subject's eye C was divided in step S500. For example, if the subject's eye C is divided into three regions as shown in FIG. 14(b) in step S500, the subject's eye C is also divided into three regions as shown in FIG. 14(b) in step S520. Next, the controller 12 calculates the total amount of blurring for each region divided in step S520 (S530).

[0057] Next, the controller 12 divides the subject's eye C into predetermined regions for the image taken a predetermined time after the eyelid opening start point (hereinafter also referred to as the third predetermined time), and determines whether or not the total amount of blur for each region has been calculated (S540). The third predetermined time is a time that indicates the period of early eyelid opening. The length of the third predetermined time is not particularly limited, but is, for example, 1 second. If the division of regions and the calculation of the total amount of blur for each region have not been performed for the image taken after the third predetermined time after the eyelid opening start point (NO in step S540), the process returns to step S520 and repeats the processes from step S520 onwards.

[0058] After dividing the image into regions and calculating the total amount of blur for each region after the third predetermined time from the eyelid opening start point (YES in step S540), the controller 12 selects one of the divided regions (S550) and determines whether the total amount of blur for the selected region exceeds a predetermined value (hereinafter, also referred to as a fourth predetermined value) during the period from the eyelid opening start point to the third predetermined time (S560). That is, the controller 12 determines whether the total amount of blur for the selected region exceeds the fourth predetermined value for all images for which the process of calculating the total amount of blur for each region has been executed in steps S510 and S530. The fourth predetermined value sets the total amount of blur estimated to have occurred in the divided regions. Note that, when the number of pixels in the divided regions is different, the fourth predetermined value may be set to the value of the amount of blur per unit pixel estimated to have occurred in the divided regions, and the controller 12 may determine whether the amount of blur per unit pixel in the selected region exceeds the fourth predetermined value.

[0059] If the total amount of shaking in the selected area exceeds a fourth predetermined value (YES in step S540), the controller 12 determines that the selected area is an area where SPK is estimated to have occurred (hereinafter also referred to as an SPK area) (S570). On the other hand, if the total amount of shaking in the selected area is equal to or less than the fourth predetermined value (NO in step S540), the controller 12 determines that the selected area is an area where SPK is estimated not to have occurred (hereinafter also referred to as a non-SPK area) (S580).

[0060] Next, the controller 12 judges whether or not it has been determined whether or not all the divided areas are SPK areas (S590). If there is an area for which it has not been determined whether or not it is an SPK area (NO in step S590), the controller 12 returns to step S550 and repeats the processes from step S550 onward.

[0061] When it is determined whether or not all the regions are SPK regions (YES in step S590), the controller 12 generates an SPK map based on the determinations in steps S570 and S580 (S600). Specifically, as shown in FIG. 15, the controller 12 superimposes a display 26 indicating an SPK region and a display 28 indicating a non-SPK region on the eye image from which the ring image has been removed. In the SPK map, the SPK region and the non-SPK region are displayed in different modes. For example, the SPK region and the non-SPK region are displayed in different colors. Then, the process proceeds to step S26 in FIG. 3.

[0062] When the process of generating the SPK map in step S24 is completed, the controller 12 displays on the display device 9 the image generated in step S16 (i.e., the pseudo-fluorescein image), a graph showing the change in the amount of blur over time (hereinafter, simply referred to as the "graph of the amount of blur"), and the determination result of the change pattern of the amount of blur determined in step S20 (S26). The graph of the amount of blur, the pseudo-fluorescein image, and the determination result of the change pattern of the amount of blur are displayed on one screen, thereby providing information useful for diagnosing dry eye. Furthermore, when the process of step S24 is executed (i.e., when the change pattern of the amount of blur is pattern c), the controller 12 further displays the image generated in step S24 (i.e., the SPK map). The SPK map may be displayed on one screen together with the graph of the amount of blur, the pseudo-fluorescein image, and the determination result of the change pattern of the amount of blur. The examiner may also operate the input device 13 to display the SPK map. Furthermore, the display device 9 is capable of displaying various eye images. For example, the controller 12 can display the modulated portion extraction image generated in step S200 by the examiner operating the input device 13.

[0063] FIG. 16 shows an example of a display screen. FIG. 16 shows a case where the change pattern of the amount of blur is pattern c. As shown in FIG. 16, the pseudo-fluorescein image 30, the SPK map 32, the table 34 showing the Breakup time, the graph 36 of the amount of blur, and the judgment result 40 of the change pattern of the amount of blur are displayed on the screen of the display device 9. The pseudo-fluorescein image 30, the SPK map 32, and the judgment result 40 of the change pattern of the amount of blur are displayed side by side in the left-right direction. The pseudo-fluorescein image 30 and the graph 36 are arranged side by side in the up-down direction, and the table 34 is arranged to the side of the pseudo-fluorescein image 30. In addition, a seek bar 38 is displayed between the pseudo-fluorescein image 30 and the graph 36. The examiner operates the input device 13 to move the slider of the seek bar 38 in the horizontal direction, and the pseudo-fluorescein image 30 of the image taken at the time specified by the slider is displayed on the display device 9. A line extending downward from the slider is displayed as a vertical line indicating time on the graph 36. A table 34, a blur amount graph 36, and a seek bar 38 are also arranged near the SPK map 32 in the same manner as the pseudo-fluorescein image 30.

[0064] FIG. 16 shows an example where the change pattern of the amount of blur is pattern c. The controller 12 displays the determination result 40 of the change pattern of the amount of blur on the display device 9 so as to inform the examiner that the change pattern of the amount of blur of the subject's eye C is pattern c. Specifically, the controller 12 displays on the display device 9 schematic diagrams of five types of change patterns, and also displays a mark 42 (a rectangle 42 surrounding the schematic diagram of pattern c in FIG. 16) indicating that the determination result is pattern c. For example, pattern a, in which the amount of blur remains small and does not change, and pattern b, in which the amount of blur is small immediately after opening the eyelids and gradually increases, are generally well known, while the other patterns c to e may not be recognized even by the examiner. By displaying all the schematic diagrams of the multiple change patterns, it is possible to inform the examiner that there are multiple change patterns that can be determined. In addition, by displaying the mark 42 together, it is possible to inform the examiner which pattern the amount of blur change is, even if all the schematic diagrams of the multiple change patterns are displayed.

[0065] In this embodiment, the display device 9 displays a graph 36 of the amount of blur together with the result of the determination of the change pattern of the amount of blur. It may be difficult for the examiner to diagnose the type of dry eye of the subject's eye C only from the graph 36 of the amount of blur. By displaying the result of the determination of the change pattern of the amount of blur determined by the controller 12 together with the graph of the amount of blur, the examiner can easily determine how the amount of blur is changing, and can easily diagnose the type of dry eye. Also, as described above, the photographing time of the pseudo-fluorescein image 30 displayed on the display device 9 can be changed by moving the slider of the seek bar 38 in the horizontal direction. As a result, the examiner can observe how the modulation part of the tear film has changed. For example, the examiner can easily grasp the manner in which adjacent modulation parts of the tear film are combined, the manner in which a single modulation part appears, and the spread and speed of the modulation part of the tear film, and thereby the type of dry eye can be determined.

[0066] In this embodiment, the display device 9 displays all the schematic diagrams of the multiple change patterns, and also displays a mark 42 indicating the judgment result on the schematic diagram of the corresponding pattern (schematic diagram of pattern c in FIG. 16), but is not limited to such a configuration. The schematic diagram of the corresponding pattern may be displayed in a different manner from the schematic diagrams of other patterns, and the display manner is not particularly limited. For example, as shown in FIG. 17(a), the periphery of the schematic diagram of the corresponding pattern (schematic diagram of pattern a in FIG. 17(a)) may be displayed in a different color. Also, the periphery of the schematic diagram of the corresponding pattern may be blinked. Also, as shown in FIG. 17(b), the schematic diagram of the corresponding pattern may be displayed with high brightness, and the schematic diagrams of the other patterns may be displayed with low brightness. Also, it is not necessary to display all the schematic diagrams of the multiple change patterns, and only the schematic diagram of the corresponding pattern may be displayed as shown in FIG. 17(c).

[0067] In this embodiment, the change patterns of the amount of shake are classified into five types, but the present invention is not limited to this configuration. The change patterns of the amount of shake may be classified into fewer than five types. For example, the change patterns may be classified into four types, patterns a to d. Even in such a case, by determining the change pattern of the amount of shake, the examiner can more easily determine how the amount of shake is changing and more easily diagnose the type of dry eye, compared to the case where dry eye is diagnosed only from the graph 36 of the amount of shake.

[0068] Moreover, when the offset value is calculated in the above-mentioned S340, the controller 12 displays the blur amount graph 36 by offsetting it downward by the calculated offset amount. As described above, the offset value is calculated so that the blur amount at the eyelid opening start point is "0". Therefore, when the blur amount graph 36 is offset downward by the offset value, the blur amount graph 36 is displayed so that the blur amount at the eyelid opening start point is "0". This makes it easier to grasp the change in the blur amount after the eyelid is opened. FIG. 18(a) shows the blur amount graph before offset of pattern c, and FIG. 18(b) shows the blur amount graph 36 offset downward by the offset value from the position shown in FIG. 18(a). In FIG. 18(a) and FIG. 18(b), the eyelid opening length 50 and the above-mentioned first predetermined value 52 (threshold value for the breakup time) are displayed together with the blur amount graph 36. In FIG. 18(a), the blur amount is large when the eyelid is opened, so the blur amount exceeds the first predetermined value immediately after the eyelid is opened. For this reason, it is difficult to determine whether the amount of blur when the eyelids are open is greater than the first predetermined value 52, or whether the amount of blur has not changed enough to exceed the first predetermined value, and it is difficult to grasp the change in the amount of blur after the eyelids are opened. On the other hand, in Fig. 18(b), the graph 36 is offset so that the amount of blur at the eyelid opening start point is "0", making it easier to grasp the amount of change in the amount of blur relative to the amount of blur when the eyelids are open. This allows the examiner to properly diagnose the type of dry eye.

[0069] If the answer is YES in step S320 (if the slope value of the amount of blur is negative), the pattern of change in the amount of blur is determined to be pattern d, and no offset value is calculated. For this reason, when the pattern of change in the amount of blur is pattern d, graph 36 of the amount of blur is not offset. In pattern d, the amount of blur becomes smaller after the eyelids are opened, so if it is offset downward, it becomes difficult to grasp the amount of blur. In this embodiment, no offset is performed in the case of pattern d, so the change in the amount of blur can be properly grasped even in the case of pattern d.

[0070] In this embodiment, when the offset value is calculated in the above-mentioned S340, the graph 36 of the amount of blur is offset downward by the calculated offset value, but the present invention is not limited to this configuration. For example, as shown in FIG. 19(b), when the offset value is calculated in the above-mentioned S340, the controller 12 may offset the first predetermined value 52 upward by the offset value without offsetting the graph 36 of the amount of blur. In this case, too, the examiner can accurately grasp from the displayed graph 36 whether the amount of blur is larger than the first predetermined value 52 with respect to the amount of blur when the eyelids are open, or whether the amount of blur has not changed to the extent that it exceeds the first predetermined value. This makes it easier for the examiner to grasp the amount of change in the amount of blur with respect to the amount of blur when the eyelids are open, and thus allows the type of dry eye to be appropriately diagnosed.

[0071] Furthermore, when the change pattern of the amount of blur is pattern c, the display device 9 displays the SPK map 32. When SPK occurs in the subject's eye C, the diagnosis result of the type of dry eye may differ depending on whether the SPK occurs in the entire subject's eye C or in a part of the subject's eye C. Furthermore, the diagnosis result of the type of dry eye may differ depending on the position (area) where the SPK occurs in the subject's eye C. By displaying the SPK map 32 when the change pattern of the amount of blur is pattern c, the examiner can easily grasp which area of ​​the subject's eye C is the SPK area, and can more appropriately diagnose the type of dry eye.

[0072] For example, as shown in FIG. 10, when the change pattern of the amount of blur is pattern c, the break pattern of the corneal staining can be narrowed down to any one of area break, partial area break, and line break. In the examples of FIG. 15 and FIG. 16, since the center part is an SPK area, the break pattern of the corneal staining can be estimated to be a partial area break. Furthermore, when the lower part is an SPK area, the break pattern of the corneal staining can be estimated to be a line break. Furthermore, when the upper part, the center part, and the lower part are all SPK areas, the break pattern of the corneal staining can be estimated to be an area break. In this way, by displaying the SPK map, the examiner can further narrow down the break pattern of the corneal staining from the above three, and can more appropriately diagnose the type of dry eye. In addition, the controller 12 may further display on the display device 9 the determination result as to whether the corneal staining breakup pattern can be estimated to be an area break, a partial area break, or a line break, depending on whether the upper, central, or lower part is the SPK region.

[0073] In this embodiment, when generating the SPK map, the iris of the subject's eye C is divided into three parts, an upper part, a central part, and a lower part, but the present invention is not limited to this configuration. For example, as shown in FIG. 20, the iris of the subject's eye C may be divided in the circumferential direction from its center 20 so as to be approximately fan-shaped. FIG. 20 shows an example in which the iris of the subject's eye C is divided into 16 parts. The number of regions into which the iris of the subject's eye C is divided is not particularly limited. For example, the iris of the subject's eye C may be divided into less than 16 parts (for example, 4 parts or 8 parts), or may be divided into more than 16 parts. In addition, the iris of the subject's eye C may be divided into approximately equal sizes, or may be divided into different sizes. In this case, by displaying the SPK map, the examiner can easily grasp which region of the subject's eye C is the SPK region, and the type of dry eye can be more appropriately diagnosed.

[0074] Although specific examples of the technology disclosed in this specification have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and variations of the specific examples exemplified above. Furthermore, the technical elements described in this specification or drawings exhibit technical utility alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Furthermore, the technology exemplified in this specification or drawings achieves multiple objectives simultaneously, and achieving one of the objectives itself has technical utility. [Explanation of symbols]

[0075] 1: Corn 2: Lighting equipment 3: Lens 4: Half mirror 6: Fixation light source 8: CCD camera 10: Memory 12: Controller

Claims

1. a projection unit that projects a pattern light onto a corneal surface of the subject's eye; an imaging unit that captures a reflected image of the pattern light projected by the projection unit from the corneal surface; a control unit that causes the imaging unit to repeatedly capture a reflected image from the corneal surface at a predetermined time interval; a calculation unit that calculates, for each captured image captured by the imaging unit, a sum of absolute values ​​of blur amounts of lines of a pattern image of the captured image caused by the pattern light; a graph generating unit that generates a graph showing a time series change in the sum of the absolute values ​​of the amounts of shaking calculated by the calculation unit; a determination unit that determines which of a plurality of preset change patterns the graph generated by the graph generation unit corresponds to; An ophthalmic apparatus comprising:

2. The ophthalmologic apparatus according to claim 1 , further comprising a display unit that displays the change pattern of the graph determined by the determination unit.

3. The ophthalmologic apparatus according to claim 2 , wherein the display unit displays a plurality of schematic diagrams each showing one of the plurality of change patterns, and displays the change pattern of the graph determined by the determination unit in a manner different from other change patterns.

4. The ophthalmologic apparatus according to claim 2 , wherein the display unit displays the graph generated by the graph generation unit and the change pattern of the graph determined by the determination unit on the same screen.

5. The eyelid opening detection unit detects whether the eyelid of the subject is opened, the display unit displays a threshold value, the sum of the absolute values ​​of the amounts of shaking of which indicates a breakdown of the tear film of the subject eye, superimposed on the graph; 3. The ophthalmic apparatus according to claim 2, wherein the graph generating unit corrects the threshold value or the sum of the absolute values ​​of the amounts of blur calculated for each of the images captured by the photographing unit immediately after the eyelid opening detection unit detects the opening of the eyelid of the test eye, so that the sum of the absolute values ​​of the amounts of blur calculated for the images captured by the photographing unit immediately after the eyelid opening detection unit detects the opening of the eyelid of the test eye becomes a predetermined value.

Images