Visual field examination device and visual field examination program

Abstract

In the present invention, a probability density function utilizing a sensitivity threshold value as a random variable is set in advance to each measurement point. An initial intensity as a projection intensity of a stimulus target to be initially projected onto the measurement point is set. The stimulus target is projected onto the measurement point at the set projection intensity, and the result of an examinee's response to the stimulus target is acquired. The probability density function is changed in accordance with the response result, and a sensitivity threshold value at the measurement point is acquired on the basis of the changed probability density function. The probability density function to be set in advance is set on the basis of OCT data on an ocular fundus being examined.

Description

Visual field testing device and visual field testing program

[0001] The present disclosure relates to a visual field testing device and a visual field testing program for testing the visual field of a subject's eye.

[0002] A visual field testing device (sometimes called a "perimeter") is a device for subjectively testing the visual field of a subject's eye. In a visual field testing device, the visual field at each measurement point on the fundus is tested based on whether the subject can visually recognize a test target projected onto the fundus.

[0003] Various technologies have been proposed to improve the accuracy and time required for visual field testing. For example, in the device described in Patent Document 1, layer thickness information from a fundus tomographic image is compared with a normal eye database, and a presentation area for a stimulus target is set based on position information for areas where the layer thickness information falls outside the normal eye range. In addition, in the device described in Patent Document 2, the initial brightness of the stimulus target is set based on the position of a lost or atrophied area in a retinal layer related to the visual function of the fundus of the subject eye.

[0004] JP 2012-100713 A JP 2014-188254 A

[0005] In order to perform a visual field test in a shorter time and with higher accuracy, the inventors of the present application conducted various studies and repeated trial and error, and discovered a new algorithm for progressing the visual field test that had not existed before.

[0006] A typical object of the present disclosure relates to a visual field testing device and a visual field testing program that are capable of performing a visual field test in a shorter time and with higher accuracy.

[0007] A first aspect of a visual field testing device provided by a typical embodiment of the present disclosure is a visual field testing device that projects a stimulus target onto each of a plurality of measurement points on the fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target, wherein a control unit of the visual field testing device includes: a prior probability density function setting step that sets in advance, for each of the measurement points, a probability density function in which the sensitivity threshold is a random variable; an initial intensity setting step that sets an initial intensity that is the projection intensity of the stimulus target to be initially projected onto the measurement point; a response result acquisition step that projects the stimulus target onto the measurement point at the set projection intensity and acquires a response result of the subject to the stimulus target; and, if the response result acquired in the response result acquisition step is visible, indicating that the stimulus target was visible, subtracts a probability density in the probability density function that is less than the projection intensity, while, if the response result is invisible, subtracts a probability density in the probability density function that is greater than or equal to the projection intensity. a function changing step of changing the probability density function by reducing the probability density of the OCT light reflected from the fundus of the subject's eye; a determination step of determining whether or not an end condition for measurement at the measurement point is met; a repeating step of repeating the response result acquiring step, the function changing step, and the determination step if it is determined in the determination step that the end condition is not met, after newly setting a projection intensity according to the previous response result; a sensitivity threshold acquiring step of acquiring a sensitivity threshold at the measurement point based on the changed probability density function if it is determined in the determination step that the end condition is met; and an OCT data acquiring step of acquiring OCT data obtained from interference light between reflected light of OCT light from the fundus of the subject's eye and reference light corresponding to the OCT light, and in the prior probability density function setting step, the probability density function is set in advance based on the OCT data for the fundus of the subject's eye acquired in the OCT data acquiring step.

[0008] A second aspect of a visual field testing device provided by a typical embodiment of the present disclosure is a visual field testing device that projects a stimulus target onto each of a plurality of measurement points on the fundus of a subject's test eye and acquires a sensitivity threshold at each of the measurement points based on the subject's response to the stimulus target, wherein a control unit of the visual field testing device executes an angio data acquisition step of acquiring OCT angio data, which is motion contrast data generated by arithmetic processing of at least two OCT signals acquired at different times for the same position on the fundus of the test eye, and an initial intensity setting step of setting an initial intensity, which is the projection intensity of the stimulus target to be initially projected onto each of the measurement points, based on the OCT angio data.

[0009] A first aspect of a visual field testing program provided by a typical embodiment of the present disclosure is a visual field testing program executed by a visual field testing control device for controlling testing by a visual field testing device, the visual field testing device being an apparatus that projects a stimulus target onto each of a plurality of measurement points on the fundus of an eye to be tested of a subject, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target, and the visual field testing program is executed by a control unit of the visual field testing control device, thereby performing the following steps: a prior probability density function setting step of setting in advance a probability density function with the sensitivity threshold as a random variable for each of the measurement points; an initial intensity setting step of setting an initial intensity which is the projection intensity of the stimulus target to be initially projected onto the measurement point; a response result acquisition step of projecting the stimulus target onto the measurement point at the set projection intensity and acquiring a response result of the subject to the stimulus target; and, when the response result acquired in the response result acquisition step is visible indicating that the stimulus target was visible, subtracting a probability density in the probability density function that is less than the projection intensity, while and a function changing step of changing the probability density function by subtracting a probability density equal to or greater than the projection intensity in the probability density function if the response result is not visible; a determination step of determining whether or not the measurement termination condition for the measurement point is met; a repeating step of repeating the response result acquisition step, the function changing step, and the determination step if it is determined in the determination step that the termination condition is not met, after newly setting the projection intensity according to the previous response result; a sensitivity threshold acquisition step of acquiring a sensitivity threshold at the measurement point based on the changed probability density function if it is determined in the determination step that the termination condition is met; and an OCT data acquisition step of acquiring OCT data obtained from the interference light between the reflected light of OCT light from the fundus of the test eye and the reference light corresponding to the OCT light, and in the prior probability density function setting step, the probability density function is set in advance based on the OCT data for the fundus of the test eye acquired in the OCT data acquisition step.

[0010] A second aspect of a visual field testing program provided by a typical embodiment of the present disclosure is a visual field testing program executed by a visual field testing control device for controlling testing by a visual field testing device, wherein the visual field testing device projects a stimulus target onto each of a plurality of measurement points on the fundus of a subject's test eye and acquires a sensitivity threshold at each of the measurement points based on the subject's response to the stimulus target, and the visual field testing program is executed by the control unit of the visual field testing control device, causing the visual field testing control device to execute an angio data acquisition step of acquiring OCT angio data, which is motion contrast data generated by arithmetic processing of at least two OCT signals acquired at different times for the same position on the fundus of the test eye, and an initial intensity setting step of setting an initial intensity, which is the projection intensity of the stimulus target to be initially projected onto each of the measurement points, based on the OCT angio data.

[0011] According to the visual field testing device and visual field testing program of the present disclosure, visual field testing can be performed in a shorter time and with higher accuracy.

[0012] In a first aspect of the visual field testing device exemplified in the present disclosure, the control unit executes a prior probability density function setting step, an initial intensity setting step, a response result acquisition step, a function changing step, a determination step, a repeating step, a sensitivity threshold acquisition step, and an OCT data acquisition step. In the prior probability density function setting step, a probability density function is set in advance for each measurement point, with the sensitivity threshold as a random variable. Hereinafter, the probability density function set before a stimulus target is projected onto the measurement point may also be referred to as a prior probability density function. In the initial intensity setting step, an initial intensity is set, which is the projection intensity (e.g., contrast intensity) of the stimulus target initially projected onto the measurement point. In the response result acquisition step, the stimulus target is projected onto the measurement point at the set projection intensity, and the subject's response result to the stimulus target is acquired. In the function changing step, if the response result acquired in the response result acquisition step is visible, indicating that the stimulus target was visible, the probability density less than the projection intensity in the probability density function is reduced. Furthermore, if the response result is invisible, the probability density greater than the projection intensity in the probability density function is reduced. In the determination step, it is determined whether an end condition for measurement at the measurement point is satisfied. In the repeating step, if it is determined that the end condition is not satisfied in the determination step, a new projection intensity is set according to the previous response result, and the response result acquiring step, function changing step, and determination step are repeated. In the sensitivity threshold acquiring step, if it is determined that the end condition is satisfied in the determination step, a sensitivity threshold at the measurement point is acquired based on the changed probability density function. In the OCT data acquiring step, OCT data obtained from interference light between reflected light of OCT light from the fundus of the test eye and reference light corresponding to the OCT light is acquired. In the prior probability density function setting step, a probability density function is set in advance based on the OCT data of the fundus of the test eye acquired in the OCT data acquiring step.

[0013] According to the technology disclosed herein, by modifying the probability density function based on the subject's response to the stimulus target and obtaining the sensitivity threshold at the measurement point based on the modified probability density function, it becomes easier to obtain the sensitivity threshold with fewer target projections than conventional methods such as the full-threshold method. Note that the full-threshold method is a method for searching for the sensitivity threshold by increasing or decreasing the projection threshold by a first value each time the same response result (visible or invisible) is repeated, and then decreasing or increasing the projection threshold by a second value that is smaller than the first value and has the opposite sign to the first value when the response result changes.

[0014] Furthermore, according to the technology disclosed herein, a prior probability density function before the projection of a stimulus target is set based on OCT data, which is likely to provide various information about the state of the fundus of the subject's eye (e.g., at least one of the state of the fundus layers and the state of vascular perfusion). Therefore, the sensitivity threshold at the measurement point is acquired based on the prior probability density function that reflects information about the state of the fundus of the subject's eye that affects visual function. This makes it easier to perform a visual field test in a shorter time with higher accuracy.

[0015] The perfusion state of blood vessels (e.g., capillary networks) in the fundus of the eye affects the visual function of the subject's eye. OCT angiographic data includes perfusion information of the vascular network in the retinal layer. Therefore, by setting a prior probability density function based on the OCT angiographic data, information on the perfusion state of retinal blood vessels, which affects visual function, is appropriately reflected in the prior probability density function. This makes it easier to perform visual field tests in a shorter time and with higher accuracy.

[0016] A specific method for setting a prior probability density function based on OCT angiographic data can be selected as appropriate. For example, the control unit may set a prior probability density function at each measurement point based on blood vessel density (e.g., blood vessel area or length per unit area) analyzed from the OCT angiographic data. Alternatively, the control unit may align a two-dimensional frontal observation image captured in real time during the examination of the fundus of the subject's eye to be examined in a visual field test with the OCT angiographic data (in this case, the frontal OCT angiographic image), and set a prior probability density function at each measurement point based on the OCT angiographic data value (e.g., blood vessel density value, etc.) at each measurement point on the fundus.

[0017] However, as will be described later, OCT data relating to the tomography of the fundus of the test eye may be used instead of or together with the OCT angiography data. In this case, the OCT data relating to the tomography may be, for example, the tomography image data itself, or data on the thickness distribution of a specific layer (sometimes referred to as a "thickness map") obtained by analyzing the tomography image. Furthermore, the OCT data relating to the tomography may be normal eye comparison data showing the difference between the thickness distribution of a specific layer in the test eye and the thickness distribution of a specific layer in a normal eye.

[0018] The retina of the fundus contains blood vessels in multiple regions at different depths. Therefore, by setting a prior probability density function based on OCT angiographic data in a specific region, information on a desired blood vessel (e.g., a blood vessel highly related to visual function) among the multiple blood vessels can be more easily reflected in the prior probability density function.

[0019] Previously, there have been studies focusing on the correlation between the thickness of the ganglion cell complex (GCC), which is composed of the nerve fiber layer (NFL), ganglion cell layer (GCL), and inner plexiform layer (IPL) among the layer structures of the fundus, and the visual function of the test eye. However, new findings by the present inventors have suggested that the perfusion state of blood vessels near the ganglion cells present in the GCC may be more closely related to the visual function of the test eye than the thickness of the GCC. Most of the blood vessels near the ganglion cells are present in the ganglion cell layer (GCL) and inner plexiform layer (IPL). Therefore, by including at least a portion of the ganglion cell layer (GCL) and inner plexiform layer (IPL) in the specific region, information on the perfusion state of blood vessels, which is likely to affect visual function, can be more appropriately reflected in the prior probability density function, making it easier to perform visual field testing in a shorter time and with higher accuracy.

[0020] The two specific boundary surfaces defining the specific region may be located between the superficial boundary of the ganglion cell layer (GCL) and the deep boundary of the inner plexiform layer (IPL). That is, the superficial (upper) boundary of the two specific boundaries may be set at a position below the superficial boundary of the ganglion cell layer, and the deep (lower) boundary may be set at a position above the deep boundary of the inner plexiform layer. In this case, the specific region does not include layers other than the ganglion cell layer and the inner plexiform layer. Therefore, information about blood vessels other than those near the ganglion cells is less likely to be reflected in the prior probability density function. Therefore, information about the perfusion status of blood vessels that are likely to affect visual function is more likely to be reflected in the prior probability density function more appropriately.

[0021] However, the specific region may be changed, for example, the specific region may include at least one of the internal limiting membrane (ILM) and the nerve fiber layer (NFL).

[0022] In the prior probability density function setting step, the prior probability density function of at least the measurement points located within the central region may be set based on the OCT angiographic data. The central region is a region including the fovea when the fundus of the test eye is viewed from the front (in the direction along the line of sight of the test eye). In the central region, the correlation between the vascular perfusion state and visual function tends to be stronger than in peripheral regions. Therefore, by setting the prior probability density function of at least the measurement points within the central region based on the OCT angiographic data, information on the vascular perfusion state is more likely to be reflected in the prior probability density function more appropriately.

[0023] In the prior probability density function setting step, the prior probability density function of at least the measurement points located within the peripheral region may be set based on OCT data relating to a tomographic layer of the fundus (e.g., a retinal layer) of the subject eye. The peripheral region is an annular region located outside (e.g., adjacent to) the central region so as to surround the central region. The peripheral region tends to have a stronger correlation between the tomographic state of the fundus (e.g., the thickness of a specific layer) and visual function than the central region. Therefore, by setting the prior probability density function of at least the measurement points within the peripheral region based on OCT data relating to the tomographic layer, the state of the fundus is more likely to be reflected in the prior probability density function.

[0024] As described above, the OCT data relating to the tomography of the fundus of the test eye may be the tomographic image data itself, or may be data on the thickness distribution of a specific layer obtained by analyzing the tomographic image. Furthermore, the OCT data relating to the tomography may be normal eye comparison data showing the difference between the thickness distribution of a specific layer in the test eye and the thickness distribution of a specific layer in a normal eye.

[0025] The specific method for setting the prior probability density functions in each of the central region and the peripheral region can be selected as appropriate. For example, the control unit may first set the prior probability density functions of all measurement points, regardless of whether they are in the central region or the peripheral region, based on OCT data related to the tomography. Then, the control unit may correct the prior probability density functions of the measurement points in the central region based on OCT angiographic data. In this case, the prior probability density functions in each of the central region and the peripheral region are appropriately set.

[0026] The control unit may set the projection intensity of the stimulus target projected onto the measurement point based on the currently set probability density function. For example, the control unit may set the initial intensity of the stimulus target projected onto the measurement point based on a prior probability density function. In this case, both the initial intensity of the stimulus target projected onto the measurement point and the probability density function for obtaining the sensitivity threshold at the measurement point are appropriately set based on the OCT data. This makes it easier to perform visual field testing in a shorter time and with higher accuracy.

[0027] The specific method for setting the projection intensity of the stimulus target based on the probability density function can be selected as appropriate. For example, the control unit may set the expected value (average) calculated from the probability density function as the projection intensity. Alternatively, the control unit may set the sensitivity threshold at which the probability density function is maximized as the projection intensity.

[0028] A second aspect of the visual field testing device exemplified in the present disclosure projects a stimulus target at each of multiple measurement points on the fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each measurement point based on the subject's response to the stimulus target. The control unit of the visual field testing device executes an angio data acquisition step and an initial intensity acquisition step. In the angio data acquisition step, OCT angio data of the fundus of the subject's eye to be examined is acquired. The OCT angio data is motion contrast data generated by processing at least two OCT signals acquired at different times for the same position on the fundus of the subject's eye. In the initial intensity setting step, the initial intensity, which is the projection intensity (e.g., contrast intensity) of the stimulus target initially projected at each measurement point, is set based on the OCT angio data.

[0029] The perfusion state of blood vessels (e.g., capillary networks) in the fundus affects the visual function of the subject's eye. OCT angiographic data includes perfusion information of the vascular network in the retinal layer. Therefore, by setting the initial intensity of the stimulus target based on the OCT angiographic data, information on the perfusion state of blood vessels in the retinal layer, which affects visual function, is appropriately reflected in the initial intensity. In other words, it becomes easier to set an appropriate initial intensity according to the perfusion state of the blood vessels. This makes it easier to perform visual field tests in a shorter time and with higher accuracy.

[0030] A specific method for setting the initial intensity based on the OCT angiographic data can be selected as appropriate. For example, the control unit may set the initial intensity at each measurement point based on the blood vessel density (e.g., the area or length of blood vessels per unit area) analyzed from the OCT angiographic data. Alternatively, the control unit may align the OCT angiographic data (in this case, the OCT angiographic front image) with a two-dimensional frontal observation image captured in real time during the examination of the fundus of the subject's eye that is the subject of the visual field test, and set the initial intensity at each measurement point based on the OCT angiographic data value (e.g., a blood vessel density value, etc.) at each measurement point on the fundus.

[0031] The retina of the fundus contains blood vessels in multiple regions at different depths. Therefore, by setting the initial intensity based on OCT angiographic data in a specific region, information on a desired blood vessel (e.g., a blood vessel highly related to visual function) among the multiple blood vessels can be more easily reflected in the initial intensity.

[0032] Previously, there have been studies focusing on the correlation between the thickness of the ganglion cell complex (GCC), which is composed of the nerve fiber layer (NFL), ganglion cell layer (GCL), and inner plexiform layer (IPL) among the layer structures of the fundus, and the visual function of the test eye. However, new findings by the present inventors have suggested that the perfusion state of blood vessels near the ganglion cells present in the GCC may be more closely related to the visual function of the test eye than the thickness of the GCC. Most of the blood vessels near the ganglion cells are present in the ganglion cell layer (GCL) and inner plexiform layer (IPL). Therefore, by including at least a portion of the ganglion cell layer (GCL) and inner plexiform layer (IPL) in the specific region, information on the state of vascular perfusion, which is likely to affect visual function, can be more appropriately reflected in the initial intensity, making it easier to perform visual field testing in a shorter time and with higher accuracy.

[0033] The two specific boundary surfaces defining the specific region may be located between the boundary of the ganglion cell layer (GCL) on the superficial retinal side and the boundary of the inner plexiform layer (IPL) on the deep retinal side. That is, the boundary on the superficial retinal side (upper side) of the two specific boundaries may be set at a position below the ganglion cell layer, and the boundary on the deep retinal side (lower side) may be set at a position above the inner plexiform layer. In this case, the specific region does not include layers other than the ganglion cell layer and the inner plexiform layer. Therefore, information about blood vessels other than those near the ganglion cells is less likely to be reflected in the initial intensity. Therefore, information about the perfusion status of blood vessels that are likely to affect visual function is more likely to be reflected in the initial intensity more appropriately.

[0034] However, the specific region may be changed, for example, the specific region may include at least one of the internal limiting membrane (ILM) and the nerve fiber layer (NFL).

[0035] In the initial intensity setting step, the initial intensities of at least the measurement points located within the central region may be set based on OCT angiographic data. The central region is a region including the fovea when the fundus of the test eye is viewed from the front (in the direction along the line of sight of the test eye). The central region tends to have a stronger correlation between vascular perfusion and visual function than peripheral regions. Therefore, by setting the initial intensities of at least the measurement points within the central region based on OCT angiographic data, information about the vascular perfusion state is more likely to be reflected in the initial intensity more appropriately.

[0036] In the initial intensity setting step, the initial intensity of at least the measurement points located within the peripheral region may be set based on OCT data relating to a tomographic layer of the fundus (e.g., a retinal layer). The peripheral region is an annular region located outside (e.g., adjacent to) the central region so as to surround the central region. The peripheral region tends to have a stronger correlation between the tomographic state of the fundus (e.g., the thickness of a specific layer) and visual function than the central region. Therefore, by setting the initial intensity of at least the measurement points within the peripheral region based on the OCT data relating to the tomographic layer, the state of the fundus is more likely to be reflected in the initial intensity more appropriately.

[0037] The OCT data relating to the tomography of the fundus of the test eye may be the tomographic image data itself, or may be data on the thickness distribution of a specific layer obtained by analyzing the tomographic image. The OCT data relating to the tomography may also be normal eye comparison data showing the difference between the thickness distribution of a specific layer in the test eye and the thickness distribution of a specific layer in a normal eye.

[0038] The specific method for setting the initial intensity in each of the central region and the peripheral region can be selected as appropriate. For example, the control unit may first set the initial intensity of all measurement points, regardless of whether they are in the central region or the peripheral region, based on OCT data related to the tomography. Then, the control unit may correct the initial intensity of the measurement points in the central region based on OCT angiographic data. In this case, the initial intensity in each of the central region and the peripheral region is appropriately set.

[0039] The control unit includes a prior probability density function setting step of setting a probability density function in advance, for each of the measurement points, based on the OCT angiographic data, with a sensitivity threshold as a random variable; a response result acquisition step of projecting the stimulus target onto the measurement point with a set projection intensity and acquiring a response result of the subject to the stimulus target; and, if the response result acquired in the response result acquisition step is visible, indicating that the stimulus target was visible, subtracting a probability density of less than the projection intensity in the probability density function, while, if the response result is invisible, subtracting a probability density of equal to or greater than the projection intensity in the probability density function. The method may further include a function changing step of changing the probability density function by reducing the density, a determination step of determining whether an end condition of measurement for the measurement point is met, a repeating step of setting a new projection intensity according to a previous response result and repeating the response result obtaining step, the function changing step, and the determination step if it is determined in the determination step that the end condition is not met, and a sensitivity threshold obtaining step of obtaining a sensitivity threshold at the measurement point based on the changed probability density function if it is determined in the determination step that the end condition is met.

[0040] In this case, the initial intensity of the stimulus target projected onto the measurement point and the probability density function for obtaining the sensitivity threshold at the measurement point are both appropriately set based on the OCT angiographic data, making it easier to perform visual field testing in a shorter time and with higher accuracy.

[0041] The specific method for setting the projection intensity of the stimulus target based on the probability density function can be selected as appropriate. For example, the control unit may set the expected value (average) calculated from the probability density function as the projection intensity. Alternatively, the control unit may set the sensitivity threshold at which the probability density function is maximized as the projection intensity.

[0042] However, even when performing a visual field test using a method that does not use a probability density function (such as the full-threshold method), it is useful to set the initial intensity of the projection target according to the OCT angiographic data. The full-threshold method is a method in which the projection intensity is increased or decreased by a first value each time the same response result (visible or invisible) is repeated, and when the response result changes, the projection intensity is increased or decreased by a second value that is smaller than the first value and has the opposite sign to the first value, thereby searching for the sensitivity threshold.

[0043] A specific method for changing the probability density function in the function changing step can be selected as appropriate. For example, the control unit may change the probability density function by multiplying the probability density function by a likelihood function corresponding to the subject's response result (visible or invisible) and normalizing the result. In this case, the probability density function is appropriately changed by simple processing. Note that the likelihood function when the response result is visible may be set so that the probability density decreases as the sensitivity threshold becomes smaller than the projection intensity. Note that the likelihood function when the response result is invisible may be set so that the probability density decreases as the sensitivity threshold becomes larger than the projection intensity. Furthermore, the intersection of the likelihood function when the response result is visible and the likelihood function when the response result is invisible may be set to be the projection intensity.

[0044] However, instead of multiplying the probability density function by a likelihood function, the probability density function may be changed by simply subtracting the probability density corresponding to the response result (a probability density less than the projection intensity if the response result is visible, or a probability density greater than or equal to the projection intensity if the response result is invisible).

[0045] The measurement termination condition used in the determination step can also be selected as appropriate. For example, the control unit may set the termination condition for the measurement for a measurement point to be that the modified probability density function falls within a predetermined standard deviation. Alternatively, the termination condition for the measurement for a measurement point may be that the measurement cycle of projecting a stimulus target, obtaining a response result, and modifying the probability density function performed for one measurement point reaches a predetermined number of times. Alternatively, the termination condition may be that either the condition of falling within a predetermined standard deviation or the condition of the measurement cycle reaching a predetermined number of times is satisfied.

[0046] A specific method for acquiring the sensitivity threshold at the measurement point based on the changed (final) probability density function can also be selected as appropriate. For example, the control unit may acquire the average value of the final probability density function as the sensitivity threshold. The control unit may acquire the average value of a range of probability density functions within a predetermined standard deviation of the final probability density function as the sensitivity threshold. Furthermore, the control unit may set the sensitivity threshold at which the final probability density function is maximized as the sensitivity threshold for the measurement point where measurement has ended.

[0047] When the retina is viewed from the front, the positions of the photoreceptor cells (cones) that convert light information into signals are offset from the positions of the ganglion cells that receive signals from the photoreceptor cells. Hereinafter, a ganglion cell that receives a signal from a certain photoreceptor cell will be referred to as a "ganglion cell corresponding to a certain photoreceptor cell." The control unit may identify the position of the ganglion cell corresponding to the photoreceptor cell present at the measurement point and set at least one of the prior probability density function and initial intensity at the measurement point based on the OCT data at the identified position. In this case, the OCT data is appropriately reflected in the visual field test progress algorithm, taking into account the positional offset between the photoreceptor cells and the ganglion cells.

[0048] The specific method for identifying the positions of ganglion cells corresponding to photoreceptor cells present at the measurement point can be selected as appropriate. For example, the positions of ganglion cells corresponding to photoreceptor cells may be identified based on a model that defines the relationship between the positions of photoreceptor cells and the positions of ganglion cells (e.g., the well-known Drasdo model or Sjostrand model).

[0049] When setting various positions (e.g., the positions of multiple measurement points) on the fundus, one method is to assume that the position where the image of a fixation target is formed on the fundus is the position of the fovea, and to set various positions based on the position of the fixation target. However, in this case, if the fixation of the subject's eye becomes unstable, the various positions set based on the position of the fixation target also become unstable. Furthermore, due to illness or other reasons, the fixation target may be viewed at a position on the fundus that is different from the fovea. As a result, it becomes difficult to perform a visual field test that is in line with the structure of the fundus.

[0050] Therefore, the control unit may acquire a fundus image of the test eye (e.g., an observation image captured in real time during the test or an OCT tomographic image, etc.) and set morphological feature positions on the fundus image, thereby setting various positions based on the morphological feature positions. In this case, even when the test eye's fixation is unstable or when the fixation target is viewed at a retinal position other than the fovea, it becomes easier to perform a visual field test that is appropriate for the fundus structure. The morphological feature positions may be set on the fundus image according to instructions input by a user. The morphological feature positions may also be identified by image processing of the fundus image. For example, the control unit may set various positions on the fundus image using the position of the fovea and the position of the optic disc on the fundus image as morphological feature positions, with a line passing through both the fovea and the optic disc as the X-axis and a line passing through the fovea and perpendicular to the X-axis as the Y-axis. In this case, it becomes easier to appropriately set various positions on the coordinate system.

[0051] A retinal region that captures visual targets instead of the fovea due to a decline in the visual function of the fovea is sometimes called a PRL (Preferred Retinal Locus). If the PRL is significantly off the fovea, the subject's fixation is likely to become unstable, and the visual field test is likely to be unstable as well. The control unit can detect the position where the fixation target is focused using an observation image of the fundus. If the position where the fixation target is focused is outside an allowable range centered on the fovea or if the position where the fixation target is focused is unstable, the control unit may execute at least one of a warning operation indicating that the test is likely to be unstable and a process to change the fixation target presentation method. The process to change the fixation target presentation method can, for example, be a process to change at least one of the shape, size, and brightness of the fixation target, or a process to invert the brightness of the fixation target and the background.

[0052] If the difference between the expected value (average) obtained from the prior probability density function set before the projection of the stimulus target onto the measurement point and the sensitivity threshold obtained as a result of the test for that measurement point exceeds a standard, the control unit may issue a warning to the examiner or may perform a retest for that measurement point, etc. In this case, the accuracy of the visual field test is further improved.

[0053] The control unit may display the visual field test results obtained for each measurement point after aligning them with various images (for example, at least one of an OCT analysis map, an OCT-A, an SLO image, a fundus camera image, a fundus fluorescence image, etc.). In this case, the user (doctor, etc.) can confirm the test results after properly understanding the position of each measurement point on the image.

[0054] 9 is a diagram illustrating the optical system configuration of the visual field examination device 1. FIG. 10 is a block diagram illustrating the electrical configuration of the visual field examination device 1. FIG. 11 is a flowchart of visual field examination control processing executed by the visual field examination device (visual field examination control device) 1. FIG. 12 is a diagram illustrating an example of three-dimensional tomographic image data 51. FIG. 13 is a diagram illustrating an example of OCT angiographic data 52. FIG. 14 is a diagram illustrating an example of thickness comparison data 53. FIG. 15 is a diagram illustrating an example of a frontal observation image 60 of the fundus in a state in which a plurality of measurement points P have been set. FIG. 16 is a diagram illustrating an example of a set prior probability density function. FIG. 17 is a diagram illustrating an example of a likelihood function for changing the prior probability density function. FIG. 18 is a diagram illustrating the result of changing the probability density function shown in FIG. 8 by the invisible likelihood function shown in FIG. 19 is a diagram illustrating a schematic view of the structure of layers and boundaries in the fundus.

[0055] (Configuration of Optical System) A typical embodiment of the present disclosure will now be described with reference to the drawings. First, the configuration of the optical system of a visual field testing device 1 of this embodiment will be described with reference to FIG. 1 . An illumination light source 11 emits infrared light. On the optical axis L1 of the illumination light source 11, a condenser lens 12, a cold mirror 13, a ring slit 14, a relay lens 15, and a perforated mirror 16 are arranged in this order from the illumination light source 11 side. The infrared light beam emitted from the illumination light source 11 illuminates the ring slit 14 via the condenser lens 12 and the cold mirror 13.

[0056] The light beam from the ring slit 14 passes through a relay lens 15 and forms an intermediate image near the opening of a perforated mirror 16. The light beam from the ring slit 14 is reflected by the peripheral surface of the perforated mirror 16. The perforated mirror 16 is disposed on an optical axis L2. The light beam reflected by the perforated mirror 16 is first imaged near the pupil of the subject's eye E by an objective lens 17, and then diffuses to illuminate the fundus Ef of the subject's eye.

[0057] The cold mirror 13 has the property of reflecting visible light and transmitting infrared light. The imaging light source 18 emits visible light for capturing a color image of the fundus Ef. For example, a flash lamp or an LED light source can be used as the imaging light source 18. The flash light emitted from the imaging light source 18 passes through a condenser lens 19, is reflected by the cold mirror 13, and is emitted to the fundus Ef of the subject's eye E via the ring slit 4, the relay lens 15, the perforated mirror 16, and the objective lens 17.

[0058] The reflected beam of infrared light from the fundus Ef passes through the objective lens 17, the perforated mirror 16, and lenses 20, 21, and 22, and is then reflected by the beam splitter 23. The reflected beam of infrared light then passes through the lens 24 and is imaged on the light receiving surface of the observation imaging element 25. The beam splitter 23 has the property of reflecting infrared light and transmitting visible light. The observation imaging element 25 is sensitive in the infrared range. The observation imaging element 25 captures a two-dimensional front observation image of the fundus Ef of the subject's eye E in real time during the visual field test.

[0059] The hole in the perforated mirror 16 is located at a position approximately conjugate with the pupil of the subject's eye E and forms a photographing diaphragm. The lens 21 is a focusing lens that moves along the optical axis L2. By adjusting the position of the lens 21 on the optical axis L2, the fundus oculi Ef and the light receiving surface of the observation photographing element 25 are placed in a conjugate relationship.

[0060] A reflected beam of visible light from the fundus Ef passes through the objective lens 17, the perforated mirror 16, and lenses 20, 21, and 22, before passing through the beam splitter 23. The reflected beam of visible light then passes through a lens 26 and is reflected by a movable mirror 27, after which an image is formed on the light-receiving surface of a color imaging element 28. The color imaging element 28 is sensitive in the visible range. The movable mirror 27 is moved onto the optical axis L2 when capturing a color image, and is retracted from the optical axis L2 during visual field testing.

[0061] The target projection optical system 30 projects a stimulus target onto the subject's eye E for performing a subjective visual field test. A reduction lens 29 is disposed between the target projection optical system 30 and the lens 26. The target projection optical system 30 of this embodiment includes a scanner that adjusts the projection position of the stimulus target on the fundus Ef. The visual field testing device 1 processes a front observation image captured in real time by the observation image capture element 25 and drives the scanner according to the processing results to project the stimulus target onto each of multiple measurement points set on the fundus Ef. As a result, even if the fundus Ef moves, the projection position of the stimulus target is tracked to an appropriate position. The projection intensity of the stimulus target (contrast intensity in this embodiment) is adjusted appropriately. The target projection optical system 30 also projects a fixation target that serves as a fixation target for the subject's eye E. In this embodiment, the brightness, size, and shape of the fixation target can be adjusted.

[0062] (Electrical Configuration) The electrical configuration of the visual field testing device 1 of this embodiment will be described with reference to Figure 2. The visual field testing device 1 includes a control unit 40 that controls various aspects of the visual field testing device 1. The control unit 40 includes a CPU 41, a RAM 42, a ROM 43, and a non-volatile memory (NVM) 44. The CPU 41 is a controller that performs various controls. The RAM 42 temporarily stores various pieces of information. The ROM 43 stores programs executed by the CPU 41, various initial values, etc. The NVM 44 is a non-transitory storage medium that can retain its contents even when the power supply is interrupted.

[0063] The visual field testing device 1 of this embodiment uses the control unit 40 to control various operations during the visual field test (e.g., the operation of capturing a front observation image using the observation imaging element 25 and the operation of projecting a stimulus target using the target projection optical system 30). Furthermore, the visual field testing device 1 of this embodiment controls the visual field test by setting parameters that determine the progress algorithm of the visual field test (e.g., the initial intensity of the stimulus target and a priori probability density function). In other words, the visual field testing device 1 of this embodiment also serves as a visual field test control device for controlling the visual field test. However, a device other than the visual field testing device 1 (e.g., a personal computer connected to the visual field testing device 1) may also function as a visual field test control device for controlling the test performed by the visual field testing device 1. Furthermore, the control units of multiple devices (e.g., the control unit of the visual field testing device 1 and the control unit of the personal computer) may cooperate to function as a visual field test control device.

[0064] The control unit 40 is connected to the illumination light source 11, the imaging light source 18, the observation imaging element 25, the color imaging element 28, the target projection optical system 30, the display unit 31, the operation unit 32, and the response switch 33. The display unit 31 displays various images. Various devices capable of displaying images can be used for the display unit 31. The operation unit 32 is operated by a user to input various instructions to the visual field testing device 1. The operation unit 32 can be, for example, at least one of a keyboard, a mouse, a touch panel, etc. Note that a microphone or the like for inputting various instructions may be used together with or instead of the operation unit 32. The response switch 33 is operated by the subject during the visual field test. In other words, the CPU 41 identifies whether the subject was able to visually recognize the stimulus target (i.e., "visible" if the stimulus target was visible, or "invisible" if the stimulus target was not visible) based on the operation signal of the response switch 33. As an example, in this embodiment, if the subject can visually recognize the stimulus target, the response switch 33 is operated, and if the subject cannot visually recognize the stimulus target, the response switch 33 is not operated. Alternatively, a switch to be operated when the stimulus target is visually recognized and a switch to be operated when the stimulus target is not visually recognized may be provided separately.

[0065] The visual field testing device 1 acquires OCT data of the subject's eye to be tested. For example, the visual field testing device 1 may acquire the OCT data from the OCT device 50 via at least one of wired communication, wireless communication, a network (e.g., the Internet), etc. Alternatively, the visual field testing device 1 may acquire the OCT data via a removable memory or the like.

[0066] The OCT device 50 acquires OCT data about the fundus Ef of the subject's eye E. The OCT data is obtained from the reflected light of OCT light from the fundus Ef and the interference light of reference light corresponding to the OCT light. In detail, the OCT device 50 includes an OCT light source, a branching optical element, an irradiation optical system, a light receiving element, etc. The OCT light source emits OCT light. The branching optical element branches the OCT light emitted from the OCT light source into measurement light and reference light. The irradiation optical system irradiates the measurement light onto the fundus Ef of the subject's eye E. The light receiving element receives the combined light of the light reflected by the fundus Ef and the reference light.

[0067] (Visual Field Test Control Processing) An example of visual field test control processing executed by the visual field test device (visual field test control device) 1 will be described with reference to Figures 3 to 11. In the visual field test control processing, parameters that determine the progress algorithm of the visual field test (e.g., the initial intensity of the stimulus target and a priori probability density function) are set, and various operations during the visual field test are controlled in accordance with the set parameters. The visual field test control processing illustrated in Figure 3 is executed by the CPU 41 in accordance with a visual field test control program stored in the NVM 44.

[0068] First, the CPU 41 acquires OCT data of the fundus oculi Ef of the subject's eye E obtained by the OCT device 50 (S1). In detail, in S1 of this embodiment, three-dimensional tomographic image data 51 (see FIG. 4), OCT angiographic data 52 (see FIG. 5), and thickness comparison data 53 (see FIG. 6) are acquired as the OCT data. Details of each OCT data will be described later.

[0069] The CPU 41 starts capturing a front observation image of the fundus Ef in real time using the observation image capturing element 25 while projecting a fixation target onto the subject's eye E using the target projection optical system 30 (see FIG. 1) (S2). At this time, the CPU 41 performs autofocusing of the front observation image of the fundus Ef. The CPU 41 also selects a reference observation image from the multiple front observation images captured intermittently. The selected reference observation image serves as the basis for various alignment processes, including tracking of the projection position of the stimulus target (so-called "tracking").

[0070] The CPU 41 executes at least one of a process for changing the fixation target presentation method and a warning process as necessary (S3). A retinal region that captures a visual target instead of the fovea due to a decline in the visual function of the fovea of ​​the subject's eye E is sometimes called a PRL (Preferred Retinal Locus). If the PRL is significantly off the fovea, the subject's fixation is likely to become unstable, and the visual field test is likely to be unstable as well. In S3, the CPU 41 detects the position where the fixation target is focused using a frontal observation image of the fundus oculi Ef. If the position where the fixation target is focused is outside an allowable range centered on the fovea, or if the position where the fixation target is focused is unstable, the CPU 41 executes at least one of a warning operation indicating that the test is likely to be unstable and a process for changing the fixation target presentation method. The process for changing the fixation target presentation method can include, for example, changing at least one of the shape, size, and brightness of the fixation target, or reversing the brightness of the fixation target and the background.

[0071] Next, the CPU 41 aligns the OCT data acquired in S1 with the front observation image (the reference observation image in this embodiment) captured by the observation image capturing element 25 (S4). That is, registration processing is performed so that the positions of the front observation image and the OCT data match when the fundus Ef is viewed from the front.

[0072] The CPU 41 sets morphological feature positions on the front observation image (the reference observation image in this embodiment), and sets a plurality of measurement points onto which the stimulus target is projected, using the set morphological feature positions as references, on the front observation image (S5). As a result, the effect on the test results when fixation becomes unstable is smaller than when various positions such as measurement points are set using the position on the fundus Ef where the image of the fixation target is formed as reference.

[0073] As an example, in this embodiment, as shown in FIG. 7 , known image processing is performed on a front observation image 60 to identify the positions of the fovea (central fovea) 61 and the optic disc 62, which are morphological feature positions. The CPU 41 determines the X-axis as a line passing through the fovea 61 and the optic disc 62. The CPU 41 also determines the Y-axis as a line passing through the fovea 61 and perpendicular to the X-axis. The CPU 41 sets multiple measurement points P on the front observation image 60 based on the set coordinate axes. As a result, the measurement points P are appropriately set on coordinates based on the morphological feature positions. However, the specific method for setting the positions of the measurement points P can be changed. For example, the CPU 41 may set the morphological feature positions on the front observation image 60 in accordance with an instruction input by the user.

[0074] The CPU 41 identifies the position of the ganglion cell corresponding to the photoreceptor cell at the set measurement point P (S6). The CPU 41 sets parameters (in this embodiment, a priori probability density function and initial intensity) that determine the progression algorithm of the visual field test based on the OCT data of the identified position (i.e., the position of the ganglion cell corresponding to the photoreceptor cell at the measurement point P) (S7, S10).

[0075] As described above, when the retina is viewed from the front, the positions of the photoreceptor cells (cones) that convert light information into signals are offset from the positions of the ganglion cells that receive signals from the photoreceptor cells. The "position of the ganglion cells corresponding to the photoreceptor cells" refers to the position of the ganglion cells that receive signals from the photoreceptor cells. The visual function at the position of a certain photoreceptor cell is related to the state of the retina at the position of the ganglion cells corresponding to the photoreceptor cells. Therefore, by setting the parameters of the visual field test based on OCT data of the position of the ganglion cells corresponding to the photoreceptor cells at measurement point P, the OCT data is appropriately reflected in the visual field test progression algorithm while taking into account the positional offset between the photoreceptor cells and the ganglion cells. As an example, in this embodiment, the positions of the ganglion cells corresponding to the photoreceptor cells are identified based on a model (e.g., the well-known Drasdo model or Sjostrand model) that defines the relationship between the positions of the photoreceptor cells and the ganglion cells.

[0076] Here, an overview of the visual field testing method employed in this embodiment will be described with reference to Figures 3 and 8 to 10. In this embodiment, statistics (Bayesian statistics, for example) are used in the visual field testing method. In detail, in this embodiment, a probability density function is changed based on the subject's response to the stimulus target, and a sensitivity threshold (in this embodiment, a contrast sensitivity threshold) at the measurement point is obtained based on the changed probability density function.

[0077] First, as shown in Fig. 8, the CPU 41 sets in advance a probability density function for each measurement point P, with the sensitivity threshold as a random variable (S7). Details of the processing of S7 will be described later. According to the probability density function shown in Fig. 8, it can be seen that there is a high probability that the sensitivity threshold is between 20 dB and 30 dB. Note that, hereinafter, the probability density function set in S7 before projecting the stimulus target onto the measurement point P is referred to as a priori probability density function.

[0078] The CPU 41 sets the initial intensity of the Nth measurement point P (initial value is "1") among the plurality of measurement points P (S10). The initial intensity is the projection intensity of the stimulus target that is first projected onto the Nth measurement point P. In Figs. 8 to 10, the projection intensity K is set to approximately 23 dB. Details of S10 will be described later.

[0079] The CPU 41 projects a stimulus target onto the Nth measurement point P at a set projection intensity K (initial intensity). The subject changes the operation of the response switch 33 depending on whether or not the stimulus target is visible (i.e., "visible" or "invisible"). As an example, in this embodiment, the subject operates the response switch 33 if the stimulus target is visible, and does not operate the response switch 33 if the stimulus target is not visible. The CPU 41 acquires the subject's response to the stimulus target (S11).

[0080] The CPU 41 changes the probability density function in accordance with the response result acquired in S11 (S12). Specifically, if the response result is "visible", the CPU 41 subtracts the probability density of values ​​less than the projection threshold K from the probability density function. If the response result is "invisible", the CPU 41 subtracts the probability density of values ​​equal to or greater than the projection intensity K from the probability density function.

[0081] In this embodiment, the probability density function is normalized by multiplying it by a likelihood function (see FIG. 9 ) corresponding to the response result, so that the probability density function is changed according to the response result. In the example shown in FIG. 9 , the likelihood function used when the response result is "invisible" is indicated by a solid line. The likelihood function when the response result is "invisible" is set so that the probability density decreases as the sensitivity threshold becomes larger than the previous projection intensity K (approximately 23 dB in the examples of FIGS. 8 to 10 ). Furthermore, the likelihood function when the response result is "visible" is set so that the probability density decreases as the sensitivity threshold becomes smaller than the previous projection intensity K.

[0082] For example, if the response result is "invisible," the probability density function shown in FIG. 8 is multiplied by the likelihood function of "invisible (solid line)" shown in FIG. 9, changing the probability density function from the dotted line (state of FIG. 9) to the solid line in FIG. 10, and then normalizing it. That is, in the example shown in FIG. 10, multiplying the likelihood function of "invisible" by the probability density function reduces the probability density of the previous projection intensity K (approximately 23 dB) or greater in the probability density function. As a result, in the changed probability density function, the probability that the sensitivity threshold will be greater than 23 dB is significantly reduced. On the other hand, if the response result is "visible," the probability density function shown in FIG. 8 is multiplied by the likelihood function of "visible (dotted line)" shown in FIG. 9 and normalized. As a result, in the changed probability density function, the probability that the sensitivity threshold will be smaller than the previous projection intensity K is significantly reduced.

[0083] The CPU 41 determines whether a measurement termination condition for the Nth measurement point P is met (S14). The measurement termination condition can also be set as appropriate. For example, the termination condition for the measurement for the Nth measurement point P may be that the changed probability density function (see, for example, FIG. 10) falls within a predetermined standard deviation. Alternatively, the termination condition for the measurement for the Nth measurement point P may be that the measurement cycle of projecting the stimulus target, obtaining the response result (S11), and changing the probability density function (S12) performed for the Nth measurement point P reaches a predetermined number of times. Alternatively, the termination condition may be that either the condition of falling within a predetermined standard deviation or the condition of the measurement cycle reaching a predetermined number of times is met.

[0084] If the measurement termination condition is not met (S14: NO), the CPU 41 sets the projection intensity of the stimulus target to be next projected at the Nth measurement point P based on the subject's response to the previously projected stimulus target (S15). In S15 of this embodiment, the CPU 41 sets the projection intensity of the stimulus target to be next projected at the Nth measurement point P based on the probability density function set at that time (i.e., the probability density function changed in S12). Specifically, the CPU 41 sets the expected value (average) calculated from the probability density function as the projection intensity. Alternatively, the CPU 41 may set the sensitivity threshold at which the probability density function is maximized as the projection intensity. As a result, the next projection intensity is appropriately determined based on the probability density function. Thereafter, the process returns to S11, and the processes of S11 to S14 are repeated for the Nth measurement point P.

[0085] If the measurement termination condition for the Nth measurement point P is satisfied (S14: YES), the CPU 41 acquires a sensitivity threshold (a contrast sensitivity threshold in this embodiment) for the Nth measurement point P based on the final probability density function changed in S12 (S16). As an example, in this embodiment, the average value of the final probability density function is acquired as the sensitivity threshold for the Nth measurement point P. As a result, an appropriate value according to the probability density function is acquired. Alternatively, the CPU 41 may acquire a sensitivity threshold at which the final probability density function is maximized as the sensitivity threshold for the Nth measurement point P.

[0086] Next, the CPU 41 determines whether the test for all (N) measurement points P has been completed (S18). If the test for all measurement points P has not been completed (S18: NO), the CPU 41 corrects the prior probability density functions for the other measurement points P based on the test results already obtained (e.g., the sensitivity threshold for the Nth measurement point P obtained in S16) (S19). For example, the CPU 41 may correct the prior probability density functions of measurement points P located near the Nth measurement point P based on the sensitivity threshold for the Nth measurement point P. In this case, the CPU 41 may increase the amount of correction the shorter the distance from the Nth measurement point P. As a result, the visual field test for the other measurement points P is performed after appropriately considering the results of the measurement for the measurement point P that has already been completed. This further improves the efficiency of the test. Thereafter, the CPU 41 increments the counter N that identifies the measurement point P by "1" and performs measurement for the next measurement point (S10 to S18). Note that, for example, if a predetermined convergence condition is satisfied as a result of performing statistical interpolation for the Nth measurement point P based on the inspection results of a measurement point P located near the Nth measurement point P, the inspection of the Nth measurement point P may be omitted. In other words, inspection of some of the multiple measurement points P may be omitted based on the inspection results of other measurement points P.

[0087] When the test for all measurement points P is completed (S18: YES), the CPU 41 determines whether the difference between the expected value (average) obtained from the prior probability density function for each measurement point P and the sensitivity threshold obtained as a result of the test for each measurement point P exceeds the standard. If the difference exceeds the standard, the CPU 41 issues a warning (alert) to the examiner or performs a retest for the measurement points P whose difference exceeds the standard. As a result, the accuracy of the visual field test is further improved.

[0088] The CPU 41 displays the visual field test results acquired for each measurement point P after aligning them with various images (for example, at least one of an OCT analysis map, an OCT angiography, an SLO image, a fundus camera image, a fundus fluorescence image, etc.). Thus, a user (doctor, etc.) can confirm the test results after properly understanding the position of each measurement point P on the image.

[0089] The CPU 41 may analyze the progression direction of the visual field defect by comparing the results of a visual field test previously performed on the same subject's eye E with the results of the current visual field test. For example, the CPU 41 may calculate the difference between the previous test result and the current test result for each measurement point P and perform ellipse analysis on the data of the four points surrounding each measurement point P. As a result of the ellipse analysis, the strength of change is obtained from the lengths of the major and minor axes, and the progression direction of the visual field defect is obtained from the angle between the major and minor axes. Therefore, by displaying the results of the ellipse analysis on the display unit 31, the user can appropriately diagnose the progression of the visual field defect.

[0090] (Parameter Setting Method) A method for setting parameters (prior probability density function and initial intensity of stimulus target) that determine the progression algorithm of the visual field test will be described in detail below. In this embodiment, the parameters of the visual field test are set based on the OCT data of the subject's eye E acquired in S1. The OCT data acquired in S1 in this embodiment includes at least one of three-dimensional tomographic image data 51 (see FIG. 4), OCT angiographic data 52 (see FIG. 5), and thickness comparison data 53 (see FIG. 6).

[0091] The three-dimensional tomographic image data 51 (see FIG. 4 ) shows the three-dimensional tomographic structure of the retinal layers of the fundus oculi Ef. There may be a correlation between the tomographic structure of the retinal layers (e.g., the thickness of a particular layer) and the visual function of the subject's eye E. Therefore, by controlling the visual field test based on the three-dimensional tomographic image data 51, the visual field test can be easily performed in a short time with high accuracy.

[0092] The OCT angiographic data 52 (see FIG. 5 ) is motion contrast data generated by arithmetic processing of at least two OCT signals acquired at different times from the same position on the subject's eye Ef. The OCT angiographic data 52 shown in FIG. 5 is a two-dimensional image of motion contrast data of a specific region (sometimes called a "slab") of the retinal layer of the fundus Ef sandwiched between two specific boundary surfaces, viewed from the front side (in other words, along the optical axis of the OCT measurement light). The OCT angiographic data 52 includes perfusion information of the vascular network of the retinal layer of the fundus Ef. The perfusion state of the blood vessels in the fundus Ef affects the visual function of the subject's eye E. Therefore, controlling a visual field test based on the OCT angiographic data 52 makes it easier to perform the visual field test in a short time and with high accuracy.

[0093] The thickness comparison data 53 (see FIG. 6 ) is a type of OCT data related to the tomography of the fundus oculi Ef. The thickness comparison data 53 of this embodiment indicates the difference between the thickness distribution of a layer (hereinafter, also referred to as a “specific layer”) in a specific region (sometimes referred to as a “slab”) of the retinal layers of the fundus oculi Ef sandwiched between two specific boundary surfaces and the thickness distribution of the layer in a specific region of a normal eye. Data indicating the thickness distribution of a specific layer (sometimes referred to as a “thickness map”) can be obtained, for example, by analyzing the three-dimensional tomographic image data 51. As mentioned above, there may be a correlation between the tomographic structure of the retinal layers and the visual function of the subject's eye E. Therefore, controlling the visual field test based on the thickness comparison data 53 makes it easier to perform the visual field test in a short time and with high accuracy. The thickness comparison data 53 illustrated in FIG. 6 is data comparing the thickness distribution of the ganglion cell complex consisting of the NFL, GCL, and IPL with the thickness distribution of a normal eye.

[0094] Here, the layer structure of the fundus oculi Ef will be explained with reference to Fig. 11. Fig. 11 shows a schematic diagram of the layer and boundary structure of the fundus oculi Ef. The upper side of Fig. 11 is the surface side of the retina of the fundus oculi Ef. In other words, the depth of the layers and boundaries increases toward the bottom of Fig. 11. Furthermore, in Fig. 11, the names of boundaries between adjacent layers are enclosed in parentheses.

[0095] The layers of the fundus Ef will be described. In the fundus Ef, from the surface side (upper side of FIG. 6 ), there are, in order, the ILM (internal limiting membrane), the NFL (nerve fiber layer), the GCL (ganglion cell layer), the IPL (inner plexiform layer), the INL (inner nuclear layer), the OPL (outer plexiform layer), the ONL (outer nuclear layer), and the ELM (external limiting membrane). membrane), IS / OS (junction between photoreceptor inner and outer segment), RPE (retinal pigment epithelium) epithelium), BM (Bruch's membrane), and Choroid.

[0096] Among the layer structures of the fundus Ef, there is a correlation between the thickness of the ganglion cell complex (GCC), which consists of the NFL, GCL, and IPL, and the visual function of the test eye. Furthermore, new findings by the present inventors have suggested that the perfusion state of blood vessels near the ganglion cells present in the GCC may be more closely related to the visual function of the test eye than the thickness of the GCC. Most of the blood vessels near the ganglion cells are present in the GCL and IPL.

[0097] (Method for Setting the Priori Probability Density Function) The method for setting the priori probability density function (S7) will be described in detail. As described above, in this embodiment, the sensitivity threshold at the measurement point P is acquired by changing the probability density function in accordance with the subject's response to the stimulus target. Therefore, the more appropriate the priori probability density function set in S7, the easier it is to perform a visual field test in a short time with high accuracy. In S7 of this embodiment, the CPU 41 sets the priori probability density function based on OCT data, which is likely to provide various information regarding the state of the fundus Ef (e.g., at least one of the structure of the fundus layers and the perfusion state of blood vessels). Therefore, the sensitivity threshold at the measurement point P is acquired based on the priori probability density function that reflects information regarding the state of the fundus Ef that affects visual function. This makes it easier to perform a visual field test in a short time with high accuracy.

[0098] In S7 of this embodiment, the CPU 41 sets a prior probability density based on the OCT angiographic data 52 (see FIG. 5 ) at a position corresponding to the measurement point P (i.e., a position taking into consideration the positional displacement of ganglion cells relative to photoreceptor cells). As a result, information on the perfusion state of retinal blood vessels, which affects visual function, is appropriately reflected in the prior probability density function. This makes it easier to perform visual field testing in a shorter time and with higher accuracy. As an example, the CPU 41 of this embodiment sets a prior probability density function at each measurement point P based on the vascular (perfusion) density (e.g., the area or length of blood vessels per unit area) of the measurement point P analyzed from the OCT angiographic data 52.

[0099] It is also possible to change the method for setting a prior probability density function based on OCT data (OCT angiographic data in this embodiment). For example, assume that certain OCT angiographic data is acquired and the sensitivity threshold for a visual field test is X dB. In this case, the number of samples for X dB is added to the prior probability density function when similar OCT angiographic data is acquired (for example, a histogram in which the horizontal axis is expressed in dB, as in FIG. 8 ). The number of samples may be accumulated for each OCT angiographic data and further normalized to set a prior probability density function corresponding to the OCT angiographic data.

[0100] Note that blood vessels exist in each of a plurality of regions at different depths in the retina of the fundus oculi Ef. In S7 of the present embodiment, the CPU 41 sets a prior probability density based on the OCT angiographic data 52 in a specific region sandwiched between two specific boundaries (slabs) in the retinal layer of the fundus oculi Ef. As a result, perfusion information of a desired blood vessel (e.g., a blood vessel highly associated with visual function) among the plurality of blood vessels is more likely to be appropriately reflected in the prior probability density function.

[0101] In detail, as described above, there is a strong possibility that the perfusion state of blood vessels near ganglion cells present in the GCC is highly correlated with the visual function of the subject's eye. Many of the blood vessels near the ganglion cells are present in the GCL and IPL. In this embodiment, the prior probability density function is set based on OCT angiographic data 52 of a specific region including at least a portion of the GCL and IPL. As a result, information on the perfusion state of blood vessels that is likely to affect visual function is more likely to be reflected in the prior probability density function more appropriately.

[0102] More specifically, the two boundaries defining the specific region are located between the boundary of the GCL on the retinal surface side (the NFL / GCL boundary in FIG. 11 ) and the boundary of the IPL on the retinal deep side (the IPL / INL boundary in FIG. 11 ). That is, of the two specific boundaries, the boundary on the retinal surface side (upper side) is set at a position below the NFL / GCL boundary, and the boundary on the retinal deep side (lower side) is set at a position above the IPL / INL boundary. In this case, layers other than the GCL and INL are not included in the specific region. Therefore, information about blood vessels other than those near ganglion cells is less likely to be reflected in the prior probability density function. Therefore, information about the perfusion state of blood vessels that likely affect visual function is more likely to be reflected in the prior probability density function more appropriately.

[0103] 7 , in S7 of this embodiment, when the fundus oculi Ef is viewed from the front, the prior probability density of measurement points P located within a central region 72 including the fovea centralis 61 is set based on the OCT angiographic data 52. The central region 72 tends to have a stronger correlation between the vascular perfusion state and visual function than the peripheral regions. Therefore, by setting the prior probability density function of at least the measurement points within the central region 72 based on the OCT angiographic data 52, information about the vascular perfusion state is more likely to be reflected in the prior probability density function more appropriately.

[0104] In S7 of the present embodiment, the prior probability density function of the measurement points P located in the peripheral region 73 surrounding the central region 72 is set based on OCT data relating to the tomography. In the peripheral region 73, there tends to be a stronger correlation between the tomographic state of the fundus Ef (e.g., the thickness of a particular layer) and visual function than in the central region 72. Therefore, by setting the prior probability density function of at least the measurement points P in the peripheral region 73 based on OCT data relating to the tomography, the state of the fundus Ef is more likely to be reflected in the prior probability density function more appropriately.

[0105] In detail, in S7 of the present embodiment, the prior probability density function of the measurement point P located within the peripheral region 73 is set based on thickness comparison data 53 (see FIG. 6 ) that compares the thickness distribution of a specific layer (the ganglion cell complex in this embodiment) in the fundus Ef with the thickness distribution of the specific layer in a normal eye. Therefore, the thickness state of the specific layer that is likely to affect visual function is more appropriately reflected in the prior probability density function.

[0106] In S7 of the present embodiment, the CPU 41 sets the prior probability density functions of all measurement points P based on the OCT data relating to the tomography (thickness comparison data 53 in the present embodiment) regardless of whether they are in the central region 72 or the peripheral region 73. Thereafter, the CPU 41 corrects the prior probability density functions of the measurement points P in the central region 72 based on the OCT angiographic data 52. As a result, the prior probability density functions in each of the central region 72 and the peripheral region 73 are set appropriately.

[0107] (Initial Intensity Setting Method) The initial intensity setting method (S10) of the stimulus target will be described. As described above, the initial intensity is the projection intensity of the stimulus target initially projected onto each measurement point P. The more appropriate the set initial intensity, the easier it is to perform the visual field test in a short time with high accuracy. In S10 of this embodiment, the CPU 41 sets the initial intensity at each measurement point P based on the OCT angiogram data 52 ( FIG. 52 ) at the position corresponding to each measurement point P. As a result, information on the perfusion state of blood vessels in the retinal layer, which affects visual function, is appropriately reflected in the initial intensity. In other words, it is easier to set an appropriate initial intensity according to the blood vessel perfusion state. This makes it easier to perform the visual field test in a short time with high accuracy. As an example, in S10 of this embodiment, the initial intensity at each measurement point P is set based on the blood vessel (perfusion) density (e.g., the area or length of blood vessels per unit area) at the measurement point P analyzed from the OCT angiogram data 52.

[0108] In S10 of this embodiment, the CPU 41 sets the initial intensity at each measurement point P based on the OCT angiographic data 52 in a specific region sandwiched between two specific boundaries (slabs) in the retinal layer of the fundus oculi Ef. As a result, perfusion information of a desired blood vessel (e.g., a blood vessel highly related to visual function) among multiple blood vessels is more likely to be appropriately reflected in the initial intensity.

[0109] Specifically, the initial intensity is set based on the OCT angiographic data 52 of a specific region including at least a part of the GCL and the IPL. As a result, information on the perfusion state of blood vessels that is likely to affect visual function is more appropriately reflected in the initial intensity.

[0110] More specifically, the two boundaries (slabs) defining the range of the OCT angiographic data 52 are located between the boundary of the GCL on the superficial retinal side (the NFL / GCL boundary in FIG. 11 ) and the boundary of the IPL on the deep retinal side (the IPL / INL boundary in FIG. 11 ). That is, of the two specific boundaries, the boundary on the superficial retinal side (upper side) is set at a position below the NFL / GCL boundary, and the boundary on the deep retinal side (lower side) is set at a position above the IPL / INL boundary. In this case, perfusion information of blood vessels other than those near ganglion cells is less likely to be reflected in the initial intensity. Therefore, information on the perfusion state of blood vessels that likely affect visual function is more likely to be reflected in the initial intensity more appropriately.

[0111] 7 , in S10 of this embodiment, the initial intensity of measurement points P located within a central region 72 including the fovea 61 is set based on the OCT angiographic data 52. The central region 72 tends to have a stronger correlation between vascular perfusion and visual function than the peripheral regions. Therefore, by setting the initial intensity of at least measurement points within the central region 72 based on the OCT angiographic data 52, information about the vascular perfusion state is more likely to be reflected in the initial intensity more appropriately.

[0112] Furthermore, the initial intensity of measurement points P located within a peripheral region 73 surrounding the central region 72 is set based on OCT data relating to the tomography (in this embodiment, thickness comparison data 53 comparing the distribution of the thickness of the ganglion cell complex with that of a normal eye). In the peripheral region 73, there tends to be a stronger correlation between the tomographic state of the fundus Ef (for example, the thickness of a specific layer) and visual function than in the central region 72. Therefore, by setting the initial intensity of at least measurement points P within the peripheral region 73 based on OCT data relating to the tomography, the state of the fundus Ef is more likely to be reflected in the initial intensity more appropriately.

[0113] In S10, the CPU 41 sets the projection intensity of the stimulus target to be projected onto the measurement point P based on the probability density function set at that time. Therefore, in S10, the CPU 41 sets the initial intensity of the stimulus target to be projected onto the measurement point P based on the a priori probability density function set in S7. As described above, in S7, the a priori probability density function is set based on the OCT data. Therefore, by setting the initial intensity based on the a priori probability density function set in S7, the OCT data is appropriately reflected in the initial intensity. As an example, in S10 of this embodiment, the expected value (average) calculated from the probability density function is set as the projection intensity.

[0114] The techniques disclosed in the above embodiments are merely examples. Therefore, the techniques exemplified in the above embodiments can be modified. For example, in the above embodiments, both the prior probability density function and the initial intensity are set based on OCT data. However, only one of the prior probability density function and the initial intensity may be set based on OCT data. For example, when a visual field examination method that does not use a prior probability density function is adopted, the initial intensity may be set based on OCT angiographic data 52.

[0115] The process of setting a priori probability density function in S7 of FIG. 3 is an example of an "priori probability density function setting step." The process of setting an initial intensity in S10 of FIG. 3 is an example of an "initial intensity setting step." The processes of acquiring a response result from the subject in S11 and S12 of FIG. 3 are an example of a "response result acquisition step." The process of changing the probability density function in S12 of FIG. 3 is an example of a "function changing step." The process of determining whether a termination condition is satisfied in S14 of FIG. 3 is an example of a "determination step." The process of repeating S11 to S14 when the termination condition is not satisfied in S14 of FIG. 3 is an example of a "repetition step." The process of acquiring a sensitivity threshold in S16 of FIG. 3 is an example of a "sensitivity threshold acquisition step." The process of acquiring OCT data in S1 of FIG. 3 is an example of an "OCT data acquisition step" and an "angio data acquisition step."

Claims

1. A visual field testing device that projects a stimulus target onto each of a plurality of measurement points on a fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target, The control unit of the visual field testing device a prior probability density function setting step of setting in advance a probability density function in which the sensitivity threshold value is a random variable for each of the measurement points; an initial intensity setting step for setting an initial intensity, which is the projection intensity of the stimulus target that is first projected onto the measurement point; a response result acquisition step of projecting the stimulus target onto the measurement point with a set projection intensity and acquiring a response result of the subject to the stimulus target; a function modification step of modifying the probability density function by subtracting a probability density less than the projection intensity in the probability density function when the response result acquired in the response result acquisition step is visible, which indicates that the stimulus target was visible, and by subtracting a probability density equal to or greater than the projection intensity in the probability density function when the response result is invisible; a determination step of determining whether a measurement termination condition for the measurement point is satisfied; a repeating step of repeating the response result obtaining step, the function changing step, and the determining step, after newly setting a projection intensity according to a previous response result when it is determined in the determining step that the termination condition is not satisfied; a sensitivity threshold acquisition step of acquiring a sensitivity threshold at the measurement point based on the changed probability density function when it is determined in the determination step that the termination condition is satisfied; an OCT data acquisition step of acquiring OCT data obtained from interference light between reflected light of OCT light from the fundus of the subject's eye and reference light corresponding to the OCT light; Run A visual field examination device characterized in that, in the prior probability density function setting step, the probability density function is set in advance based on the OCT data about the fundus of the test eye acquired in the OCT data acquisition step.

2. 2. The visual field examination device according to claim 1, A visual field examination device characterized in that the OCT data includes OCT angiographic data, which is motion contrast data generated by processing at least two OCT signals acquired at different times for the same position on the fundus of the test eye.

3. 3. The visual field examination device according to claim 2, A visual field examination device characterized in that in the prior probability density setting step, the probability density function is set in advance based on the OCT angiographic data in a specific region of the retinal layer of the fundus of the test eye that is sandwiched between two specific boundary surfaces.

4. 3. The visual field examination device according to claim 2, A visual field examination device characterized in that in the prior probability density function setting step, a probability density function of measurement points located within a central region including the fovea when the fundus of the test eye is viewed from the front is set based on the OCT angiographic data.

5. 5. The visual field examination device according to claim 1, A visual field testing device characterized in that the control unit sets the projection intensity of the stimulus target to be projected onto the measurement point based on the probability density function set at that time.

6. A visual field testing device that projects a stimulus target onto each of a plurality of measurement points on a fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target, The control unit of the visual field testing device an angio data acquisition step of acquiring OCT angio data, which is motion contrast data generated by processing at least two OCT signals acquired at different times for the same position on the fundus of the subject eye; an initial intensity setting step of setting an initial intensity, which is the projection intensity of a stimulus target initially projected onto each of the measurement points, based on the OCT angiographic data; A visual field testing device characterized by performing the above.

7. A visual field examination program executed by a visual field examination control device for controlling an examination by a visual field examination device, the visual field examination device projects a stimulus target onto each of a plurality of measurement points on a fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target; The visual field examination program is executed by the control unit of the visual field examination control device, a prior probability density function setting step of setting in advance a probability density function in which the sensitivity threshold value is a random variable for each of the measurement points; an initial intensity setting step for setting an initial intensity, which is the projection intensity of the stimulus target that is first projected onto the measurement point; a response result acquisition step of projecting the stimulus target onto the measurement point with a set projection intensity and acquiring a response result of the subject to the stimulus target; a function modification step of modifying the probability density function by subtracting a probability density less than the projection intensity in the probability density function when the response result acquired in the response result acquisition step is visible, which indicates that the stimulus target was visible, and by subtracting a probability density equal to or greater than the projection intensity in the probability density function when the response result is invisible; a determination step of determining whether a measurement termination condition for the measurement point is satisfied; a repeating step of repeating the response result obtaining step, the function changing step, and the determining step, after newly setting a projection intensity according to a previous response result when it is determined in the determining step that the termination condition is not satisfied; a sensitivity threshold acquisition step of acquiring a sensitivity threshold at the measurement point based on the changed probability density function when it is determined in the determination step that the termination condition is satisfied; an OCT data acquisition step of acquiring OCT data obtained from interference light between reflected light of OCT light from the fundus of the subject's eye and reference light corresponding to the OCT light; causing the visual field examination control device to execute A visual field examination program characterized in that, in the prior probability density function setting step, the probability density function is set in advance based on the OCT data about the fundus of the test eye acquired in the OCT data acquisition step.

8. A visual field examination program executed by a visual field examination control device for controlling an examination by a visual field examination device, the visual field examination device projects a stimulus target onto each of a plurality of measurement points on a fundus of a subject's eye to be examined, and acquires a sensitivity threshold at each of the measurement points based on a response result of the subject to the stimulus target; The visual field examination program is executed by the control unit of the visual field examination control device, an angio data acquisition step of acquiring OCT angio data, which is motion contrast data generated by processing at least two OCT signals acquired at different times for the same position on the fundus of the subject eye; an initial intensity setting step of setting an initial intensity, which is the projection intensity of a stimulus target initially projected onto each of the measurement points, based on the OCT angiographic data; A visual field examination program that causes the visual field examination control device to execute the above.