JP6436293B2 - Fundus imaging device with wavefront compensation - Google Patents

Description

The present disclosure relates to the wavefront compensation with fundus photography equipment for photographing a fundus image of the eye in a state of correcting the wavefront aberration of the eye.

  2. Description of the Related Art There is known an apparatus that detects wavefront aberration of an eye using a wavefront sensor such as a Shack-Hartmann sensor, controls a wavefront compensation device based on the detection result, and photographs a fundus image after wavefront compensation at a cellular level. The fundus image obtained by such an apparatus can be used for image processing relating to cells of the fundus, such as cell density analysis (see Patent Document 1).

JP 2014-110825 A

  In order to properly perform image processing related to cells of the fundus, it is necessary to use a fundus image with high reliability. However, since the fundus photographing apparatus with wavefront compensation has a high magnification, the image quality of the fundus image is easily affected by the movement of the eye.

The present disclosure, in view of the problems of the prior art, and an object thereof is to provide a fundus photographing equipment with wavefront compensation can be satisfactorily image processing relates to a fundus of the cells.

In order to solve the above problem, the fundus imaging apparatus with wavefront compensation according to the first aspect of the present disclosure receives reflected light from the fundus in a state where the wavefront aberration is compensated for by the wavefront compensation device by the light receiving element. A fundus imaging apparatus with wavefront compensation comprising: a fundus imaging optical system for capturing a fundus image by: and an image forming unit that forms the fundus image based on a light reception signal from the light receiving element. The fundus is moved between each fundus image by setting a cell image forming the retina as a reference region and comparing image regions corresponding to the reference region included in each fundus image repeatedly formed by the image forming means. a moving speed measuring means for measuring the speed, based on a measurement result by the moving speed measurement unit, double that is repeatedly formed by the image forming means An image selecting means to sift the fundus image, comprising a.

According to the wavefront compensation with fundus photography equipment of the present disclosure, there is an effect that it is possible to satisfactorily perform the image processing relates to a fundus of the cells.

It is the figure which showed the external view of the imaging device 1 of this embodiment. 2 is a schematic diagram illustrating an optical system of the photographing apparatus 1. FIG. It is the block diagram which showed the control system of the imaging device 1 of this embodiment. 5 is a flowchart illustrating main processing executed by a control unit 800. It is the figure which showed an example of the reference | standard image. It is the flowchart which showed the operation | movement of the apparatus at the time of continuous imaging | photography of an AOSLO image. 6 is a graph showing a positional shift of a shooting position of each frame with respect to a reference image. It is the graph which showed the relative speed of the fundus and the fundus imaging optical system at the photographing timing of each frame.

  In the following, exemplary embodiments will be described with reference to the drawings. The imaging apparatus 1 is a fundus imaging apparatus with wavefront aberration compensation (AOSLO) that captures a fundus image of the subject's eye while correcting the wavefront aberration of the subject's eye.

  First, a schematic configuration of the photographing apparatus 1 will be described with reference to FIG. The photographing apparatus 1 includes a base 510, a face support unit 600, and a photographing unit 500. The face support unit 600 is attached to the base 510. The imaging unit 500 houses an optical system described later, and is provided on the base 510. The face support unit 600 is provided with a chin rest 610. The chin rest 610 is moved in the left / right direction (X direction), the up / down direction (Y direction), and the front / rear direction (Z direction) with respect to the base of the face indicating unit 600 by operating a chin rest driving means (not shown).

  Next, the optical system of the photographing apparatus 1 will be described with reference to FIG. The imaging apparatus 1 of the present embodiment includes a fundus imaging optical system 100, a wavefront aberration detection optical system (hereinafter referred to as an aberration detection optical system) 110, aberration compensation units 20, 72, and a second imaging unit 200. A tracking unit (position detection unit) 300 and an anterior ocular segment observation unit 700.

  The fundus imaging optical system 100 projects laser light (illumination light) on the eye E and receives reflected light from the fundus of the laser light to capture a fundus image of the eye E. The fundus of the eye E is photographed by the fundus imaging optical system 100 with high resolution (high resolution) and high magnification. As described below, the fundus imaging optical system 100 may have a configuration of a scanning laser ophthalmoscope using a confocal optical system, for example. The fundus imaging optical system 100 includes a first illumination optical system 100a and a first imaging optical system 100b. In the present embodiment, the aberration compensation units 20 and 72 are arranged in the fundus imaging optical system 100 in order to correct the aberration of the eye to be examined. The aberration compensation unit includes a diopter correction unit 20 for correcting low-order aberrations (diopter: for example, spherical power) of the eye to be examined and high-order aberration compensation for correcting high-order aberrations of the eye to be examined. Part (wavefront compensation device) 72.

  The first illumination optical system 100a illuminates the fundus two-dimensionally by irradiating the eye E with laser light and scanning the laser light on the fundus. The first illumination optical system 100a includes a light source 11, a lens 12, a polarization beam splitter (PBS) 14, a beam splitter (BS) 71, a concave mirror 16, in an optical path from the light source 11 (first light source) to the fundus. Concave mirror 17, flat mirror 18, aberration compensation unit 72 (wavefront compensation device 72), beam splitter (BS) 75, concave mirror 21, concave mirror 22, scanning unit 25, concave mirror 26, concave mirror 27, flat mirror 31, Lens 32, plane mirror 33, aberration compensation unit 20 (diopter correction unit 20), plane mirror 35, concave mirror 36, deflection unit 400, dichroic mirror 90, concave mirror 41, plane mirror 42, plane mirror 43, and concave surface A mirror 45.

  The light source 11 emits laser light. In the present embodiment, the laser light has a near-infrared wavelength that is difficult to be visually recognized by the eye to be examined. For example, in the present embodiment, the light source 11 is an SLD (Super Luminescent Diode) having a wavelength of 840 nm. The light source 11 may be any light source that emits spot light having a high convergence property, and may be, for example, a semiconductor laser.

  The laser light emitted from the light source 11 is converted into parallel light by the lens 12 and then enters the wavefront compensation device 72 via the PBS 14, BS 71, concave mirrors 16 and 17, and the flat mirror 18. In the present embodiment, the laser light passes through the PBS 14 and becomes a light beam having only an S-polarized component. The wavefront compensation device 72 corrects higher-order aberrations of the eye to be examined by controlling the wavefront of the incident light. The detailed configuration of the wavefront compensation device 72 will be described later. In the present embodiment, the laser beam is guided from the wavefront compensation device 72 to the BS 75, reflected by the concave mirror 21 and the concave mirror 22, and travels toward the scanning unit 25.

  In the present embodiment, the scanning unit 25 is used together with the deflecting unit 400 in order to scan laser light two-dimensionally on the fundus. The scanning unit 25 is a resonant mirror used for main scanning of laser light. The laser light is scanned in the X direction on the fundus by the scanning unit 25.

  The light that has passed through the scanning unit 25 enters the diopter correction unit 20 via the concave mirrors 26 and 27, the plane mirror 31, the lens 32, and the plane mirror 33.

  The diopter correction unit 20 is a unit for performing diopter correction. The diopter correction unit 20 includes a pair of lenses and plane mirrors in addition to the drive unit 20a. The optical path length is adjusted by moving the plane mirror and lens of the diopter correction unit 20 in a predetermined direction by the drive unit 20a. As a result, the diopter error of the eye E is corrected.

  The illumination light guided from the diopter correction unit 20 to the plane mirror 35 is reflected by the concave mirror 36 and travels toward the deflection unit 400.

  The deflection unit 400 scans the laser light emitted from the light source 11 in the vertical direction (Y direction) on the fundus. Furthermore, the deflecting unit 400 is also used to move the scanning range of the laser light on the fundus. For example, in the present embodiment, the deflecting unit 400 may include two optical scanners (specifically, an X galvanometer mirror and a Y galvanometer mirror) having different directions of deflecting laser light.

  The light that has passed through the deflecting unit 400 is guided into the pupil of the eye E through the dichroic mirror 90, the concave mirror 41, the plane mirrors 42 and 43, and the concave mirror 45. The laser light is collected on the fundus of the eye E. As described above, the laser beam is two-dimensionally scanned on the fundus by the operations of the scanning unit 25 and the deflecting unit 400.

  Further, the dichroic mirror 90 has a characteristic of transmitting light beams from the second photographing unit 200 and tracking unit 300 described later and reflecting light beams from the light source 11 and the light source 76 described later. Note that the emission ends of the light sources 11 and 76 and the fundus of the eye E are conjugate. In this way, the first illumination optical system 100a is formed.

  Next, the first photographing optical system 100b will be described. The first imaging optical system 100 b receives the reflected light of the laser light irradiated on the fundus by the light receiving element 56. The imaging apparatus 1 acquires a first fundus image (in this embodiment, an AO-SLO image) based on a signal from the light receiving element 56. The first imaging optical system 100b shares the optical path from the eye E to the BS 71 with the first illumination optical system 100a. The first imaging optical system 100 includes elements arranged in the reflection side optical path of the BS 71, that is, a plane mirror 51, a PBS 52, a lens 53, a pinhole plate 54, a lens 55, and a light receiving element 56. . In this embodiment, the light receiving element 56 is an APD (avalanche photodiode). Further, the pinhole plate 54 is placed at a position conjugate with the fundus.

  The fundus reflection light of the laser light from the light source 11 traces the first illumination optical system 100 a in the reverse direction, is reflected by the BS 71 and the flat mirror 51, and only the S-polarized light is transmitted by the PBS 52. This transmitted light is focused on the pinhole of the pinhole plate 54 via the lens 53. The reflected light focused at the pinhole is received by the light receiving element 56 through the lens 55. A part of the illumination light is reflected on the cornea, but most of the illumination light is removed by the pinhole plate 54. Therefore, the light receiving element 56 can receive the reflected light from the fundus while suppressing the influence of corneal reflection.

  The first fundus image is acquired by processing the light reception signal of the light receiving element 56 by an image processing unit (for example, the control unit 800). In this embodiment, a fundus image of one frame is formed by main scanning of the scanning unit 25 and sub-scanning of a Y-scan galvanometer mirror provided in the deflection unit 400. Note that the deflection angle (swing angle) of the mirror in the scanning unit 25 and the deflection unit 400 is determined so that the angle of view of the fundus image (fundus image) acquired by the first imaging unit 100 becomes a predetermined angle. Here, in order to observe and photograph a predetermined range of the fundus at a high magnification (here, observation at a cell level or the like), the angle of view is set to about 1 to 5 degrees. In this embodiment, the angle is 1.5 degrees. Although depending on the diopter of the eye to be examined, the imaging range of the first fundus image is about 500 μm square.

  Further, when the reflection angle of the X-scanning galvanometer mirror and the Y-scanning galvanometer mirror provided in the deflection unit 400 is moved larger than the imaging field angle of the first fundus image, the first fundus image is captured on the fundus. The position (that is, the scanning range of the laser beam) is changed.

  The second photographing unit 200 is a unit for obtaining a fundus image (second fundus image) having a wider angle of view than the angle of view of the first photographing unit 100. The second fundus image is used, for example, as an image for position designation and position confirmation for obtaining the first fundus image. The second imaging unit 200 of the present embodiment preferably has a configuration capable of acquiring and observing a fundus image of the eye E to be examined in real time with a wide angle of view (for example, about 20 to 60 degrees). For example, as the second imaging unit 200, an observation / imaging optical system of an existing fundus camera and an optical system and a control system of a scanning laser ophthalmoscope (SLO) may be used.

  The tracking unit 300 detects a temporal change in positional deviation due to fixation eye movement of the eye E and obtains movement position information. The tracking unit 300 sends the light reception result obtained at the start of tracking as reference information to the control unit 800, and then sequentially transmits the light reception result (light reception information) obtained for each scan to the control unit 800. The control unit 800 compares the received light information obtained thereafter with the reference information, and obtains the movement position information by calculation so that the same received light information as the reference information is obtained. The control unit 800 drives the deflection unit 400 based on the obtained movement position information. By performing such tracking, the deflection unit 400 is driven so that even if the subject eye E moves slightly, the movement of the deflection unit 400 is performed, and thus the movement of the fundus image displayed on the monitor 850 is suppressed. It becomes. The dichroic mirror 91 has a characteristic of transmitting the light beam from the second photographing unit 200 and reflecting the light beam from the tracking unit 300.

  The anterior segment observation unit 700 is a unit that illuminates the anterior segment of the eye E with visible light and captures a front image of the anterior segment. An image captured by the anterior segment observation unit 700 is output to the monitor 850. The anterior ocular segment image acquired by the anterior ocular segment observation unit 700 is used for alignment between the imaging unit 500 and the eye E to be examined. The dichroic mirror 95 has a characteristic of transmitting the light flux from the second imaging unit 200 and the tracking unit 300 and reflecting the light flux from the anterior ocular segment observation unit 700.

Next, the aberration detection optical system 110 will be described. The aberration detection optical system 110 includes a wavefront sensor 73. The aberration detection optical system 110 projects measurement light onto the fundus of the eye E, and receives (detects) the fundus reflection light of the measurement light as an index pattern image with the wavefront sensor 73. The aberration detection optical system 110 has a part of optical elements on the optical path of the first illumination optical system 100a and the first photographing optical system 100b (in the present embodiment, on the common optical path), and the optical paths of the optical systems 100a and 100b. Some are shared. That is, the aberration detection optical system 110 of the present embodiment shares the BS 71 to the concave mirror 45 arranged on the optical path of the optical systems 100a and 100b with the optical systems 100a and 100b. Further, the aberration detection optical system 110 includes a light source 76, a lens 77, PBS 78, BS 75, BS 71, a dichroic mirror 86, a PBS 85, a lens 84, a plane mirror 83, and a lens 82.

  The light source 76 is used for detecting the aberration of the eye E. In the present embodiment, the light source 76 emits light having a wavelength different from that of the light source 11. As an example, in this embodiment, a laser diode that emits laser light having a wavelength of 780 nm is used as the light source 76 as measurement light. The measurement light emitted from the light source 76 is converted into a parallel light beam by the lens 77 and then incident on the PBS 78.

  The PBS 78 is an example of first polarization means provided in the wavefront compensation unit. The PBS 78 polarizes the light emitted from the light source 76 in a predetermined direction. More specifically, the PBS 78 is polarized in a direction (ie, P-polarized light) orthogonal to the polarization direction (ie, S-polarized light) of the PBS 14. The light that has passed through the PBS 78 is reflected by the BS 75 and guided to the optical path of the first illumination optical system 100a. As a result, the measurement light is condensed on the fundus of the eye E through the optical path of the first illumination optical system 100a.

  The measurement light is reflected at the condensing position of the fundus (for example, the retina surface). The fundus reflection light of the measurement light follows the optical path of the first illumination optical system 100a (that is, the optical path of the first photographing optical system 100b) in the opposite direction to that during projection. On the way, the measurement light is reflected by the wavefront compensation device 72. Thereafter, the measurement light is reflected by the BS 71, thereby deviating from the optical path of the first illumination optical system 100a. Thereafter, the measurement light is reflected by the dichroic mirror 86 and guided to the wavefront sensor 73 through the PBS 85, the lens 84, the plane mirror 83, and the lens 82.

  The PBS 85 is a second polarization unit provided in the wavefront compensation unit. The PBS 85 is used to guide light to the wavefront sensor 73 by transmitting light polarized in one direction (here, S-polarized light) out of the light irradiated to the eye E from the light source 76. The The PBS 85 blocks a component polarized in a direction orthogonal to the transmitted component (P-polarized light). The dichroic mirror 86 has a characteristic of transmitting light (840 nm) having the wavelength of the light source 11 and reflecting light (780 nm) having the wavelength of the light source 76 for detecting aberration. Therefore, the wavefront sensor 73 detects light having an S-polarized component from the fundus reflection light of the measurement light. In this way, light reflected by the cornea and the optical element is suppressed from being detected by the wavefront sensor 73.

  The wavefront sensor 73 receives fundus reflection light of measurement light for aberration measurement in order to detect wavefront aberration of the eye E. As the wavefront sensor 73, an element that can detect wavefront aberration including low-order aberration and high-order aberration (more specifically, a Hartmann Shack detector, a wavefront curvature sensor that detects a change in light intensity, and the like) is used. May be. In the present embodiment, the wavefront sensor 73 includes, for example, a microlens array composed of a number of microlenses, a two-dimensional imaging element 73a (or a two-dimensional light receiving element) for receiving a light beam that has passed through the microlens array, Have The microlens array of the wavefront sensor 73 is disposed at a position substantially conjugate with the pupil of the eye E to be examined. Further, the imaging surface (light receiving surface) of the two-dimensional imaging element 73a is disposed at a position substantially conjugate with the fundus of the eye E to be examined.

  An index pattern image 61 (Hartmann image in this embodiment) is formed on the imaging surface of the two-dimensional imaging element 73a by the light beam that has passed through the microlens array (not shown). Therefore, the fundus reflection light passes through the microlens array and is received by the two-dimensional imaging element 73a, thereby being imaged as a Hartmann image (dot pattern image). In the present embodiment, the aberration information of the eye to be examined is acquired from the Hartmann image, and the wavefront compensation device 72 is controlled based on the aberration information. Details of the Hartmann image will be described later.

  The wavefront compensation device 72 is disposed in the optical path of the fundus imaging optical system 100, and compensates the wavefront aberration of the eye E by controlling the wavefront of the incident light. In the present embodiment, a liquid crystal spatial light modulator may be used for the wavefront compensation device 72. As an example, a description will be given below assuming that a reflective LCOS (Liquid Crystal On Silicon) or the like is used. In this case, the wavefront compensation device 72 has predetermined laser light (S-polarized light) from the light source 11, fundus reflected light (S-polarized light) of the laser light, reflected light (S-polarized component) of wavefront aberration detection light, and the like. Are arranged in a direction capable of correcting the aberration with respect to the linearly polarized light (S-polarized light). As a result, the wavefront compensation device 72 can modulate the S-polarized component of the incident light. In the present embodiment, the reflection surface of the wavefront compensation device 72 is arranged at a position that is substantially conjugate with the pupil of the eye to be examined.

  In the wavefront compensation device 72 of the present embodiment, the alignment direction of the liquid crystal molecules in the liquid crystal layer is substantially parallel to the polarization plane of incident reflected light. The predetermined plane on which the liquid crystal molecules rotate according to the change in the voltage applied to the liquid crystal layer includes the incident optical axis and the reflected optical axis of the fundus reflection light with respect to the wavefront compensation device 72, and the mirror layer of the wavefront compensation device 72. And a plane that includes the normal line.

  In this embodiment, the wavefront compensation device 72 is a liquid crystal modulation element, and in particular, a reflective LCOS or the like is used, but the present invention is not limited to this. Other reflective wavefront compensation devices may be used. For example, a deformable mirror that is a form of MEMS (Micro Electro Mechanical Systems) may be used. Further, a transmissive wavefront compensation device may be used instead of the reflective wavefront compensation device. In the transmission type device, the wavefront aberration is compensated by transmitting the reflected light from the fundus.

  In the above description, a light source that emits illumination light having a wavelength different from that of the first light source is used as the aberration detection light source. However, the first light source may also serve as the aberration detection light source.

  In the present embodiment described above, the wavefront sensor and the wavefront compensation device are the pupil conjugate of the eye to be examined. However, the position may be a position that is substantially conjugate with a predetermined part of the anterior eye portion of the eye to be examined. There may be.

  In the present embodiment described above, the wavefront sensor and the wavefront compensation device are the pupil conjugate of the eye to be examined. However, the position may be a position that is substantially conjugate with a predetermined part of the anterior eye portion of the eye to be examined. There may be.

  Next, with reference to FIG. 3, a control system of the photographing apparatus 1 in the present embodiment will be described. The photographing apparatus 1 has a control unit 800. The control unit 800 is a processor (for example, a CPU) that controls the entire apparatus of the photographing apparatus 1. In the present embodiment, a storage unit 801, an operation input unit 802, an image processing unit 803, and a monitor 850 are electrically connected to the control unit 800. The control unit 800 includes a light source 1, a driving unit 10a, a scanning unit 15, a light receiving element 56, a wavefront compensation device 72, a wavefront sensor 73, a light source 76, a second imaging unit 200, a tracking unit 300, a deflection unit 400, Are electrically connected.

  The storage unit 801 stores various control programs and fixed data. Further, the storage unit 801 may store an image captured by the imaging device 1, temporary data, and the like.

  The control unit 800 controls each member described above such as the first imaging unit 100 based on the operation signal output from the operation unit 802. The operation unit 802 is connected to a mouse or the like (not shown) as an operation member operated by the examiner.

  The image processing unit 803 forms a fundus image to be examined with different angles of view, that is, a first fundus image and a second fundus image, on the monitor 850 based on signals received by the light receiving element 56 and the second imaging unit 200.

  The monitor 850 may be a display monitor mounted on the photographing apparatus 1 or a general-purpose display monitor that is separate from the photographing apparatus 1. Moreover, the structure in which these were used together may be sufficient. The monitor 850 can display fundus images (first fundus image and second fundus image) captured by the imaging apparatus 1 as moving images and still images.

  The control of the wavefront compensation device 72 by the controller 800 is executed based on the wavefront aberration detected by the wavefront sensor 73. In the present embodiment, feedback control of the wavefront compensation device 72 based on the detection signal from the wavefront sensor 73 is performed. By controlling the wavefront compensation device 72, the high-order aberrations of the laser light emitted from the light source 1 and the fundus reflection light of the laser light are removed together with the S-polarized component of the reflected light of the light source 76. In this way, the aberrations of the laser light emitted from the light source 1 and the fundus reflection light of the laser light are removed. As a result, the imaging apparatus 1 acquires a high-resolution first fundus image (that is, a wavefront compensated image) from which high-order aberrations of the eye E have been removed (wavefront compensated). At this time, the low-order aberration is corrected by the diopter correction unit 10.

  The operation of this apparatus having the above configuration will be described with reference to the flowchart shown in FIG. The imaging apparatus 1 operates according to a control program stored in the storage unit 801 after the power is turned on. In the following operation, the imaging apparatus 1 captures the first fundus image (AOSLO image) a plurality of times at the same position of the eye E. Moreover, the imaging device 1 may perform image processing (for example, analysis processing (for example, measurement of cell density)) regarding cells forming the retina using a plurality of first fundus images captured at the same position. .

  First, the imaging unit 500 is aligned with the eye E (S1). For example, the control unit 800 acquires a real-time anterior segment image using the anterior segment observation unit 700 and causes the monitor 850 to display the anterior segment image. While observing the anterior ocular segment image on the monitor 850, the examiner manually or automatically adjusts the position of the chin rest 610 to perform rough alignment. In this case, the examiner instructs the subject to fixate a fixation target (not shown).

  In the present embodiment, after the rough alignment is completed, when the start switch provided in the operation unit 802 is operated by the examiner, acquisition of the fundus image using the first imaging unit 100 and the second imaging unit 200 is started. (S2).

  In this case, first, the diopter correction of each of the photographing units 100 and 200 may be performed by the control unit 800 (S3). For example, diopter correction in the first photographing unit 100 is performed using the diopter correction unit 10.

  The imaging position on the fundus of the eye E (that is, the acquisition range of the first fundus image (AOSLO image)) by the first imaging unit 100 may be set (S5). In this case, the imaging position may be set to a position desired by the examiner. For example, the control unit 800 displays a wide-area second fundus image acquired by the second imaging unit 200 on the monitor 850 and displays an enlarged image (that is, the first fundus image) in the second fundus image. The location to be acquired is received via the operation input unit 802. The control unit 800 determines a driving range (that is, a scanning range) of the deflection unit 400 based on a signal from the operation input unit 802. As a result, the approximate photographing position of the first fundus image is determined.

  By the way, even if the fixation is performed well, the eyes are constantly moving. Therefore, when the first fundus image (that is, the AOSLO image) is captured a plurality of times for the same part, it is preferable to consider the eye movement.

  Therefore, first, in the present embodiment, a reference image is selected (S6). The reference image is a first fundus image of the eye E and is a still image in which a reference region (for example, ROI 901) described later is set (see FIG. 5). The reference image may be an image acquired based on an image acquisition trigger signal, or may be an image stored in advance in the storage unit 801. In this case, for example, a first fundus image may be newly taken in order to obtain a reference image. The reference image may be an averaged image composed of a plurality of first fundus images. When the first fundus image of the eye to be examined is stored in the storage unit 801 in advance, the first fundus image of the eye to be examined that has been taken in the past may be used as the reference image. In any case, when there are a plurality of reference image candidates, for example, the control unit 800 may select an image designated by the examiner via the operation unit 802 as the reference image.

  When the reference image is selected, the ROI 901 is set for the reference image (S7). The ROI 901 is used, for example, as a reference area for selecting and selecting the first fundus image that is repeatedly formed. The ROI 901 is set at a specific position on the reference image, for example. The ROI 901 may be formed from a partial image area on the reference image. The setting of the ROI 901 may be automatically set by the control unit 800, or may be set manually by the examiner via the operation unit 802.

  In other words, the image information of the ROI 901 in the reference image is used as a template for determining the presence or absence of an image of the same tissue as the ROI 901 in the other first fundus image. In the present embodiment, in the process of S7, it is assumed that a predetermined size is set in the center of the reference image. In the present embodiment, the cell image forming the retina in the fundus image is set as the ROI 901. The cell image is an image showing an arrangement structure of a large number of cells or tissues forming the retina. With the arrangement structure, it is possible to accurately distinguish from other tissues. Cells that form the retina can include photoreceptors, pigment epithelial cells, and ganglion cells. In the present embodiment, the ROI 901 uses a photoreceptor cell image showing an array structure of photoreceptor cells. In the case where a photoreceptor cell image is used as the ROI 901, if an image 902 of a blood vessel or its shadow is included, the reliability of image processing described later is lowered. In this case, the control unit 800 may select a reference image so that a cell image that does not include a blood vessel image (here, a photoreceptor cell image) is set as the ROI 901. The control unit 800 may select a cell image that does not include a blood vessel image at the set position of the ROI 901 as the reference image. In the present embodiment, the ROI 901 may be preferably set as a cell image having an area of 1/8 or more of the fundus image with respect to a 100 μm square fundus image. In addition, when a photoreceptor cell image is used as the ROI 901, the photoreceptor cell may preferably be contained in a range of 50% or more of the area of the ROI 901.

  Once the ROI 901 is set, the control unit 800 then continuously captures the same position (S8). Here, the operation of the apparatus in S8 will be described in more detail with reference to FIG.

  First, the first fundus image for one frame is sequentially formed (sequentially acquired) by the control unit 800 (S21). Next, the control unit 800 determines whether or not to save the acquired image (in this embodiment, the image is stored in the storage unit 801).

  In the present embodiment, for example, the control unit 800 extracts the ROI 901 from the reference image, and determines by image processing whether or not the image region corresponding to the extracted ROI 901 is included in each first fundus image. Also good. In this case, for example, the first fundus image determined to include the image region corresponding to the ROI 901 is selected, and the first fundus image determined to include the image region corresponding to the ROI 901 is excluded. The selected image is stored in the storage unit 801 as a still image, and the excluded image is discarded (for example, data is deleted and not stored in the storage unit 801). At this time, the image data to be excluded may be erased without being stored in a storage device (for example, the storage unit 801) such as an HDD, or after the data is temporarily stored in the storage unit 801, the data may be deleted. It may be erased. Further, it is not always necessary to delete the data of the excluded image, and it may be stored in the storage unit 801 as excluded image data.

  More specifically, for example, the control unit 800 detects the correlation (degree of coincidence) between the ROI 901 and another first fundus image (in the present embodiment, the image acquired in the process of S21 immediately before) by image processing. Then, the presence / absence of an image area corresponding to the ROI 901 is determined (S22). Here, the correlation of the other first fundus image with respect to the ROI 901 is used. The correlation between the ROI 901 of the reference image and other fundus images is calculated by image processing. For example, the correlation can be detected using various image processing techniques. For example, the control unit 800 translates the ROI 901 of the base image one pixel at a time relative to the other first fundus image (reference image) or rotates it by a small angle, etc. The position at which the image data of the images match (position where the correlation is highest) is obtained. Then, the control unit 800 determines whether or not there is an image area corresponding to the ROI 901 depending on whether or not the correlation exceeds a predetermined threshold.

  Note that the method of obtaining the correlation is not necessarily limited to this, and the correlation with other images may be obtained using image information of the entire reference image. In addition, various image processing methods (a method using various correlation functions, a method using Fourier transform, and a method based on feature point matching) can be used for correlation detection. In addition, it is possible to use the position shift detection technique between images for the detection of a correlation.

  Next, based on the correlation acquired by the process of S22, a first exclusion determination is performed (S23). In the present embodiment, the first exclusion determination is performed in order to select a first fundus image including an image region corresponding to the reference region. When the correlation acquired in the process of S22 is lower than the threshold value, the control unit 800 excludes the first fundus image from the selection target. That is, when it is determined that there is an image area corresponding to the ROI 901, it is set as a selection target, and when it is determined that there is no image area corresponding to the ROI 901, it is excluded from the selection target.

  Next, the control unit 800 may measure the moving speed of the fundus between the fundus images by comparing the image areas corresponding to the ROI 901 in each first fundus image (S24). Then, the control unit 800 may select each fundus image that is repeatedly formed based on the measurement result.

  The moving speed is acquired by, for example, obtaining a positional shift between fundus images acquired adjacent to each other. In a plurality of adjacent first fundus images acquired by the control unit 800, the position (position coordinates) of the image region corresponding to the ROI 901 in one first fundus image (first reference image) and the other You may obtain | require the deviation | shift amount with the position (positional coordinate) of the image area | region corresponding to ROI901 in a 1st fundus image (2nd reference image). Here, the amount of positional deviation between images can be detected using various image processing techniques. For example, the position where the correlation with the ROI 901 is highest in the first fundus image is obtained by, for example, translating the ROI 901 of the base image pixel by pixel or rotating it by a slight angle with respect to the first fundus image (reference image). , The position of the image area corresponding to the ROI 901 may be detected. Then, the moving speed is calculated by comparing the position of the image area corresponding to the ROI 901 with respect to the adjacent fundus image. In the present embodiment, as described above, the movement speed of the fundus is measured by comparing two fundus images using the cell arrangement structure.

  More specifically, in S24, the relative velocity between the fundus and the fundus imaging optical system at the timing when the first fundus image (in the present embodiment, the image acquired in the process of S21 immediately before) is acquired is shown. Even if information (speed information) is acquired. The speed information may be information indicating a speed in a direction orthogonal to the photographing optical axis of the fundus imaging optical system. In the present embodiment, as the speed information, a positional shift amount of the image area corresponding to the ROI 901 between frames adjacent to each other at the image formation timing is used. In the present embodiment, this is based on the amount of positional deviation between the image area corresponding to the ROI 901 in the first fundus image acquired immediately before and the image area corresponding to the ROI 901 in the immediately preceding first fundus image.

  A second exclusion determination is performed on the first fundus image using the speed information acquired in the process of S24 (S25). In the second exclusion determination, the first fundus image is captured while the fundus is moving greatly. Although details will be described later, in the present embodiment, images that have been moved by microsaccades in fixation fine movement are excluded from selection targets.

  Next, if it is not determined as an image to be excluded by the processing of S23 and S25, the first fundus image is stored in the storage unit 802 as a selected image (S26).

  Next, the control unit 800 determines whether or not a condition for ending the continuous shooting is satisfied (S27). For example, input of an end instruction from the examiner, completion of storage of a predetermined number of first fundus images, and the like may be conditions for ending the imaging. If the conditions for termination are satisfied (S27: Yes), the control unit 800 next executes the process of S9 (see FIG. 4). On the other hand, if the conditions for termination are not satisfied (S27: No), the control unit 800 returns to the process of S21 and repeatedly executes the processes of S21 to S27. As a result, a plurality of fundus images photographed at the same position are selected, and these images are stored in the storage unit 802.

Here, a case has been described in which only the image information of the ROI 901 is used in the reference image to obtain a positional deviation from other images. However, the present invention is not limited to this, and the image information of the entire reference image is not limited to this. It may be used to determine the positional deviation from other images. In addition, various image processing methods (a method using various correlation functions, a method using Fourier transform, and a method based on feature point matching) can be used for detecting the amount of positional deviation. Here, FIG. 7 and FIG. The 100 first fundus images acquired continuously will be illustrated as an example, and the image selected by the controller 800 will be described in more detail. In the present embodiment, as an example, about 50 first fundus images per second are acquired by the imaging device 1.

  FIG. 7 shows the transition of the positional deviation amount of the photographing position between each first fundus image and the reference image. In FIG. 7, the horizontal axis corresponds to the frame number indicating the shooting order. The vertical axis corresponds to the positional deviation amount. In the present embodiment, a threshold value α in the first exclusion determination is set. The threshold value α is a value smaller than the half value (0.75 deg in the present embodiment) of the angle of view of the first fundus image. As an example, in the present embodiment, 0.1 deg is set as the threshold value α.

  It is determined in the first exclusion determination that a frame whose positional deviation amount with respect to the reference image is larger than the threshold value α should be excluded from the selection target. Therefore, in the example of FIG. 7, it is determined that the first fundus image other than the 40th frame to the 59th frame should be excluded.

  FIG. 8 shows the relative speed transition between the eye to be examined and the optical system based on the positional shift amount between each first fundus image and the first fundus image one frame before the first image. Is shown. In FIG. 8, the horizontal axis corresponds to the frame number indicating the shooting order. The vertical axis corresponds to speed. In FIG. 8, the speed of the nth frame can be obtained by, for example, dividing the amount of positional deviation between the (n−1) th frame and the nth frame by the frame period. In the present embodiment, a threshold value β in the second exclusion determination is set. β is a threshold value for excluding an image distorted by the influence of eye movement from the selection target. For example, in the present embodiment, β is set to exclude an image formed when microsaccade is generated from selection targets. In this case, β is preferably a value less than or equal to the microsaccade speed. In the present embodiment, a value of 1.5 deg / sec is used as β. There are various theories of the speed of microsaccades based on literature and papers. In the present embodiment, β is set so that an image formed during the movement can be excluded from the selection target even if the movement is relatively slow among various theories. Note that β may be a threshold for suppressing the influence of eye movements other than microsaccade. For example, it may be set based on the speed of fixation fine movement other than microsaccade, or may be set based on the speed of nystagmus other than fixation fine movement. Note that the nystagmus other than the fixed eye movement may include, for example, a psychological movement that causes the gaze direction to follow the light projected from the imaging device 1 to the eye E to be examined.

  Frames whose moving speed is larger than the threshold value β are excluded from selection targets in the second exclusion determination. In FIG. 8, for example, it is shown that the eye moved greatly between 39 frames and 40 frames and between 59 frames and 60 frames. That is, it is considered that the eye moved by the microsaccade between 39 frames and 40 frames and between 59 frames and 60 frames. As a result, 39 frames and 40 frames, and 59 frames and 60 frames are likely to have large distortion due to eye movement. Therefore, it is determined that these first fundus images should be excluded from selection targets in the second exclusion determination.

  Therefore, in the examples of FIGS. 7 and 8, images of 41 frames to 59 frames are selected as a result of the first determination process and the second determination process. Other images are excluded from the selection targets. As a result, among the plurality of first fundus images captured continuously, a favorable first fundus image with a small positional deviation of the photographing position with respect to the reference image is selected.

  Returning to FIG. 4, the description will be continued. In the present embodiment, after the end of S8, the control unit 800 executes a photoreceptor cell analysis process as an image process related to the fundus cells using the first fundus image selected in S8 (S9). In the photoreceptor cell analysis process (S9), for example, an addition average image is generated by adding and averaging the plurality of first fundus images selected in S8. Further, the control unit 800 detects photoreceptor cells from the addition average image. Furthermore, information on the density of photoreceptor cells (for example, photoreceptor cell density, hexagonal cell appearance rate, etc.) is calculated based on the photoreceptor cell detection result.

  As described above, the imaging apparatus 1 according to the present embodiment includes the same image area as the ROI 901 filled with photoreceptor cells in a plurality of first fundus images formed in succession, and positional deviation between adjacent frames. Is selected as the selected image. Therefore, an image with little distortion due to eye movement is selected as the selected image. The selected image is used for image processing relating to cells of the fundus. As a result, highly reliable image processing (for example, analysis processing, image composition processing, etc.) results (for example, analysis results, image composition results, etc.) are obtained.

  In particular, in the present embodiment, positional deviation detection is performed when a desired image is obtained using a cell image showing an array structure of a large number of cells or tissues forming the retina as the ROI 901. Thereby, in the present embodiment, better positional deviation measurement accuracy is realized as compared with the case where the positional deviation is obtained using the running state of the blood vessel (for example, the intersection of the blood vessels). Therefore, image processing can be performed on a selected image that is selected favorably.

In the present embodiment, a fundus image having a small positional shift between adjacent frames as compared to a threshold value for excluding an image formed when microsaccade occurs is selected as an image processing target.
Therefore, the image distorted by the microsaccade can be favorably excluded from the image processing target. Since microsaccades are fast movements among various eye movements, it is particularly difficult to obtain a good cell level image at the timing of their generation. On the other hand, in the present embodiment, the first fundus image acquired at the microsaccade generation timing is automatically excluded from the image processing target, so that the reliability of the image processing can be improved satisfactorily. .

  As mentioned above, although demonstrated based on embodiment, this indication is not limited to the said embodiment, A various deformation | transformation is possible.

  For example, in the above-described embodiment, when the fundus has moved greatly since the reference image was captured, or when blinking occurred during imaging (in the present embodiment, during scanning of laser light), the first fundus image There may be a case where the reference position is not included in. In addition, even when the first fundus image is distorted due to the movement of the fundus during photographing, a region having a high degree of coincidence with the ROI 901 of the reference image may not be detected from the first fundus image. Therefore, in this embodiment, there may be a case where the positional deviation information cannot be acquired by the process of S22. On the other hand, the control unit 800 may also exclude the first fundus image for which position shift information has not been acquired from the selection targets.

  In addition, tracking by the tracking unit 300 may be performed while the first fundus image is continuously captured. The tracking unit 300 may perform tracking using the wide-angle second fundus image captured by the second imaging unit 200. In this tracking, the deflection units 400 and 410 are driven so that even if the eye E is moved slightly, the movement is canceled out. For the detailed configuration of the tracking unit 300 and the operation of the entire apparatus in this case, refer to, for example, Japanese Patent Application Laid-Open No. 2011-115301. Note that, at the timing when the deflection units 400 and 410 are driven by tracking, the control unit 800 may not acquire the first fundus image. In addition, when the first fundus image is acquired also at the timing, the image may be excluded from the selection target.

  Moreover, although the said embodiment demonstrated the case where an image process was performed immediately after the selection image was memorize | stored in the memory | storage part 801, it is not necessarily restricted to this. For example, when the selected image is stored in the storage unit 801, the processing related to imaging is terminated (for example, automatic imaging), and then stored in the storage unit 801 at an arbitrary timing based on an instruction from the examiner. A configuration may be employed in which image processing is performed on a selected image.

  In the above-described embodiment, the selected image is selected based on the amount of positional deviation between two adjacent frames. However, the positional shift between images corresponding to the eye movement speed may be obtained based on the positional shift amount between each image and an image formed before or after a certain frame with respect to each image. It is not limited to. For example, an image may be selected based on the amount of positional deviation from the image before or after two frames or more of each image.

  In the above embodiment, the case where the image is selected in parallel with the continuous photographing of the first fundus image has been described, but the present invention is not necessarily limited thereto. The photographing of the first fundus image and the selection of the image may be performed at different timings. For example, the image selection may be performed on the result of continuous shooting stored in the storage unit 801 in advance.

  In the above embodiment, the ROI 901 is set at a certain position (for example, the center of the image) of the reference image. However, it is not necessarily limited to this. For example, the control unit 800 may detect a region having a certain size filled with a cell image in a certain first fundus image, and set the region as the ROI 901. In this case, the control unit 800 performs processing for setting the ROI 901 while avoiding the blood vessel image. In this case, the ROI can be set according to the cell distribution state in the image.

56 Photodetector 72 Wavefront Compensation Device 73 Wavefront Sensor 100 Fundus Imaging Optical System 110 Wavefront Aberration Detection Optical System 800 Control Unit 801 Storage Unit 802 Operation Input Unit 850 Monitor 901 ROI

Claims (3)

A fundus imaging optical system for capturing a fundus image by receiving reflected light from the fundus in a state where the wavefront aberration is compensated by the wavefront compensation device,
An image forming means for forming the fundus image based on a light receiving signal from the light receiving element;
A cell image forming a retina in the fundus image is set as a reference region, and an image region corresponding to the reference region included in each fundus image repeatedly formed by the image forming unit is compared between the fundus images. A moving speed measuring means for measuring the moving speed of the fundus of the eye,
A fundus imaging apparatus with wavefront compensation , comprising: image selection means for selecting each fundus image repeatedly formed by the image forming means based on a measurement result by the moving speed measurement means . An image processing means for performing addition processing or photoreceptor cell density analysis processing on the fundus image selected by the image selection means from among a plurality of fundus images repeatedly formed by the image forming means. Item 1. A fundus photographing apparatus with wavefront compensation according to Item 1. Image determining means for determining by image processing whether or not an image area corresponding to a reference area on the fundus image set in advance as a reference image is included in each fundus image repeatedly formed by the image forming means; With
The fundus photographing with wavefront compensation according to claim 1, wherein the moving speed measuring unit measures a moving speed between fundus images determined by the image determining unit to include an image region corresponding to the reference region. apparatus.

Images