JP2019042375A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

To generate a proper EnFace image.SOLUTION: An image processing apparatus comprises: EnFace image generation means which generates an EnFace image on the basis of a three-dimensional image of a subject eye; target range selection means which selects a target range for generating the EnFace image; and projection method determination means which determines a projection method for generating the EnFace image in response to the target range.SELECTED DRAWING: Figure 3

Description

  The disclosure of the present specification relates to an image processing apparatus, an image processing method, and a program.

  In recent years, angiography called OCT angiography (hereinafter referred to as OCTA) using OCT (Optical Coherence Tomography) has been proposed (Patent Document 1).

  In OCTA, an OCTA image (Enface image) of an arbitrary depth range can be obtained by projecting only motion contrast data of a part of the depth range among three-dimensional motion contrast data on a two-dimensional plane.

  Various ranges are used as the depth range. For example, according to the distribution of blood vessels in the retina, the superficial layer of the retina (Superficial Capillary Plexus), the deep layer of the retina (Deep Capillary Plexus), the outer layer of the retina (Outer Retina), radial peripapillar capillaries (Radial Peripapillar Capillaries), the choroidal capillary plate ( Many types of depth ranges are used, such as Choriocapillaris.

JP, 2015-131107, A

  Various depth ranges are used to generate Enface images, but when projecting with a specific projection method, depending on the depth range, a preferable image may not be obtained.

  The disclosure of the present specification aims to generate an appropriate Enface image.

  In addition, it is not only the said objective but it is an effect derived by each structure shown to the form for implementing the invention mentioned later, Comprising: It is also another object of this indication to have an effect which can not be obtained by a prior art. It can be positioned as one of

  An image processing apparatus disclosed in the present specification includes an acquisition unit for acquiring a three-dimensional image of an eye to be examined, an EnFace image generation unit for generating an EnFace image based on the three-dimensional image, and an object for generating the Enface image. A target range selecting means for selecting a range, and a projection method determining means for determining a projection method for generating the Enface image according to the target range.

  According to the disclosure of the present specification, it is possible to generate an appropriate Enface image.

FIG. 1 is a diagram illustrating an example of an image processing system. It is a figure showing an example of an image processing device. It is a figure which shows an example of the control block of an image generation part. It is a figure which shows an example of the control block of an image generation part. It is a flowchart which shows an example of a basic process of this specification. It is an example of information buried when projected by AIP. It is an example of the Enface image projected by AIP. . It is an example of the Enface image projected by MIP. 5 is a flowchart illustrating an example of a process according to the first embodiment. 15 is a flowchart illustrating an example of processing in Embodiment 2. FIG. 16 is a flowchart illustrating an example of processing in Embodiment 3. FIG. FIG. 18 is a diagram illustrating an example of a control block of the third embodiment. 5 is an example of a tomogram used in Example 1;

First Embodiment
A preferred embodiment will be described. FIG. 1 is a diagram showing an image processing system according to the present embodiment. The image processing system shown in FIG. 1 includes a light interference unit 100, a scanning optical system 200, an image processing apparatus 300, a monitor 310, a pointing device 320, and a keyboard 321. The image processing apparatus 300 may be configured integrally with the light interference unit 100 and the scanning optical system 200, or may be configured separately. The light interference unit 100 and the scanning optical system 200 constitute an OCT.

  The light interference unit 100 includes a light source 101 that is a low coherence light source that emits near-infrared light. The light emitted from the light source 101 propagates through the optical fiber 102 a and is branched into the measurement light and the reference light by the light branching unit 103. The measurement light branched by the light branching unit 103 is incident on the optical fiber 102 b and guided to the scanning optical system 200. On the other hand, the reference light branched by the light branching unit 103 is incident on the optical fiber 102 c and is guided to the reflection mirror 113.

  The reference light incident on the optical fiber 102 c is emitted from the end of the fiber, is incident on the dispersion compensation optical system 112 via the collimating optical system 111, and is guided to the reflection mirror 113. The reference light reflected by the reflection mirror 113 follows the reverse optical path and enters the optical fiber 102c again. The dispersion compensation optical system 112 corrects the dispersion of the scanning optical system 200 and the optical system of the eye to be examined E which is the object to be measured. The reflection mirror 113 is configured to be drivable in the optical axis direction by the optical path length control unit 114 (not shown), and can change the optical path length of the reference light relative to the optical path length of the measurement light . On the other hand, the measurement light incident on the optical fiber 102b is emitted from the end of the fiber. The light source 101 and the optical path length control unit 114 are controlled under the control of the control unit 130 (not shown).

  Next, the scanning optical system 200 will be described. The scanning optical system 200 is an optical system configured to be movable relative to the eye E. The drive control unit 205 (not shown) of the scanning optical system is configured to be able to drive the scanning optical system 200 in the vertical and horizontal directions with respect to the eye axis of the eye E.

  The light emitted from the fiber end of the optical fiber 102 b is substantially collimated by the optical system 202 and enters the scanning unit 203. The scanning unit 203 has two galvano mirrors capable of rotating the mirror surface, one deflects light in the horizontal direction, the other deflects light in the vertical direction, and the light incident under the control of the drive control unit 205 Bias. As a result, the scanning unit 203 can scan in two directions of the main scanning direction in the plane of the drawing and the sub-scanning direction in the direction perpendicular to the plane of the drawing. The light scanned by the scanning unit 203 forms an illumination spot on the eye E via the lens 204. When the scanning unit 203 receives in-plane deflection, each illumination spot moves on the eye E. The reflected light at this illumination spot position travels along the reverse optical path, enters the optical fiber 102 b, and returns to the light branching portion 103.

  As described above, the reference light reflected by the reflection mirror 113 and the measurement light reflected from the eye E are returned to the light branching portion 103 as return light to generate light interference. The light interfering with each other passes through the optical fiber 102 d, and the interference light emitted to the lens 122 is substantially collimated and enters the diffraction grating 123. The diffraction grating 123 has a periodic structure, and disperses the input interference light. The dispersed interference light is imaged on the line sensor 125 by the imaging lens 124 capable of changing the focusing state. The line sensor 125 is connected to the image processing apparatus 300. The configuration of the OCT is not limited to the above-described example, and it is acceptable as long as a tomographic image of the subject's eye can be imaged.

  FIG. 2 is a diagram for explaining the image processing apparatus 300. As shown in FIG. 2, the image processing apparatus 300 includes a reconstruction unit 301 as means for generating tomographic image data. The present embodiment is a Fourier domain method, and the reconstruction unit 301 generates tomographic image data of an eye to be examined by performing wave number conversion and Fourier transformation on output data of the line sensor 125. In the present embodiment, the image processing apparatus is provided with the light interference unit 100 of the Fourier domain method, but the image processing apparatus may be provided with the light interference unit of the time domain method.

  The image processing apparatus 300 further includes a motion contrast generation unit 302 for generating motion contrast data from a plurality of tomographic image data.

  The image processing apparatus 300 further includes an image analysis unit 303 that analyzes the generated tomographic image data. The image analysis unit 303 can analyze tomographic image data of an eye to be examined and analyze a structure of the eye to be examined included in the tomographic image data. For example, the image analysis unit 303 can extract layer boundaries included in tomogram data by analyzing the tomogram data using a known method.

  Then, the image generation unit 304 generates an image for display from the generated tomographic image data and motion contrast data, and the control unit 306 outputs the generated image for display to the monitor 310. The storage unit 305 stores definitions of a plurality of depth ranges, a definition of a depth range to be applied by default, and the like, and the image generation unit 304 generates an OCTA image according to the depth ranges acquired from the storage unit 305.

  Here, the image processing apparatus includes a CPU (not shown), and the CPU executes, for example, a program stored in the ROM on the RAM, whereby the reconstruction unit 301, the motion contrast generation unit 302, the image analysis unit 303, and the image It functions as the generation unit 304. Note that the image processing apparatus may have a plurality of CPUs, and a plurality of ROMs may be used to record programs. That is, when at least one or more processors and at least one memory are connected, and at least one or more processors execute a program stored in at least one or more memories, the image processing apparatus 300 executes the reconstruction unit 301, It functions as functioning as the motion contrast generation unit 302, the image analysis unit 303, and the image generation unit 304. Note that the processor is not limited to the CPU, and may be a GPU or the like, or a combination of processors of different types such as the CPU and the GPU may be used.

  The CPU controls the entire computer using programs and data stored in the RAM and the ROM. In addition, the functions of each part are realized.

  The RAM includes, for example, an area for temporarily storing programs and data loaded from a storage medium drive, and also includes a work area required for the CPU to perform various processes.

  The ROM generally stores computer programs and setting data.

  Further, a pointing device 320 and a keyboard 321 are connected to the image processing apparatus 300. The pointing device 320 is a mouse provided with a rotatable wheel and a button, and can designate an arbitrary position on the monitor 310. Although a mouse is used as a pointing device in the present embodiment, any pointing device such as a joystick, a touch pad, a track ball, a touch panel, a stylus pen, etc. may be used.

  As described above, the image processing system according to the present embodiment includes the light interference unit 100, the scanning optical system 200, and the image processing apparatus 300.

  Further, at least a part of each part of the image processing apparatus 300 may be realized as an independent apparatus. Alternatively, each function of each unit of the image processing apparatus 300 may be installed in one or more computers and realized by a CPU (not shown) of each computer. In the present embodiment, it is assumed that the functions of the units of the image processing apparatus 300 are installed in the same computer. The image processing apparatus 300 can also be configured as an electric circuit by an image processing board.

  Next, a control method for capturing a tomographic image of an eye to be examined using the present apparatus will be described.

  First, the examiner places a patient in front of the scanning optical system 200 and starts OCT imaging. The light emitted from the light source 101 is divided into measurement light traveling to the subject's eye and reference light traveling to the reference mirror 113 by the light branching means 103 after passing through the optical fiber 102 a.

  The measurement light traveling toward the eye to be inspected passes through the optical fiber 102 b and is emitted from the end of the fiber, is substantially collimated by the optical system 202, and is incident on the scanning means 203. The scanning means 203 has a galvano mirror, and the measurement light deflected by the mirror illuminates the eye through the optical system 204. Then, the reflected light reflected by the eye to be examined follows the reverse path and is returned to the light branching unit 103.

  On the other hand, the reference light traveling toward the reference mirror passes through the optical fiber 102 c and is emitted from the end of the fiber, and reaches the reference mirror 113 through the collimating optical system 111 and the dispersion compensating optical system 112. The reference light reflected by the reference mirror 113 follows the reverse path and is returned to the light branching unit 103.

  The measurement light and the reference light returned to the light branching portion 103 interfere with each other, become interference light, enter the optical fiber 102 d, are substantially collimated by the optical system 122, and enter the diffraction grating 123. The interference light input to the diffraction grating 123 is imaged on the line sensor 125 by the imaging lens 124, and an interference signal at one point on the subject's eye can be obtained.

  The interference signal acquired by the line sensor 125 is output to the image processing apparatus 300. The interference signal output from the line sensor 125 is 12-bit integer data. The reconstruction unit 301 performs wave number conversion, fast Fourier transformation (FFT), and absolute value conversion (acquisition of amplitude) on the 12-bit integer format data, and a tomographic image in the depth direction at one point on the subject's eye Generate data.

  After acquiring the interference signal at one point on the subject's eye, the scanning unit 203 drives the galvano mirror to generate another single point of interference light on the subject's eye. The other interference light is generated as tomographic image data in the depth direction at another point on the subject's eye via the line sensor 125 and the reconstruction unit 301. By repeating this series of control, it is possible to generate one tomographic image data of the eye to be examined.

  Further, the scanning unit 203 drives the galvano mirror, scans the same part of the eye to be examined a plurality of times, and acquires a plurality of tomographic image data on the same part of the eye to be examined. Then, the scanning unit 203 minutely drives the galvano mirror in the sub-scanning direction orthogonal to the main scanning direction, and acquires a plurality of tomographic image data in another part of the eye to be examined. By repeating this control, a plurality of tomographic image data in a predetermined range of the eye to be examined can be acquired.

  In the above, one tomographic image data at one point of the eye to be examined is acquired by subjecting a set of interference signals obtained from the lie sensor 125 to FFT processing. However, the interference signal may be divided into a plurality of sets, and FFT processing may be performed on each of the divided interference signals to obtain a plurality of tomographic image data from one interference signal. According to this method, it is possible to acquire more tomographic image data than the number of times the same part of the eye to be examined is actually scanned.

  Next, a method of generating motion contrast data from tomographic image data in this image processing apparatus will be described.

  The data in the complex number format generated by the reconstruction unit 301 is output to the motion contrast generation unit 302. First, the motion contrast generation unit 302 corrects the positional deviation of a plurality of tomographic image data at the same position of the eye to be examined.

  Then, the motion contrast generation unit 302 obtains a decorrelation value, for example, by the following equation (1) between the two tomographic image data whose positional deviation has been corrected.

  Here, Axy indicates the amplitude at the position (x, y) of the tomographic image data A, and Bxy indicates the amplitude at the same position (x, y) of the tomographic data B. The resulting decorrelation value Mxy takes a value from 0 to 1, and the value becomes closer to 1 as the difference between the two amplitude values increases.

  Then, a plurality of decorrelation values are obtained by repeating the above-described decorrelation operation for the number of acquired tomographic image data, and an average value of the plurality of decorrelation values is obtained to obtain final motion contrast data.

  Here, although motion contrast data is determined based on the amplitude of complex data after FFT, the method of determining motion contrast data is not limited to the above method. Motion contrast data may be determined based on phase information of complex data, or motion contrast may be determined based on both amplitude and phase information. Also, motion contrast can be determined based on the real part or imaginary part of complex data.

  Furthermore, although motion contrast data is obtained by calculating the decorrelation value of two values in the above, motion contrast data may be obtained based on the difference between the two values, or based on the ratio of the two values. It is also possible to obtain motion contrast data.

  Furthermore, although the final motion contrast data is obtained by obtaining the average value of the plurality of acquired decorrelation values in the above, the maximum value of the plurality of decorrelation values, differences, and ratios is the final motion contrast data. As well.

  Next, a procedure for generating an OCTA image from motion contrast data will be described. In the motion contrast data, the upper end and the lower end are set in the depth direction of the target location, and an OCTA image is generated by performing projection in the region between them.

  An example of the function of the image generation unit 304 in this specification will be described using FIG. The function illustrated in FIG. 3 is realized by a processor or the like as in the image generation unit 304.

  The original data input unit 401 acquires original data for generating an Enface image, such as tomographic image data and motion contrast data, and notifies the target range selection unit 402 and the Enface image generation unit 404 of the original data. Specifically, the original data input unit input unit 401 acquires the three-dimensional tomographic image data generated by the reconstruction unit 301 and the three-dimensional motion contrast data generated by the motion contrast generation unit 302, and the target range selection unit 402, And the Enface image generation unit 404. Here, the three-dimensional tomographic image data and the three-dimensional motion contrast data correspond to an example of a three-dimensional image. The original data input unit input unit 401 corresponds to an example of an acquisition unit that acquires a three-dimensional image of the eye to be examined. The original data input unit 401 may input the original data only to the Enface image generation unit 404. The original data includes three-dimensional tomographic image data and three-dimensional motion contrast data.

  The target range selection unit 402 selects a target range for generating an Enface image in the original data, and notifies the projection method selection unit 403 and the Enface image generation unit 404 of the selected target range. The target range selection unit 402 selects a target range for generating an Enface image according to an input from the user via the pointing device 320 or the keyboard 321, for example. The target range is, for example, a range in the depth direction and / or a range in the fundus surface direction. The target range selection unit 402 may select a target range using an arbitrary shape such as a rectangle or a Bezier curve in the original data according to the user's input, or may select the target range by another arbitrary method. good. The target range selection unit 402 can automatically determine the target range when there is a predetermined target range for generating the Enface image.

  The projection method selection unit 403 selects a projection method based on the target range, and notifies the Enface image generation unit 404 of the selected projection method. That is, the projection method selection unit 403 corresponds to an example of the projection method determination unit that determines the projection method for generating the Enface image according to the target range. The projection method selection unit 403 selects a projection method based on the depth range in the selected target range. The projection method selection unit 403 can select the projection method based on, for example, information in which the target range stored in the memory and the projection method are associated with each other.

  The Enface image generation unit 404 generates an Enface image based on the original data, the target range, and the projection method, and notifies the display unit 405 of the generated Enface image. The display unit 405 displays the Enface image. That is, the Enface image generation unit 404 corresponds to an example of an EnFace image generation unit that generates an EnFace image based on a three-dimensional image.

In the present embodiment, the projection method selection unit 403 determines which of maximum value projection (MIP) and average value projection (AIP) to use, for example, based on the width of the depth range of the target range. In this embodiment, AIP and MIP are defined as follows.
・ AIP: The average value of the values in the depth direction in the target range is projected ・ MIP: The maximum value is projected after smoothing by moving average in the depth direction as a measure against noise Note that the top few in the depth direction as MIP It is also possible to project the maximum value of the values after excluding. Further, the smoothing method in MIP is not limited to the moving average, and may be smoothed by another method. Also, in MIP, it is not necessary to perform smoothing.

  In Enface images, projection may be performed with AIP in order to observe a less noisy and smooth structure. On the other hand, in the case where the depth range in the projection range is wide, there is a possibility that information of a locally tone-protruded area such as choroidal neovascularization may be buried. Specifically, in the tomographic image shown in FIG. 6, when the layer 601 and the layer 602 are set as the target range of projection, the choroidal neovascular vessel 603 is averaged in the range of the target range 604 of projection indicated by the dashed arrow. Be done. As a result, in the Enface image after projection shown in FIG. 7, the choroidal neovascularization 701 may not be able to obtain sufficient contrast. Therefore, as shown in FIG. 8, even if the noise increases, it may be desirable in some cases to project with MIP that allows the observation target (choroidal neovascularization 801) to be observed.

  The projection method selection unit 403 sets the threshold X (temporarily 20 μm) of the depth range based on, for example, the diameter size (temporarily 5 μm) of the new blood vessel to be observed, and the maximum value of the depth range where the Enface image is generated. If the threshold value X exceeds the threshold value X, the MIP is selected, and if the depth range for generating the Enface image is equal to or smaller than the threshold value X, the AIP is selected. Note that the threshold may be determined in advance regardless of the size of the observation target. That is, when the size of the depth range for generating the Enface image exceeds a predetermined threshold, the projection method selection unit 403 selects MIP as the projection method, and the size of the depth range for generating the nface image is in advance. If it is below the defined threshold value, AIP is selected.

  Although the present embodiment will be described focusing on AIP and MIP as a projection method, a projection method such as median projection or mode projection may be used.

  In addition, the projection method selection unit 403 may determine, for example, whether or not the removal process of the projection artifact is necessary based on the depth of the target range. In the motion contrast image, for example, in the motion contrast image of the retinal pigment subepithelium, the same blood vessel as the upper blood vessel is depicted as a false blood vessel, or in the motion contrast image of a low brightness region such as the outer granular layer, the upper blood vessel Projection artifacts such as shadows are easily generated. Therefore, when the depth specified by the target range selection unit 402 is deeper than the predetermined threshold (in the case of the choroid), it is desirable to perform processing for removing the projection artifact. That is, when the depth of the target range selected by the target range selection unit 402 includes a position deeper than a predetermined threshold, the projection method selection unit 403 determines that the projection artifact is to be removed, and the target range selection unit 402 selects If the depth of the target range does not include a position deeper than a predetermined threshold, it is determined that the removal of the projection artifact is not performed. Note that the projection method selection unit 403 may not determine whether to remove the projection artifact. In addition, it is possible to use a well-known various method of the removal method of a projection artifact. For example, it is possible to identify a blood vessel included on the vitreous side from the depth range in which the Enface image is generated, and to subtract the identified blood vessel from the Enface image to remove a projection artifact.

  The basic form of the process flow in this specification will be described using the flowchart of FIG.

  In S501, the target range selection unit 402 selects a target range to be projected according to the user's input or automatically. In step S502, the projection method selection unit 403 determines whether the depth range in the target range is sufficiently wide relative to the observation target. Note that the projection method selection unit 403 may simply determine whether or not the depth range is larger than the threshold in S502. If the depth range in the target range is sufficiently narrow relative to the observation target, the target range selection unit 402 selects AIP as the projection method in S504. If the depth range in the target range is wider than the observation target in S502, the target range selection unit 402 selects MIP as the projection method in S503. Thereafter, in step S505, an Enface image is generated by the selected projection method.

  According to this embodiment, since the projection method is selected according to the size of the depth range for generating the Enface image, it is possible to generate an Enface image more appropriate than when the projection method is fixed.

  In addition, since an appropriate projection method is automatically selected, it is possible to easily generate an appropriate Enface image.

<Modification of Embodiment 1>
Next, a modification of the first embodiment will be described. In the present embodiment, a target range (depth range) for generating the Enface image is selected according to the image processing mode for generating the Enface image.

  An apparatus configuration according to a modification of the first embodiment will be described with reference to FIG. In the present embodiment, an image processing mode selection unit 410 is added to the basic form described in FIG. 3 in order to make it easy for the user to select a target range. In the present embodiment, when the user selects the image processing mode, the image processing apparatus 300 automatically selects the target range and determines the projection method. The image processing mode selection unit 410 selects an arbitrary image processing mode from a plurality of image processing modes in accordance with an input from the user via the pointing device 320 or the keyboard 321, for example. That is, the image processing mode selection unit 410 corresponds to an example of a selection unit that selects a mode indicating a target range for generating an Enface image from a plurality of modes.

  In this embodiment, as an example of the image processing mode, a CNV mode suitable for depiction of choroidal neovascularization and a superficial blood vessel mode suitable for depiction of superficial blood vessels are defined. The CNV mode is the target of Enface image generation between the boundary (upper end) of OPL (Outer Plex Layer) and ONL (Outer Nuclear Layer) to the Bruch's membrane (lower end) as a place where choroidal neovascularization may exist. This is a mode to select as a region. The surface vascular mode is selected as a target region between the boundary (lower end) of ICL (Inner Nuclear Layer) (upper end) to GCL (Ganglion Cell Layer) and IPL (Inner Plexiform Lyaer) to facilitate observation of the superficial blood vessel. Do. At this time, margins may be provided at the upper end and the lower end in consideration of the possibility that each blood vessel exists beyond the boundary. Here, the CNV mode is an example of a first mode in which a range defined based on the boundary between OPL (Outer Plex Layer) and ONL (Outer Nuclear Layer) and the Bruch film is associated as a target range for Enface image generation. Equivalent to. In the superficial vascular mode, a range defined based on an ILM (Inner Nuclear Layer) (upper end) and a boundary between a GCL (Ganglion Cell Layer) and an IPL (Inner Plexiform Lyaer) corresponds as a target range for Enface image generation. This corresponds to an example of the second mode.

  Note that the image processing mode is not limited to the above example, and three or more modes may be provided. The image processing mode selection unit 410 selects a target region of CNV, and corresponds to an example of a target region selection unit that selects a target region for generating an Enface image.

  The flow of processing in the present embodiment will be described using the flowchart of FIG. In step S901, the image processing mode selection unit 410 selects an image processing mode in accordance with an input from the user. In step S501, the image processing mode selection unit 410 automatically selects a target range associated with the selected image processing mode.

  In step S902, the projection method selection unit 403 switches the process for generating the Enface image according to the selected image processing mode. That is, the projection method selection unit 403 corresponds to an example of a projection method determination unit that determines the projection method for generating the Enface image according to the selected mode.

  Here, choroidal neovascularization may be accompanied by ridges 1301 of RPE, for example as shown in FIG. In such a case, since the depth range 1304 in the target range shown by thick lines 1302 and 1303 in FIG. 13 is partially enlarged, there are cases where the AIP can not visualize the choroidal neovascularization. Further, in the case of the CNV mode of the present embodiment, the width of the depth range in the target range is sufficiently large for the choroidal neovascularization, so the projection method selection unit 403 selects the MIP in S503. In addition, since the depth of the target range is deeper than the predetermined threshold, the projection method selection unit 403 is set to perform projection artifact removal processing in S903. On the other hand, since superficial blood vessels are generally thicker and highly reflective than choroidal neovascularization, they are likely to be visualized with high brightness in motion contrast images. Therefore, in the case of the superficial blood vessel mode, the projection method selection unit 403 selects AIP in S504. Further, since the depth of the target range is shallower than the predetermined threshold, the projection method selection unit 403 does not perform the projection artifact removal process in S904. In step S505, the Enface image generation unit 404 generates an Enface image based on the projection method selected in S503, S903, or S504, and S904, and the display control unit 405 causes the monitor 310 to display the Enface image. That is, in the present embodiment, MIP is associated with the CNV mode as a projection method as a default, and AIP is associated as a projection method as a default.

  As described above, in the present embodiment, the depth range and the projection method for automatically generating the Enface image are selected according to the image processing mode selected by the user, so the user can easily select an appropriate Enface image. It becomes possible to observe.

  In addition, although the example of the CNV mode for observing a wide range choroidal neovascularization was shown in a present Example, it may be desired to observe the retinal neovascularization which exists in a specific layer. Therefore, another CNV mode may be defined in which the target region is limited only to, for example, the Bruch film periphery. In this case, since the target range is not wide relative to the choroidal neovascularization to be observed, the information is unlikely to be buried even if AIP is used. Therefore, AIP can be selected with priority given to noise reduction. That is, even in the image processing mode in which a new blood vessel is to be observed, the projection method selection unit 403 may select the projection method based on the width of the depth range for generating the Enface image. .

  In this embodiment, the target range and the projection method are automatically determined when the image processing mode is selected. However, the present invention is not limited to this. For example, after detecting a layer boundary from tomographic image data of an eye to be examined, a layer thickness may be calculated to determine a projection method. In addition, as a projection method to be determined, the contrast, the noise reduction parameter, and the like may be determined according to the target range and the layer thickness. This makes it possible to generate an Enface image according to the observation target.

  In the present embodiment, the projection method is uniquely determined based on the depth of the target range, but is not limited to this. For example, when an Enface image generated for the same subject eye already exists, the projection method to be preferentially used may be determined by the above method.

  Also, projection artifact removal is not an essential process.

  In the present embodiment, the CNV mode is associated with MIP as a projection method as a default, and the superficial blood vessel mode is associated with AIP as a projection method as a default. However, according to an instruction from the user, the projection method is It can be changed in any way.

Second Embodiment
In the first embodiment, the configuration for selecting the projection method based on the information in the depth direction of the target range has been described. In this embodiment, a configuration will be described in which the projection method is selected based on the area of the target range in the plane perpendicular to the depth direction. That is, in the present embodiment, the projection method for generating the Enface image is determined based on the range in the plane orthogonal to the depth direction.

  The apparatus configuration in the present embodiment is the same as that in the first embodiment. Further, the target range in the depth direction in the present embodiment is, for example, the entire retina.

  In the present embodiment, a macular mode and an optic disc mode are defined as the image processing mode. Lesions around the macula can greatly affect the subject's vision. The macular mode in which the macula is the back generation target of the Enface image and the optic nerve head mode in which the optic nerve head is the generation target of the Enface image are modes indicating different target ranges in a plane orthogonal to the depth direction.

  Therefore, in the case of the macular mode, the projection method selection unit 403 selects MIP as a projection method in order to observe local features as much as possible. On the other hand, the area around the optic papilla has a different structure from other parts, such as blood vessels running in the depth direction. For this reason, in the case of the optic disc mode, the projection method selection unit 403 selects the AIP in order to observe the entire optic disc area. That is, the projection method selection unit 403 selects MIP as the projection method when the image processing mode selection unit 410 selects the macular mode, and AIP as the projection method when the image processing mode selection unit 410 selects the optic disc mode. Choose

  The flow of the process at this time will be described using the flowchart of FIG. 10 and the difference from the flowchart of FIG. 9 described in the first embodiment. In step S902, the projection method selection unit 403 determines the selected image processing mode. If the macular mode is selected, the projection method selection unit 403 selects MIP as the projection method in S503. In the case of the optic disc mode, the projection method selection unit 403 selects AIP as the projection method in S504. Further, since the entire retinal layer is assumed as the target range in the depth direction, projection artifact removal is unnecessary. Therefore, in S904, it is selected that projection artifact removal is unnecessary. Note that projection artifact removal is not an essential process.

  In this embodiment, the method of selecting the projection method according to the part is shown. The value used as the feature of the fundus is not limited to site information, and lesion information and the like may be acquired and used by any method.

  According to the above-described method, it is possible to generate an Enface image appropriate for the target range in a plane perpendicular to the depth direction.

Embodiment 3
In the first and second embodiments, an example in which the target range and the projection method are selected based on the image processing mode and the Enface image is generated is described. In this embodiment, an example will be described in which a plurality of Enface images are generated and displayed by changing a part of the configuration of the basic form shown in the first embodiment.

  As described in the first embodiment, when the MIP is selected as the projection method, the contrast is high, and a fine structure can be observed, but there is a tendency that there are many noises. On the contrary, when AIP is selected as the projection method, although a smooth structure with less noise can be observed, there is a possibility that the information of the portion where the gradation is locally projected may be buried. By comparing the Enface image generated by projecting with MIP (hereinafter referred to as MIP-Enface image) with the Enface image generated by projecting with AIP (hereinafter referred to as AIP-Enface image), it is possible to support a part that is difficult to observe alone. It becomes possible to observe. In the present embodiment, the Enface image generation unit 404 generates both the MIP-Enface image and the AIP-Enface image. Then, the display control unit 405 causes the monitor 310 to display both the MIP-Enface image and the AIP-Enface image. Thereby, the user can adaptively select the projection method of the Enface image according to the state of the input original data with reference to the displayed images of the plurality of projection methods.

  The apparatus configuration in the present embodiment will be described with reference to FIG. A priority determination unit 1201 is added to the basic form described in FIG. The projection method selection unit 403 selects a plurality of projection methods and notifies the Enface image generation unit 404 and the priority determination unit 1201. The Enface image generation unit 404 generates a plurality of Enface images based on a plurality of projection methods. The priority determining unit 1201 gives priorities to Enface images generated by the respective projection methods, and notifies the display unit 405 of the Enface images. The display control unit 405 initially displays, for example, a high-priority Enface image. When the user wants to observe another EnFace image different from the EnFace image currently displayed on the monitor, the display control unit 405 may switch to another EnFace image according to an instruction from the user and cause the monitor 310 to display the image. it can. At this time, since EnFace images based on a plurality of projection methods have been generated, it is possible to immediately confirm EnFace images by another projection method.

  Here, the flow of processing will be described using the flowchart of FIG. In S501, a target range including a depth range for generating the Enface image is selected according to the user's input or automatically. In step S1101, the Enface image generation unit 404 generates a MIP-Enface image and an AIP-Enface image. In step S502, when the depth range in the target range is, for example, sufficiently narrow relative to the observation target, the priority determining unit 1201 sets the priority of the AIP-Enface image higher than the priority of the MIP-Enface image in S1103. In step S502, when the depth range in the target range is wider than the observation target, the priority determining unit 1201 increases the priority of the MIP-Enface image to the priority of the AIP-Enface image in step S1102. In step S1104, a high-priority Enface image is displayed. If the depth range for generating the Enface image is larger than the threshold, the priority determining unit 1201 makes the priority of the MIP-Enface image higher than the priority of the AIP-Enface image, and the depth range for generating the Enface image is If it is smaller than the threshold, the priority of the AIP-Enface image may be higher than the priority of the MIP-Enface image.

  In the present embodiment, the EnFace image with high priority is initially displayed, but the present invention is not limited to this. For example, a plurality of generated Enface images may be displayed side by side on one screen. In this case, it is possible to add a symbol such as a circle to the target Enface image or display the Enface image larger than other Enface images so that the Enface image having high priority can be identified.

  Although this embodiment shows a method of generating a plurality of Enface images by changing MIP and AIP, even if a plurality of Enface images are generated by changing the presence or absence of projection artifact removal setting, contrast, and noise reduction parameters. good.

  According to the above method, Enface images generated by another projection method can be additionally observed, and the projection method of EnFace images can be adaptively changed according to the input original data.

<Other Embodiments>
The exemplary embodiments have been described above in detail, but the disclosed technology can be embodied as, for example, a system, an apparatus, a method, a program, or a recording medium (storage medium). Specifically, the present invention may be applied to a system configured of a plurality of devices (for example, a host computer, interface device, imaging device, web application, etc.), or may be applied to a device including one device. good.

  Further, it goes without saying that the object of the present invention can be achieved by the following. That is, a recording medium (or storage medium) storing a program code (computer program) of software for realizing the functions of the above-described embodiments is supplied to a system or apparatus. Such storage media are, of course, computer readable storage media. Then, the computer (or CPU or MPU) of the system or apparatus reads out and executes the program code stored in the recording medium. In this case, the program code itself read from the recording medium implements the functions of the above-described embodiments, and the recording medium recording the program code constitutes the present invention.

Claims (12)

Acquisition means for acquiring a three-dimensional image of an eye to be examined; EnFace image generation means for generating an EnFace image based on the three-dimensional image;
Target range selecting means for selecting a target range for generating the Enface image;
An image processing apparatus comprising: projection method determination means for determining a projection method for generating the Enface image according to the target range. Acquisition means for acquiring a three-dimensional image of an eye to be examined; EnFace image generation means for generating an EnFace image based on the three-dimensional image;
Selecting means for selecting a mode indicating a target range for generating the Enface image from a plurality of modes;
An image processing apparatus comprising: projection method determination means for determining a projection method for generating the Enface image according to the selected mode.   The image processing apparatus according to claim 1, wherein the target range is a range in a depth direction.   The image processing apparatus according to claim 1, wherein the target range is a range in a plane orthogonal to the depth direction.   The image processing apparatus according to claim 2, wherein the plurality of modes are modes representing target ranges of different depths.   The image processing apparatus according to claim 2, wherein the plurality of modes are modes indicating different target ranges in a plane orthogonal to the depth direction. The target range is a range in the depth direction,
The plurality of modes include a first mode and a second mode,
A range defined based on the boundary between OPL (Outer Plex Layer) and ONL (Outer Nuclear Layer) and the Bruch film is associated with the first mode as the target range, and the second mode is associated with the ILM (Inner A range defined based on the boundary between Nuclear Layer (upper end) and GCL (Ganglion Cell Layer) and IPL (Inner Plexiform Lyaer) is associated as the target range,
The projection method determining means determines the projection method to be the maximum value projection when the first mode is selected, and determines the projection method to be the average value projection when the second mode is selected. The image processing apparatus according to claim 2, characterized in that:   The image processing apparatus according to any one of claims 1 to 7, further comprising display control means for causing the display unit to display the Enface image generated according to the projection method determined by the projection method determination means. Acquisition means for acquiring a three-dimensional image of an eye to be examined; EnFace image generation means for generating a plurality of EnFace images using a plurality of projection methods based on the three-dimensional image;
A priority determination unit that gives priority to the plurality of EnFace images according to a target range for generating each of the plurality of Enface images;
An image processing apparatus, comprising: display control means for causing a display unit to display the plurality of EnFace images based on the priorities assigned to the plurality of EnFace images. Acquisition means for acquiring a three-dimensional image of an eye to be examined; EnFace image generation means for generating an EnFace image based on the three-dimensional image;
Target site selecting means for selecting a target site for generating the Enface image;
An image processing apparatus comprising: projection method determination means for determining a projection method for generating the Enface image according to the target region. An acquisition step of acquiring a three-dimensional image of an eye to be examined; and an EnFace image generation step of generating an EnFace image based on the three-dimensional image;
A target range selection step of selecting a target range for generating the Enface image;
A projection method determination step of determining a projection method for generating the Enface image according to the target range.   A program that causes a computer to execute the steps of the image processing method according to claim 11.

Images